1. The Puzzle
Drawing on the work of Hambrick, Christensen, Moore, Tushman, Osterwalder, and others, the past several decades have produced many highly developed and well-validated frameworks for the work of new-product strategy and business growth. Hambrick and Frederickson (2001) gave us a clean schema for what a strategy actually is. Christensen (1997, 2003) gave us the disruption framing that resolved why incumbents lose, and Christensen and colleagues (1997, 2003, 2016) along with Ulwick (1991+) developed Jobs to Be Done as the demand-side discipline that separates customer needs from existing product categories. Moore (1991, 2015) gave us the chasm-crossing and zone-based transformation disciplines for taking new products from early adopters to mainstream and for organizing the enterprise to run new businesses alongside established ones. Porter (1980) gave us the structural analysis of industry forces and competitive dynamics. Kim and Mauborgne (2005) gave us the Blue Ocean / Red Ocean choice between contesting existing demand and creating new demand space. Cooper (1986+) gave us stage-gate process. Ries (2011) and Blank (2005) gave us lean experimentation. McGrath and MacMillan (1995) gave us discovery-driven planning under uncertainty. Brown and IDEO (2009) gave us the desirability / feasibility / viability evaluation triangle. Osterwalder and Pigneur (2010) gave us business-model articulation. Bain & Company and Zook (2001) gave us adjacency analysis. Tushman and O’Reilly (1997) gave us the ambidextrous organization and prescriptive approaches to tailored governance for explore work. Collins and Porras (1994) gave us the Vision Framework for building the foundational clarity of purpose and measurable definition of success that long-lived companies anchor to.
Each is well-developed for the specific question it answers. None covers everything needed for the end-to-end work of taking a new business idea from initial hypothesis to validated, scaled execution in a mature enterprise. The complex question of how do these frameworks fit together is what every sophisticated innovation leader eventually asks. It is a puzzle that can be broken down into three layers, each with a distinct audience that feels the pain most acutely.
Layer 1: Strategy hypothesis definition and evaluation
The first layer of the puzzle belongs to the product manager and the venture lead — the practitioner staring at a blank canvas with a new business idea, a small team, and the question of how do I turn this kernel of an idea into a clear strategy hypothesis I can evaluate and test? The borrowed frameworks at this layer each answer part of the question. Collins and Porras’s Vision Framework brings clarity of purpose and measurable definition of success rooted in commitment-inducing collective values. Hambrick and Frederickson’s Strategy Diamond gives a five-element schema: arenas, vehicles, differentiators, staging, economic logic. Christensen’s Jobs to Be Done framework separates customers’ unmet needs from existing product categories. Moore’s whole-product construct gives the demand-side discipline for what counts as a complete offering rather than a partial one. Porter’s Five Forces expose competitive dynamics within the industry structure. Kim and Mauborgne frame the strategic choice between contesting existing demand and creating new demand space. Bain and Zook’s adjacency analysis frames where the new business sits relative to the company’s core. Osterwalder’s business-model canvas gives a structured surface for revenue-streams and cost-structure articulation.
Each of these is rigorous. None reconciles directly with the others. A practitioner stitching three of them together informally ends up with a collection of strategy artifacts with no explicit account of how the pieces relate. The puzzle at this layer is making them coherent enough to write down a strategy that is comprehensive, self-consistent, testable, and holds up to scrutiny — a strategy whose target markets, competitive differentiation, whole solution, implementation approach, financial logic, and staging explicitly fit together, rather than being left to stakeholders to interpret or assume. Or, worse, containing significant gaps and unreconciled or conflicting assumptions hidden inside the gaps.
Layer 2: The nested loops — inner-loop methodology inside stage-gated investment
The second layer of the puzzle belongs jointly to the project lead running the inner loop and the investment decision-maker running the outer loop. The inner loop is hypothesize-test-learn iteration on a tight cadence. Lean Startup’s build-measure-learn (Ries, 2011) is the early-stage variant; McGrath and MacMillan’s discovery-driven planning is the financial-uncertainty variant; Brown and IDEO’s design-thinking prototype-test-iterate cadence is the desirability variant. A practitioner typically picks one of these variants rather than trying to operate three in parallel — and the harder question is how the chosen inner-loop method integrates with the strategy hypothesis schema in Layer 1, with the outer-loop investment cadence below, and with the portfolio governance the next layer addresses. The outer loop is stage-gated investment decision: Cooper’s stage-gate process applied at the pace at which new businesses actually mature, with outcomes that resolve to Continue / Pivot / Pause / Stop rather than the binary Go / Kill posture that pushes practitioners to either overstate marginal projects or prematurely kill ones with hidden potential.
Each loop has a rigorous tradition behind it. The puzzle at this layer is what goes wrong at the loops themselves. On the inner loop, the most consequential failure pattern is the team that falls in love with an initial product concept and races toward development without testing the whole strategy first — sometimes without even articulating a strategy hypothesis beyond the product concept. On the outer loop, the most consequential failure patterns are sunk-cost bias keeping a weak concept funded past the point evidence justifies, an empty pipeline that leaves the gate decision-maker with no credible alternatives to redirect resources toward, and evaluation criteria too thin or inconsistent to surface what the evidence actually says. These failures don’t come from the methods being wrong; they come from the methods being applied in isolation, without the dimensional discipline of the strategy hypothesis schema, the integration discipline between the loops, and the portfolio context that makes stopping decisions practically defensible.
Layer 3: Portfolio strategy as the wrapper
The third layer of the puzzle belongs to the portfolio manager — the head of innovation, the head of corporate development, the executive team responsible for the company’s growth pipeline as a whole. The borrowed frameworks at this layer give pieces of the answer. Tushman and O’Reilly’s ambidextrous organization (1997) provides the structural insight that explore work and exploit work require different organizational treatment, and is prescriptive about creating tailored governance approaches for the explore side. Bain and Company and Zook’s adjacency analysis (2001) gives a portfolio-allocation lens across core, adjacent, and disruptive territory — though the allocation guidance is based on correlated adjacency outcomes rather than causal mechanisms, and the foundational stick to the core advice is conditional on the core having sufficient growth potential and not being under near-term disruption threat. Moore’s Zone to Win (2015) articulates the operational discipline of running a continuously-updated portfolio view across performance / productivity / incubation / transformation zones, and is prescriptive about how governance approaches need to evolve as ventures mature and scale across the zones.
These three frameworks together cover substantial territory. What none of them provides — even with Moore’s zone-based governance evolution included — is a clear diagnostic for predictively characterizing the risks of pursuing specific adjacencies, and a methodology for allocating investment under the different failure-rate assumptions that different adjacencies or maturity stages actually carry. Those two gaps are what Layer 3 has to close, and they are where the practical work of running a portfolio gets hardest in mature enterprises.
Layers 2 and 3 together — the nested loops plus portfolio strategy — comprise what BRI calls Staged Innovation Methodology, the bundled offering that pairs inner-loop iteration and outer-loop stage-gated investment with portfolio governance. Together with the BRI Strategy Framework at Layer 1, the Staged Innovation Methodology forms the integrated system the article describes.
Four observations the synthesis is up against
The borrowed frameworks are individually helpful. When a practitioner tries to apply them in the context of new-product or new-business growth inside a mature company, things get complicated very quickly. Four observations surface in practice that explain why.
1. The frameworks overlap and the overlaps are not clean. Lean Startup’s customer-validation experimentation overlaps with Discovery-Driven Planning’s milestone-and-assumption-test discipline at the inner-loop iteration level. Cooper’s stage-gate process overlaps with Lean Startup at the project-progression level. The overlaps are not redundant; they are where the frameworks talk past each other. A practitioner using Lean Startup’s vocabulary and Cooper’s vocabulary in the same sentence is, more often than not, equivocating on what counts as evidence and what counts as a decision.
2. The frameworks are calibrated to different starting conditions. Lean Startup was developed for the startup founder iterating on a single new business under capital constraints. Cooper’s stage-gate was developed for the corporate R&D pipeline running multiple new products through structured commitment phases. Discovery-Driven Planning was developed for the new-venture leader operating inside an established corporation. Design Thinking was developed for the design consultancy treating desirability as the primary lens. Each is calibrated to the conditions it emerged in. Applying any one to a different starting condition imports more than the method; it imports the assumptions of the originating context. An industrial-equipment manufacturer running an inner-loop methodology developed for a software startup runs into the four-quarter capital-equipment budget cycle, the unionized engineering workforce, the regulated product-safety regime, and the company’s existing capital structure — none of which the borrowed method’s underlying assumptions account for.
3. Inconsistent or unreconciled foundational assumptions. Frameworks built from different epistemic starting points carry assumptions about what counts as evidence, what counts as a strategy choice, and what counts as a decision. When a practitioner stitches three of them together informally, the assumption inconsistency shows up in practice as the team that cannot agree on whether to Continue / Pivot / Pause / Stop a project. One reviewer is operating on customer-validation criteria, another on financial-defensibility criteria, another on a desirability-feasibility-viability balance. Each is using a defensible framework. They reach different conclusions because the frameworks rest on unreconciled assumptions about what evidence carries the most weight, and the practitioner stitching them together never made the reconciliation explicit.
4. The hidden methodology / host-organization interlock. A fourth observation surfaces in practice but is rarely named cleanly. The methodology that worked in one company stalls or fails in another, even when the strategy is sound and the team is capable. Symptoms surface: projects that should advance pause inexplicably; gates that should clear get stuck on issues nobody can articulate; methodology adoption that lands clean on the slide but collapses in the work. The root cause is almost always the interlock between the methodology and the host organization’s resources, processes, and priorities — what Christensen and Kaufman (2006) named the RPP lens — and the cumulative impact of small misalignments accumulating into compounded execution friction. Few practitioners diagnose the pattern; fewer still know how to design around it.
A new piece of the puzzle
The puzzle has a recent addition. The integration question, having been answered for the borrowed frameworks of the past several decades, now extends to a new capability set the field is actively absorbing: generative AI — large language models, chatbots, agents, and AI-enabled applications and services. The frameworks of the past several decades did not anticipate this capability set, because the capability set did not exist when the frameworks were developed. Where the new capability fits — and where the structural integration we describe still matters as AI capability continues to evolve — is the work of the article’s closing section.
The synthesis claim
The integrated system we describe in the article is what we wish had existed when we were inside corporate innovation programs trying to make the field’s best frameworks work together in practice. Drawing on decades of practitioner experience leading new-business strategy and innovation work, we describe the system not as a replacement for any of the borrowed frameworks but as the integrated picture of how the pieces fit. What each framework is rigorously right about. Where the seams are. What our practice has added to make the puzzle hang together. How the integrated system serves all three audience layers without forcing one to absorb the complexity of all three.
The article starts at Layer 1 — what is a strategy — because that is the layer where the pieces don’t really fit, and where the first contribution of our practice lands.
2. The BRI Strategy Framework
At Layer 1, the practitioner’s work is to turn a new-business idea into a strategy hypothesis that holds up to scrutiny. The acute pain at this layer is the question every new-business strategy hypothesis has to answer eventually and often answers badly: what is our strategy here, is it a real strategy or a wish list, and how do we know? The borrowed frameworks at this layer give partial answers.
The BRI Strategy Framework treats the strategy itself as a hypothesis. The framing is deliberate. A hypothesis carries explicit and testable choices and assumptions, not vague commitments or hidden interpretations. A hypothesis is comprehensive across the consequential dimensions and self-consistent within and across them. A hypothesis serves as the organizing framework that absorbs more narrowly focused borrowed frameworks coherently, rather than leaving the practitioner to stitch them together informally. And a hypothesis is revisable: every dimension carries assumptions, every assumption is testable, every test result revises the strategy.
Our starting point for the framework is Hambrick and Frederickson’s Strategy Diamond (2001), articulated in Are You Sure You Have a Strategy? — five elements that together, they argued, capture the consequential choices any strategy contains: Arenas (where will the firm be active), Vehicles (how will it get there), Differentiators (how will it win), Staging (what will be the speed and sequence of moves), and Economic Logic (how will returns be obtained). The Diamond is rigorous, well-validated, and the cleanest available schema for what a strategy actually is at the highest level of abstraction. The BRI framework extends it in two substantive ways.
First, the BRI framework splits Hambrick’s Vehicles into two distinct dimensions — Whole Solution (the what) and Implementation Approach and Execution (the how) — recognizing that the two answer different strategic questions and that conflating them tends to leave one of the two underspecified in the strategy artifact. Moore’s whole-product construct from Crossing the Chasm (1991) anchors the Whole Solution dimension; build-vs-buy-vs-partner-vs-license discipline anchors Implementation Approach.
Second, the BRI framework treats the dimensions as a strategy hypothesis schema rather than as a static description of strategy choices. The schema is structured to support both definition and evaluation within a single artifact, where the Diamond and adjacent frameworks (Business Model Canvas, Lean Canvas) are primarily definition tools that require separate evaluation instruments.
Strategy Objective — the foundation underneath the dimensions
Underneath the six dimensions sits the Strategy Objective — what outcome the project aspires to achieve, in what timeframe, measured by what metrics. The objective is not a seventh dimension; it is the foundation the dimensions rest on. Without a clear objective and clear measures of success, there is no basis for evaluating the relative merit of different strategy alternatives. The objective is what makes the rest of the schema decision-useful.
In a mature enterprise, the Strategy Objective for any specific new-business initiative ladders up to the company’s foundational vision of itself — its core purpose, its core values, and the envisioned future it is built to pursue. The Vision Framework that Collins and Porras articulated in Built to Last (1994) is the canonical reference here. The Vision Framework brings clarity of purpose, measurable definition of success, and the commitment-inducing collective values that align an organization to long-term outcomes the strategy is serving. A new-business Strategy Objective that has no coherent line of sight to the company’s Vision Framework is a Strategy Objective that will struggle for organizational commitment when the work gets hard. A Strategy Objective that ladders cleanly to the company’s vision can mobilize the resources, attention, and patient capital that new businesses need to mature.
What follows are the six dimensions of the BRI Strategy Framework, each with attribution to the borrowed frameworks the dimension absorbs and an explicit account of what the BRI framework extends.
Dimension 1 — Target Markets and Unmet Need
The first dimension describes the target markets the strategy is intended to serve and the factors driving or creating the product or business opportunity: target use cases, customer challenges with current solutions, unmet needs, and Jobs to Be Done.
The dimension draws on Hambrick’s Arenas and on the Jobs-to-Be-Done tradition that Ulwick developed as outcome-driven innovation (Ulwick, 1991+) and that Christensen elaborated as the formal JTBD framing in The Innovator’s Solution (Christensen and Raynor, 2003) and synthesized in Competing Against Luck (Christensen, Hall, Dillon, and Duncan, 2016). The Diamond’s Arenas element answers the where will we be active question at the level of product categories, market segments, geographies, and value-chain stages. JTBD answers the deeper why does the customer hire a solution at all question at the level of the underlying job the customer is trying to accomplish and the unmet needs that surface when current solutions fall short. The two questions interact: the answer to where will we be active is rigorous only when the answer to what job is the customer hiring us for is rigorous.
The BRI JTBD treatment calibrates a middle path between Ulwick’s granular outcome-driven approach and basic single-main-JTBD treatments. A Job to Be Done is what the customer is hiring a solution to accomplish, expressed in a standard syntax of Verb + Object + Clarifying Context, independently of any particular solution, technology, or product category. Move the lawn into a more presentable state every week without consuming the homeowner’s Saturday is a Job to Be Done; robotic mower is one possible solution. The Job persists across solutions; the solutions compete for the hiring decision. The discipline is solution-agnostic and stable over time — the same Job survives strategy revisions, prototype iterations, and even pivots that change the solution shape entirely.
The BRI framework’s contribution at this dimension is to treat JTBD as the demand-side discipline that resolves who is the customer and what are they hiring this for into evidence rather than positioning prose. Most failures at this dimension come from a strategy hypothesis whose target-market definition is descriptive rather than evaluable. A team that names small and medium businesses in the manufacturing vertical who need to modernize their operations as their target market has named a positioning frame, not a Job. The Job — what the manufacturing operations leader is actually trying to accomplish that current solutions are not accomplishing well — is what makes the rest of the hypothesis evaluable. The dimension carries explicit Jobs to Be Done in the standard syntax, with associated unmet needs and Market Requirements scored on importance and current-alternatives satisfaction. The Implied Opportunity Score that combines those two scores is what surfaces underserved territory the strategy can credibly target.
Dimension 2 — Competitive Differentiation
The second dimension describes the current and potential alternative solutions target markets can choose from, the basis of competition, and how the whole solution will be differentiated.
The dimension draws on Hambrick’s Differentiators and on the broader competitive-strategy tradition. Porter’s Competitive Strategy (1980) gave the field the Five Forces frame for industry-structural analysis and the categorical taxonomy of competitive alternatives — direct rivals, substitutes, potential entrants, customer-side bargaining alternatives. Kim and Mauborgne’s Blue Ocean Strategy (2005) framed the strategic choice between contesting existing demand and creating new demand space. Christensen’s disruption framing from The Innovator’s Dilemma (1997) explained how new entrants displace established offerings through different value vectors that incumbents systematically underestimate. Each of these covers a substantive piece of the competitive picture, and each is independently rigorous.
What the BRI framework adds at Dimension 2 is the cross-dimensional connection. Competitive differentiation is not evaluated against direct-competitor specs and features in isolation; it is evaluated against the customer requirements and unmet needs derived from the Jobs to Be Done work in Dimension 1. The competitive question is not how do our features compare to the rival’s features? but which alternative satisfies more of the customer’s prioritized requirements better, and where do we have the credible opportunity to differentiate on requirements that matter and are not being well served? Porter’s structural analysis, the Blue Ocean / Red Ocean strategic choice, and disruption theory each inform pieces of the answer — but the competitive evaluation only becomes decision-useful when the four buckets of competing alternatives are scored against the customer’s prioritized requirements rather than against each other in a vacuum.
The four buckets are non-negotiable. The hardest competitor a new B2B software tool faces is rarely a feature-comparable rival; it is the spreadsheet the buyer is using today plus the inertia of switching. The hardest competitor a new consulting offering faces is rarely another firm; it is the prospect’s decision to keep doing what they are already doing or to handle the work in-house. Both are do-nothing losses, and both are typically invisible to a competitive analysis that only enumerates direct competitors and substitutes. The dimension forces the analysis to cover the full surface of the customer’s decision, with the do-nothing option treated as a first-class fourth bucket because customers evaluating a strategy or product purchase often weight inertia, switching cost, and risk-of-disruption heavily.
Dimension 3 — Whole Solution
The third dimension describes the complete capability stack required to satisfy market requirements across the entire customer journey and product lifecycle — main product or production units plus complementary elements provided by the organization or by other actors in the value network.
The dimension draws on Moore’s whole-product construct from Crossing the Chasm (1991), with the recognition that a customer’s purchase decision rests not on the core product alone but on the complete experience the core product is embedded in: channels that deliver it, integrators who incorporate it, complementors whose offerings make it more valuable, payers who finance it, influencers who shape opinion about it. A core product without the whole-solution layer is a feature, not an offering.
The BRI framework adds two cross-dimensional connections at Dimension 3. First, the value network is part of the whole solution, not a separate consideration. Every element of the whole solution that the organization does not deliver directly is delivered by a value-network actor, and the value network’s structure, incentives, and bargaining position are first-class strategic considerations rather than execution details. The classic Blockbuster and Netflix comparison in their early days illustrates the point: both provided DVD rentals as the core product, but the whole solutions were entirely different. Blockbuster delivered through physical stores with local inventory, store-clerk recommendations, and in-person browsing. Netflix delivered through the postal service with a web-based catalog, queue management for ordering, and recommendation algorithms. Same core product, different value networks, different whole solutions, different strategies.
Second, every element of the whole solution traces back to the customer requirements from Dimension 1. The performance of each element is evaluated against the prioritized requirements the customer is hiring the solution to satisfy. The element is critical if it materially affects high-priority requirement performance; the element is risky if there is low confidence the implementation will deliver the required performance; the element is differentiating if the implementation is more exclusive than what competitors can offer for the same requirement. The dimension surfaces, by design, the gap between what we will deliver and what the customer needs to buy in total — and forces the strategy to resolve that gap explicitly rather than letting it surface as a delivery surprise late in the project.
Dimension 4 — Implementation Approach and Execution
The fourth dimension describes how the organization will develop, produce, market, and sell the whole solution and operate the business — the approach choices, the value-network performers, and the integration with Company Fit and the interlock between methodology and host organization.
The dimension draws on Hambrick’s Vehicles — the build-vs-buy-vs-partner-vs-license decision at the level of the strategy as a whole — and on Tushman and O’Reilly’s ambidextrous organization framing from Winning Through Innovation (1997) for the how to organize question for new-business work that requires structurally different resources, processes, and priorities than the core business. Bain and Zook’s adjacency analysis from Profit from the Core (2001) informs the strategic territory the implementation work has to navigate, with the caveats from the borrowed-frameworks discussion above on the adjacency framework’s limitations.
The BRI framework’s contribution at Dimension 4 integrates the implementation choices with Company Fit / RPP assessment. The right Vehicle for a given solution element depends on the element’s strategic importance, the organization’s existing competencies, the relative exclusivity available through internal versus external paths, and the fit between the element’s required capabilities and the organization’s Resources, Processes, and Priorities. Treating the implementation choices as tactical produces strategy artifacts that say the company will enter market X without specifying whether the entry is via internal build (slow but exclusive), licensing (fast but margin-constrained), partnership (fast but contingent on partner commitment), or acquisition (fast and exclusive but expensive and integration-risky). The strategic differences across those four paths are large, and the differences in execution risk are larger still — execution risk varies with how well each path fits the host organization’s RPPs. The dimension forces the implementation choices into view at the strategy level, evaluated against Company Fit explicitly rather than left to discover at execution time.
The dimension also surfaces Internal Requirements the strategy must satisfy: strategic alignment, technical and operational constraints, capacity constraints, governance constraints, alignment with the priorities of the core business. Internal Requirements are the feasibility lens that determines whether a strategically attractive Implementation Approach can actually be executed inside the organization that will pursue it. The integration of Implementation Approach with Company Fit is what makes the dimension diagnostic rather than aspirational.
Dimension 5 — Financial Logic
The fifth dimension describes the financial assumptions and implications of all elements of the strategy hypothesis and how they relate to each other financially.
The dimension draws on Hambrick’s Economic Logic, on McGrath and MacMillan’s Discovery-Driven Planning (1995, HBR), and on Osterwalder and Pigneur’s Business Model Generation (2010). Hambrick’s Economic Logic names the question: how will returns be obtained? McGrath and MacMillan operationalize the question under uncertainty with the reverse income statement and the explicit assumption checklist — start from the required outcome and work backward to what would have to be true. Osterwalder and Pigneur structure the question around revenue streams, cost structure, key resources, and key partnerships in a single visual canvas.
The BRI framework’s contribution at this dimension is to treat financial logic as a hypothesis layer explicitly linked to specific assumptions in the other five dimensions, rather than as a free-standing forecast. A revenue projection is not a Financial Logic claim; it is a calculated outcome that depends on assumptions about Target Markets (segment size and reachable share), Whole Solution (production units and pricing), Implementation Approach (cost structure and operational efficiency), and Staging (the trajectory through which scale is achieved). The dimension carries those linkages explicitly: every financial number traces back to the dimensional assumptions it rests on, and a change in those assumptions propagates into the financial model rather than orphaning the forecast.
A second refinement at this dimension is the vocabulary distinction between business model and strategy. In Osterwalder’s frame, business model refers to the entire nine-block strategic logic. In the BRI framework, business model refers narrowly to how production units are sold — one element within Financial Logic, alongside cost structure, financial advantage, and the broader unit-economics question. The broader strategic logic is captured by the full Strategy Hypothesis across all six dimensions, not by the term business model. The distinction matters for working with client teams: a team debating the business model often means three different things across the room (the revenue mechanism, the financial logic, the strategy as a whole), and the synthesis problem reappears at the level of vocabulary if the distinction is not held.
Dimension 6 — Staging
The sixth dimension describes how strategy choices across the other five dimensions will change based on different triggering conditions. Staging is the dimension that recognizes a strategy is rarely a single point-in-time configuration; it is a sequence of conditional configurations that respond to evidence, market shifts, capability development, and the cumulative state of the other five dimensions.
The dimension draws on Hambrick’s Staging. Hambrick named the question: what is the speed and sequence of moves? The borrowed framework leaves open how staging choices get made and what triggers a move from one staging configuration to the next.
The BRI framework’s contribution at Dimension 6 is to link staging choices and actions explicitly to specific triggering conditions — defined market signals, capability availability thresholds, competitive moves, customer-acquisition-cost stabilization, capital availability, regulatory clarity — rather than to calendar time or to hidden implicit assumptions about when the conditions will obtain. The discipline applies across the full lifecycle of a new business but carries the most weight during scaling, where most strategic staging choices are about how the business expands after initial market validation: when to extend from beachhead segments to adjacent segments, when to vertically integrate versus continue with value-network partners, when to license existing assets versus continue building, when to expand geographically, when to introduce adjacent products or services. Each of those choices has a triggering condition that, if named explicitly, lets the team and the investment decision-maker watch for the signal rather than rely on calendar drift or untested timing assumptions.
The triggering-conditions discipline is what separates a staging plan that anchors strategy work from a staging plan that papers over uncertainty. A staging configuration that names expand to adjacent segments in year three without naming what evidence about beachhead success would justify the expansion is a staging configuration whose timing assumption is implicit and untested. A staging configuration that names expand to adjacent segments when beachhead segment penetration exceeds 25% with positive unit-economics retention at six months is a staging configuration whose triggering condition is explicit, testable, and shared between project leadership and the investment decision-maker.
3. The Nested Loops: Inner-Loop Innovation Methodology and Stage-Gated Pipeline
At Layer 2, the work belongs jointly to the project lead and the investment decision-maker. The acute pain at this layer is what every team running new-business work actually has to do, day to day: develop and test the strategy hypothesis defined by the Strategy Framework, on a tight cadence of evidence-gathering, and submit the resulting evidence to investment-decision points that scale as the project advances. Two distinct rhythms operate here, and the puzzle is making them reinforce each other rather than fight each other. They compose as a nested loop: many inner-loop iterations within each pipeline stage; many pipeline progressions inside the enclosing portfolio governance.
What follows is the structure of each loop, the contributions the BRI methodology makes to integrating them, and the evaluation discipline that runs at every stage gate.
The scientific method as the underlying pattern
The inner loop’s epistemic foundation is the scientific method: hypothesize, design a test that could prove the hypothesis wrong, run the test, observe, revise the hypothesis, decide. This is the deepest, most-validated pattern of knowledge production in human experience, and it is the pattern every modern iterative-method tradition in innovation work is a domain-specific variant of.
Lean Startup (Ries, 2011, The Lean Startup; Blank, 2005, The Four Steps to the Epiphany) operationalizes the scientific method for startup-context customer development. The build-measure-learn loop is hypothesize-test-observe-revise wearing software-startup clothes, with the hypothesis narrowed to customer-problem-solution fit and the experiment narrowed to MVP-driven validation. Lean’s contribution is the discipline of running the loop fast and cheaply on a narrow surface; its scope is intentionally limited to the early-stage customer-development question. Applied in a mature-enterprise context, Lean’s startup-context calibration needs adjustment for the longer evidence horizons, the multi-stakeholder buying processes, the host-organization governance, and the regulatory and operational constraints typical of mature businesses.
Discovery-Driven Planning (McGrath and MacMillan, 1995, HBR) operationalizes the scientific method for new-venture financial forecasting under uncertainty. The reverse income statement and the explicit assumption checklist are hypothesize-test scaffolding for the financial dimension specifically: start from the required outcome, work backward to what must be true for the outcome to obtain, classify each must-be-true as an assumption, and design milestones that test the assumptions in priority order. DDP’s contribution is the discipline of making implicit financial assumptions explicit and testable; its scope is the financial-logic dimension.
Design Thinking (Brown / IDEO, 2009, Change by Design) operationalizes the scientific method for desirability validation through prototyping and user observation. The prototype-test-iterate cadence is hypothesize-test-observe-revise applied to the desirability question: what does the customer actually want, how do we learn what they want without asking them directly (because asking does not reliably surface the answer), how do we converge on a desirable solution through iteration. Design Thinking’s contribution is the discipline of empathy-driven customer-needs discovery; its scope is the desirability dimension.
Each is a powerful framework for the question it answers. Each is, by design, narrow. The integrated inner-loop methodology inherits from the scientific method directly and integrates contributions from each of the contemporary variants where they are most rigorously developed — Lean’s customer-development discipline, DDP’s financial-assumption discipline, Design Thinking’s desirability discipline — and adds the holistic-evaluation discipline that ties them together across all six Strategy Framework dimensions. The inner loop is not a fourth iterative-method tradition alongside the others. It is the integration of the others under the underlying epistemic pattern that produced them.
The four-step inner-loop cycle
Within every investment stage, the inner loop runs a four-step cycle, repeated as many times as the stage’s evidence requirements demand or the project’s resource limits allow:
Step 1 — Define or refine the strategy hypothesis. Comprehensive across all six dimensions of the Strategy Framework, with strategy elements explicitly classified as Choices (within the team’s control to make), Assertions (about other actors’ choices, requiring research and validation), or Uncertainties (quantitative assumptions defined as ranges representing the distribution of potential values). The classification matters operationally: each category implies a different evidence-gathering posture, and treating an assertion as a choice or hiding an uncertainty as a point estimate is one of the most common ways strategy work goes silently wrong. The hypothesis at this step carries the team’s working model of what the strategy is and what it depends on, expressed at a fidelity appropriate to the current stage.
Step 2 — Evaluate holistically. Across Desirability, Feasibility, and Viability simultaneously, applied to the full hypothesis rather than to one dimension at a time. Critical Assumption identification is integral to this evaluation pass — not a separate downstream activity. Every assumption in the hypothesis is scored on impact-on-outcome and current-confidence; the high-impact-and-low-confidence subset becomes the Critical Assumptions that get prioritized resources for evidence-gathering in Step 3. The evaluation is what makes the experimental work in Step 3 directed and disciplined rather than reactive: the team always knows which assumptions are worth which kind of investment, before they spend the investment.
Step 3 — Gather evidence. Prioritized experiments and research targeting the Critical Assumptions identified in Step 2. Evidence depth scales with stage: directional and anecdotal at early stages; rigorous and structured at mid stages; definitive and quantitatively substantiated at launch. This is where the bulk of the team’s investment and effort is applied — and should be grounded in the real world as much as possible. Lean’s MVP discipline lives here when the Critical Assumption is customer-problem-solution fit. DDP’s milestone-and-assumption-test discipline lives here when the Critical Assumption is a financial-logic uncertainty. Design Thinking’s prototype-and-observe discipline lives here when the Critical Assumption is desirability. The inner loop selects from the contributing variants based on what kind of assumption is being tested, rather than committing to one variant as the universal experimental form.
Step 4 — Decision. An explicit choice at the end of each iteration: keep iterating within the current stage, refine the hypothesis meaningfully, or push toward the stage gate. At the stage gate itself, the decision becomes one of the four canonical outcomes the outer loop runs on: Continue / Pivot / Pause / Stop.
A continuous apply-learnings thread runs across all four steps. Every iteration carries forward what was learned in the previous one. Within each step, new evidence feeds immediately back into the working hypothesis. The four steps are an analytical decomposition; in practice the work runs as a continuous cycle with the steps rebalanced iteration by iteration.
The outer loop — stage-gated investment
The inner loop runs within an investment stage. The outer loop is the stage-gated investment structure that bounds the inner loop’s iterations and decides when the project advances. The outer loop’s epistemic ancestors are Cooper’s stage-gate process from Winning at New Products (Cooper, 1986+) and the tranched-investment discipline developed in venture capital — both structures of progressive investment commitment scaling with demonstrated risk reduction. Intel’s pioneering work in Corporate Venture Capital was one of the substantive contributing experiences that brought the tranched-investment posture into internal new-business development; the BRI outer loop draws on both lineages.
At each stage gate, the investment-decision input is three things in combination: the cumulative evidence the team has accumulated across the multiple inner-loop iterations run inside the stage; the project’s relative strength against other projects in the portfolio (provided by the governance layer above); and the project’s read against the stage-and-class-specific evaluation criteria treated in the next subsection. The three inputs together resolve the gate decision into one of the four canonical outcomes the outer loop runs on.
The first refinement at the outer loop is the four canonical stage-gate outcomes — Continue / Pivot / Pause / Stop — replacing the binary Go / Kill posture or the Go / Kill / Hold / Recycle variant that some stage-gate practice carries forward. The vocabulary distinction is substantive, not cosmetic. Continue advances to the next stage with committed investment. Pivot adjusts hypothesis direction meaningfully — may re-enter the current stage with a revised hypothesis or move forward with changed scope. Pause halts active investment pending a future condition: a market change, internal resource availability, or evidence availability. Stop terminates the project and reallocates resources.
Two of the four are organizationally consequential in ways the binary form does not capture. Stop as a first-class outcome is portfolio optimization, not failure; this framing is what makes early-stage stopping organizationally normal rather than a political event. The team that Stops a project at Stage 2 has not failed; they have generated evidence that the bet is not worth the increment of investment the next stage would require, which is one of the most resource-optimizing outcomes any inner-loop work can produce. Pivot as a first-class outcome reflects the fundamental hypothesis revision common in new-business work; it is distinct from Cooper’s Recycle, which connotes returning the project to an earlier stage with the same hypothesis. Pivot preserves the project but revises the hypothesis — which is what discovery-driven work actually does when evidence contradicts the working assumptions.
The second refinement at the outer loop is to gate on evidence-readiness rather than on the calendar. Projects advance when the evidence required by the stage’s stage-and-class-specific criteria is in hand, taking only the time and resources minimally needed to gather that evidence. This evidence-readiness cadence is fundamentally different from the typical corporate calendar-based budget cycle that allocates resources by fiscal period regardless of project state.
The third refinement is that fidelity rises with investment scale at each gate. Each gate does not just gate on more evidence; it gates on higher-fidelity evidence, because the next stage’s investment increment justifies more rigorous hypothesis testing. The granularity of the strategy hypothesis model and the depth and rigor of supporting evidence both rise stage by stage, while the underlying four-step inner-loop cycle remains the same. Three fidelity levels are the default articulation. Low Fidelity (LoFi) is rapid order-of-magnitude framing — a few key choices and assumptions per strategy framework dimension, simple ranges on the financial uncertainties, anecdotal evidence drawn from existing experience and adjacent observation. Mid Fidelity (MidFi) is fuller hypothesis with back-of-envelope financials, explicit assumption work, and structured evidence from interviews, prototypes, or market research. High Fidelity (HiFi) is the full toolset: market segmentation models, market sizing with multi-variable conversion tables, unit economics, operating-expense and cashflow forecasts, definitive customer evidence, fully specified production units, and finalized Implementation Approach choices for every solution element. A LoFi cycle and a HiFi cycle are recognizably the same four steps; what scales is scope, modeling depth, evidence standards, and investment.
Evaluation discipline at each gate — DFV mapped to the six Strategy Framework dimensions
The criteria the gate decision draws on are organized along the Desirability / Feasibility / Viability lens — DFV — drawn from Brown and IDEO’s Change by Design (2009). DFV is rigorous, well-attributed, intuitive to client teams, and resolves the broader how do we evaluate this concept question into three distinct evidence questions: does the strategy respond to a real, motivating unmet need in a viable target market (Desirability), can the organization actually build, deliver, and operate the whole solution (Feasibility), and is the strategy financially sustainable and internally coherent for the organization that will pursue it (Viability).
DFV is the right evaluation lens because the three questions are the right three questions to keep in simultaneous view at every gate. Concepts that look strong on Desirability but weak on Feasibility are stranded by execution; concepts that look strong on Desirability and Feasibility but weak on Viability are stranded by the host organization’s ability or willingness to sustain them. A gate decision that does not balance all three lenses produces predictable failure modes — the desirable-but-infeasible concept that gets greenlit on customer enthusiasm, the feasible-but-undesirable concept that gets greenlit on technical confidence, the desirable-and-feasible-but-unviable concept that gets greenlit on momentum and stalls in the host-organization interdependencies.
The BRI refinement of DFV is to map the three lenses onto the six Strategy Framework dimensions rather than substituting for them. Desirability draws on Target Markets and Unmet Need plus Competitive Differentiation. Feasibility draws on Whole Solution plus Implementation Approach and Execution. Viability draws on Financial Logic plus the viability-side concerns inside Implementation Approach and Execution — strategic alignment, internal organizational constraints, Company Fit / RPP signals. Each dimension carries DFV-evaluable assumptions; each gate decision draws on dimension-specific DFV evidence. The mapping is what gives the configurable-criteria structure its operational granularity.
The criteria themselves number in the dozens. They are derived from decades of practitioner experience working with leading frameworks across many engagements, customized and prioritized per client and per portfolio. The right criteria for a Stage 2 Core project in an established consumer goods company are not the same as the right criteria for a Stage 2 Disruptive project in an emerging-technology venture portfolio, and the relevant quantity grows across stages, fidelity, and investment level: a Stage 1 LoFi gate operates on a smaller, more directional subset; a Stage 4 HiFi gate operates on a much larger, much more rigorous subset.
A worked example of Continue / Pivot / Pause / Stop
Consider a Stage 2 Mid-Fidelity gate decision on a B2B software hypothesis being evaluated inside an established industrial-equipment manufacturer. The project has completed three inner-loop iterations: customer-development sprints with twelve prospects, a back-of-envelope financial model with priced unit economics, and a working prototype tested on three of the prospects.
The Desirability evidence is mixed: eight of the twelve prospects expressed strong interest, three expressed moderate interest, one declined to engage. The Feasibility evidence is positive: the prototype demonstrated the core technical claim. The Viability evidence is uncertain: the financial model shows a viable unit economics at scale, but the Implementation Approach assumes an internal sales motion the company has never run before, and the Company Fit work has surfaced an RPP mismatch around the sales-team priorities the host organization currently has in place.
A binary Go / Kill decision on this hypothesis is misleading either way. Go overstates Viability confidence and pushes the hypothesis into Stage 3 with an unresolved sales-motion question. Kill terminates a hypothesis whose Desirability and Feasibility evidence is genuinely strong.
The four-outcome version surfaces three credible decisions. Pivot — adjust the Implementation Approach to a partner-led sales motion that uses a complementor with the relevant sales infrastructure, re-test the unit economics under the partner share, re-enter Stage 2 with the revised hypothesis. Pause — hold the hypothesis pending a clearer signal from corporate strategy on whether the host organization is about to invest in a direct-sales build-out that would resolve the RPP mismatch. Continue — advance to Stage 3 with explicit recognition that the sales-motion question is the largest residual uncertainty and that Stage 3 evidence-gathering will prioritize it.
The Stop outcome is also live — if the Viability assessment lands strongly enough on the negative side that the hypothesis is judged not worth the Stage-3 investment increment under any reasonable Implementation Approach. None of these four is a defeat; each is the right answer to a different reading of the evidence. The four-outcome vocabulary is what lets the team and the investment decision-maker have the actual conversation, rather than collapsing it into the binary defense the Go / Kill posture forces.
4. Innovation Portfolio Strategy
At Layer 3, the work belongs to the portfolio manager — the head of innovation, the head of corporate development, the executive team allocating capital across multiple in-progress new-business projects. The acute pain at this layer is the question every executive sponsor of an innovation program eventually has to answer: is the aggregate result of this portfolio actually going to meet the strategic intent the company set out to pursue?
That single question unpacks into the operational questions the portfolio manager works with day to day: Is the pipeline flow healthy, or is the portfolio heading into feast and famine? Does the current mix manage risk and return appropriately for the strategic moment? Are the right kinds of bets in the right stages? Is the company getting a defensible return on its investment in the portfolio as a whole? These are the questions a portfolio strategy has to answer continuously, with current evidence, in a form the executive team can act on.
The borrowed frameworks at this layer each contribute pieces. Tushman and O’Reilly’s ambidextrous organization (1997, Winning Through Innovation) provides the structural insight that explore work and exploit work require different organizational treatment, and is prescriptive about creating tailored governance approaches for the explore side. Bain and Company and Zook’s Profit from the Core (Zook, 2001) frames portfolio allocation across core, adjacent, and disruptive territory — with the limitations that the adjacency guidance is based on correlated rather than causal outcomes, and the foundational stick to the core advice is conditional on the core having sufficient growth potential and not facing near-term disruption threats. Moore’s Zone to Win (2015) articulates the operational discipline of running a continuously-updated portfolio view across performance / productivity / incubation / transformation zones, and is prescriptive about how governance approaches need to evolve as ventures mature and scale across the zones.
These three frameworks together cover substantial portfolio-layer territory. What none of them provides — even with Moore’s zone-based governance evolution included — is a clear diagnostic for predictively characterizing the risks of pursuing specific adjacencies, and a methodology for allocating investment under the different failure-rate assumptions that different adjacencies or maturity stages actually carry. The portfolio strategy at Layer 3 has to close those gaps, and what follows is how the integrated system does that work.
A portfolio is not a roadmap
The first move at this layer is to draw a distinction the borrowed vocabulary does not always hold. A roadmap is a sequence of intended deliverables: features the team plans to release, projects the company plans to launch, milestones the leadership plans to hit. A portfolio is a population of opportunities being progressively winnowed under uncertainty. The two artifacts look superficially similar — both are visual representations of multiple in-flight projects across time — and they are governed by completely different logic.
A roadmap commits to the deliverables; the management challenge is execution. A portfolio commits to the betting structure; the management challenge is intentional risk distribution with strategic intent. Most projects in a portfolio will not reach launch — by design — and the disciplined stopping of weak projects early is the mechanism that frees resources for the strongest. Treating a portfolio as a roadmap is one of the most common diagnostic errors in mature companies whose innovation operations look like pipelines but lack the portfolio governance that gives pipelines their meaning.
What a portfolio strategy specifies
A complete portfolio strategy specifies seven things: a Portfolio Objective, a Portfolio Mix, a Class Profile for each project class, resource allocation targets, pipeline targets, time-in-stage targets, and evaluation criteria at the intersection of stage and project class.
The Portfolio Objective is the specific growth outcome the portfolio is designed to achieve, expressed in terms leadership can hold the portfolio accountable to: revenue contribution by year N; market entry into specific adjacencies; defensive coverage against specific disruption threats; capability development that supports the core business’s long-term position. The objective is what gives the rest of the portfolio strategy its meaning — without an explicit objective, the portfolio is a description of activity rather than a structured bet on a strategic outcome.
The Portfolio Mix is the intentional distribution across project classes. Following the three-class portfolio taxonomy that Nagji and Tuff (2012, HBR) articulated and that draws on Christensen’s (1997) Disruptive Innovation framing and Bain and Zook’s (2001) adjacency framing, three classes are operative: Core Innovation (enhancements to existing products or services within existing markets), Adjacent Innovation (extensions into new markets or adjacent products for existing markets), and Disruptive Innovation (entirely new products or services creating new markets). Each class has a distinct risk profile, time horizon, investment scale, and expected survival rate. The Mix is the calibration of the portfolio against the Portfolio Objective and the organization’s risk tolerance — a portfolio whose objective is defensive maintenance of the current core needs a different Mix than a portfolio whose objective is opening a new platform-scale franchise.
The Class Profile for each class makes the Mix operational. For each project class, the Class Profile specifies expected per-project investment, expected time horizon, expected survival rate, and expected revenue characteristics if the project succeeds. These are the underlying assumptions a portfolio plan is built on; they are also what gets compared against actuals as the portfolio runs.
Resource allocation targets, pipeline targets, and time-in-stage targets translate the Mix and Class Profiles into operational guidance. How many Disruptive-class projects should be in Stage 1 at any time. How long should a Core-class project be expected to spend in Stage 2. What share of the portfolio’s annual investment goes to Adjacent versus Disruptive bets.
The seventh element — evaluation criteria at the intersection of stage and project class — is where the portfolio layer reaches down into the nested-loop work. The right evaluation lens for a Stage 2 Core project is fundamentally different from the right lens for a Stage 2 Disruptive project. A Disruptive project evaluated against Core-business criteria almost always looks bad: the customer base is unproven, the financial model carries wider uncertainty ranges, the Implementation Approach requires capabilities the host organization does not yet have. Holding the Disruptive project to the Core project’s evidence standards effectively kills it before it has a chance to develop. The portfolio strategy is what specifies the right evidence standards for each class, at each stage, so the inner-loop and stage-gate work actually serves the portfolio’s strategic intent rather than working against it.
The portfolio as alternative governance
Stepping back from the seven specifications: what is actually being defined in a complete portfolio strategy is, in substance, an alternative governance model. The Portfolio Objective specifies different objectives than the company’s default corporate objectives. The Portfolio Mix specifies different risk distribution than the company’s default risk posture. The evaluation criteria specify different evidence standards than the company’s default decision-making norms. The cadence of stage-gate decisions specifies different reporting and decision rhythms than the company’s default budget cycle.
Innovation portfolio strategy is fundamentally uncertain and exploratory: a population of bets being progressively winnowed under uncertainty, where most projects will not reach launch — by design. That exploratory posture is what creates the governance tension. The host organization’s default Resources / Processes / Priorities are typically calibrated for predictable Core-business performance — budget cycles tied to forecast quarters, metrics designed for proven business models, decision-making norms tuned to high-confidence reporting. The portfolio’s exploratory work runs on different rhythms, demands different evidence, and produces different signals. That structural gap is the source of portfolio-level RPP mismatch — and the reason a complete portfolio strategy has to design alternative governance as a first-class strategic deliverable rather than relying on the host organization’s defaults.
Moore’s zone-based governance approach is one of the more rigorously developed treatments of how governance should evolve as ventures move from transformation through incubation into productivity and performance zones — the appropriate governance pattern is different in each zone, and the transitions are governance-change events that the portfolio strategy has to plan for. The integrated system draws on this lineage and extends it with the predictive adjacency-risk diagnostic and the failure-rate-adjusted allocation methodology the borrowed frameworks leave open.
A portfolio strategy that papers over the RPP mismatch by claiming it can run exploratory Disruptive bets under Core-business governance is the portfolio strategy that most often produces the executive frustration of ten projects in flight and nothing reaching launch.
What this means operationally is captured by the distinction between pipeline and portfolio: the pipeline is process (how a single project moves through stages); the portfolio is intentional risk distribution with strategic intent and matching governance. Conflating the two collapses the portfolio strategy into a project-tracking spreadsheet.
Pipeline aggregation and cross-project resource allocation
Two operational disciplines run continuously inside the portfolio layer.
The first is pipeline aggregation. The Innovation Pipeline gives stage-gated discipline at the project level — each project moves through stages with growing fidelity, growing investment, and growing evidence requirements. The portfolio aggregates across the in-flight set: how many projects are in which stages, with which fidelity, against which class profiles, with what cumulative investment to date, with what cumulative evidence-readiness for the next gate. The aggregate view is what tells the executive team whether the portfolio is balanced for the strategic moment — whether the early-stage funnel is wide enough to feed late-stage launches three years from now, whether the late-stage queue carries the kinds of bets the Portfolio Objective requires, whether the cross-class Mix is drifting from the calibration the strategy specified.
The second is cross-project resource allocation. The investment-decision-maker discipline shifts at this layer. The question at the project level is should this project Continue / Pivot / Pause / Stop given the evidence on this project? The question at the portfolio level is of these N projects competing for resources, which deserve the next increment of investment given the Portfolio Objective and the cross-portfolio evidence picture? The two questions are related but distinct, and the portfolio-level question is the harder one. It is also where most innovation portfolios fail in practice: the project-level questions get answered honestly project by project, the cross-project comparison never happens, and the portfolio drifts into whatever Mix emerges from the path-of-least-resistance decisions made at each gate independently.
The portfolio view as executive-layer strategic instrument
The portfolio view is the artifact the C-suite reviews. It must speak the strategic-objective language — growth rates, return profiles, time-horizons, capability development, defensive coverage, adjacency expansion — while still resolving down to project-level evidence the executive team can interrogate when a particular bet is the question of the moment. The integrated system makes that resolution explicit: the portfolio view rolls up from the same Strategy Hypothesis Model the project work runs in, the same DFV evaluation lens the stage gates run on, the same configurable evaluation criteria each project is judged against. There is no translation layer between project-level evidence and the portfolio’s executive view, because both views read from the same underlying structure.
The absence of that translation layer is the design feature that makes the portfolio actually decision-useful at the executive layer. A portfolio view that requires the team to manually summarize ten projects into a deck for each executive review session is a portfolio view that is always slightly out of date, always slightly biased toward whatever framing the deck author chose, and always slightly disconnected from the evidence the project teams are actually working with. The integrated system replaces that translation layer with a continuously-updated portfolio view that resolves to project-level evidence on demand.
5. What the Integrated System Adds
The integrated system is more than the borrowed frameworks held together. The most important contribution is not any of the operational pieces named below. It is the integration discipline itself — the systematic work of designing the interfaces and relationships between the borrowed frameworks so the overlaps land intentionally rather than accidentally, the boundaries are explicit rather than fuzzy, and the foundational assumptions get reconciled rather than carried forward implicitly. The five specific contributions that follow are instances of this discipline applied to particular interfaces.
The integration discipline as the meta-contribution
Every borrowed framework in the field has a periphery where it touches another. Lean Startup’s customer-validation experimentation has a periphery where it touches Cooper’s stage-gate investment cadence. Discovery-Driven Planning’s reverse income statement has a periphery where it touches Osterwalder’s value-proposition canvas. Hambrick’s Strategy Diamond has a periphery where it touches Moore’s whole-product construct. The peripheries are where the integration question lives, and they are where most informal stitching of frameworks falls down: the periphery gets treated as the leftover space between two well-developed bodies of work, rather than as a design surface that needs deliberate attention.
The work BRI has done over the years on this front has the character of interface design — the explicit treatment of each periphery as something to be specified, evaluated, and refined. What evidence does the inner-loop methodology produce that the outer-loop stage-gate consumes? What format must that evidence be in? What dimension of the strategy hypothesis does each framework’s contribution map onto? Where does a contribution from one framework supersede a contribution from another, and where do they coexist on different surfaces of the same dimension? The questions are not glamorous. They are the operational substrate on which a working integration depends.
The reason the integration discipline matters more than any of the specific contributions that follow is that the specific contributions are themselves outputs of the discipline. The dimensional integration that absorbs multiple borrowed frameworks under each Strategy Framework dimension, the assumptions taxonomy, the fidelity model, the portfolio-configurable methodology layer, the RPP / Company Fit alignment — each is a place where the integration discipline was applied to a particular interface and produced a refinement that is now treated as canonical. Without the discipline, the contributions would not have surfaced. With the discipline, more contributions will surface as the practice continues — and the integrated system will continue to absorb new framework variants without losing its structural coherence.
Contribution 1 — Borrowed frameworks integrated under a coherent dimensional whole
The first contribution sits at Layer 1 itself: the six-dimension Strategy Framework provides a coherent structural whole that absorbs multiple narrowly-focused borrowed frameworks under each dimension, rather than leaving the practitioner to stitch them together informally.
The Target Market and Unmet Need dimension absorbs Jobs-to-Be-Done in the Ulwick and Christensen lineage, Customer Value Propositions, and segment-selection thinking — each becoming an evaluable dimension of the strategy hypothesis rather than a freestanding method applied in parallel. The Competitive Differentiation dimension absorbs Porter’s Five Forces, the Blue Ocean and Red Ocean framing, and Christensen’s disruption theory — each contributing to a single Competitive Differentiation claim evaluated against the customer requirements from Dimension 1. The Whole Solution dimension absorbs Moore’s whole-product construct alongside the value-network treatment that ties solution performance to customer requirements. The Implementation Approach and Execution dimension absorbs Hambrick’s Vehicles, Bain and Zook’s adjacency thinking, and Tushman and O’Reilly’s ambidextrous-organization framing — integrated with Company Fit / RPP assessment that informs the build / buy / partner / license choices and their associated execution risk. The Financial Logic dimension absorbs McGrath and MacMillan’s Discovery-Driven Planning alongside Osterwalder’s business-model articulation. The Staging dimension takes Hambrick’s Staging and applies the triggering-conditions discipline that links staging choices explicitly to specific signals rather than to calendar time. Underneath all six, Collins and Porras’s Vision Framework anchors the Strategy Objective to the company’s foundational purpose.
The contribution is not that any individual borrowed framework was wrong about what it covers; each is rigorously right within its scope. The contribution is that the six-dimension structure makes it possible to integrate them coherently — every borrowed framework’s contribution lands at its rightful dimension with attributes that interact across dimensions through a shared underlying structure rather than colliding informally in a practitioner’s working notes.
Contribution 2 — The assumptions taxonomy (cross-framework foundational reconciliation)
The first contribution named the dimensional structure that absorbs multiple borrowed frameworks coherently. The second contribution gives that integration its operational discipline: the assumptions taxonomy creates a consistent classification of every element of a strategy hypothesis — as a Choice (within the team’s control to make), an Assertion (about other actors’ choices, requiring research and validation), or an Uncertainty (a quantitative assumption defined as a range representing the distribution of potential values) — that holds across the borrowed-framework contributions absorbed under each dimension.
The discipline looks structural and is structural — but the operational payoff is what it surfaces about how the absorbed borrowed frameworks rest on different foundational assumptions. A team running Lean Startup is operating under an assumption about what counts as evidence (validated learning from MVP-driven customer-development sprints). A team running Discovery-Driven Planning is operating under an assumption about what counts as a strategy choice (a milestone-bounded commitment to a financial trajectory). A team running Design Thinking is operating under an assumption about what counts as a decision (the convergence of an iterative prototyping process on a desirable solution). When a practitioner stitches three frameworks together informally, these foundational assumptions stay implicit, and the team that cannot agree on whether to Continue / Pivot / Pause / Stop a project is usually a team whose reviewers are operating under different unreconciled foundational assumptions about what evidence carries the most weight.
The taxonomy makes the reconciliation explicit. Choices are committed; assertions are validated; uncertainties are quantified as ranges that tighten with evidence. Each category carries a different evidence-gathering posture — one of the consequential implications, both for decision-makers and for project leads, that uncertainty-as-ranges produces. For decision-makers, ranges surface inherent uncertainty directly and discourage anchoring on a specific expected outcome; funding decisions become appropriate to the level of uncertainty actually present. For project leads, ranges remove the implicit pressure to deliver on a specific number — leaders are typically more willing to engage in quantitative modeling when they don’t feel obligated to defend a single point estimate as a personal commitment. The taxonomy frames uncertainty as the team’s shared object of work rather than as a failure mode each individual is held personally accountable to overcome. The cross-framework discipline of making implicit assumptions visible at framework boundaries — and then choosing the appropriate evidence-gathering posture for each — is the load-bearing value the contribution provides.
Contribution 3 — The fidelity model
The third contribution applies across every dimension of the strategy hypothesis. The fidelity model treats LoFi to MidFi to HiFi concept maturation as an explicit dimension of methodology progression, with fidelity-appropriate evidence requirements at each stage and across every dimension simultaneously.
The contribution is what makes the same underlying inner-loop cycle work across the full lifecycle of a new business — from earliest concept exploration through committed market launch. Without explicit fidelity scaling, teams face an impossible choice between a methodology rigorous enough to support a Stage-4 launch decision (and far too heavy for a Stage-1 ideation pass) and a methodology lean enough to support Stage-1 work (and far too thin for a Stage-4 commitment). With explicit fidelity scaling, the same four-step cycle runs at every stage, with hypothesis granularity, modeling depth, and evidence standards that scale to the stage’s investment level.
The interface logic the contribution surfaces is subtle but consequential. A LoFi hypothesis is not a worse hypothesis than a HiFi hypothesis; it is the right hypothesis for the stage. A Stage-1 LoFi hypothesis that survives an honest evaluation against LoFi-appropriate evidence is exactly as decision-useful as a Stage-4 HiFi hypothesis evaluated against HiFi-appropriate evidence. The mistake teams make without the fidelity model is treating LoFi work as a rough draft to be polished into HiFi work, rather than as the right work for the stage it lives in.
The fidelity model applies dimension-by-dimension, with each dimension scaling its fidelity in step with the others so the strategy hypothesis as a whole maintains internal coherence at every stage. The contribution makes the fidelity question an explicit feature of the methodology across all dimensions, not an emergent property practitioners have to figure out engagement by engagement.
Contribution 4 — The portfolio-configurable methodology layer
The fourth contribution sits at the interface between the methodology and the portfolio. The portfolio-configurable methodology layer means that the methodology adapts its emphasis, criteria, and cadence to the host company’s portfolio strategy and Innovation Pipeline configuration.
The contribution responds to a failure pattern visible repeatedly in practice. A methodology that worked beautifully for a software startup’s customer-discovery sprints implodes when applied to an industrial-equipment manufacturer’s hardware-platform play. The same four-step cycle, the same scientific-method foundation, the same DFV evaluation lens — and the methodology fails at the manufacturer because the configuration is wrong. The hardware-platform play needs longer Stage-1 fidelity work than the software startup did; the evidence standards for hardware feasibility differ from the evidence standards for software feasibility; the stage durations are calibrated to capital-equipment timelines rather than software release cycles; the evaluation criteria at each stage-class intersection differ.
The contribution is to make the methodology configuration an explicit deliverable of the portfolio strategy, not an emergent property of the project teams’ habits. Stage definitions and counts; available tool sets at each stage; evaluation criteria at the intersection of stage and project class; evidence standards; target ranges for investment, headcount, survival rates, and durations — all are configurable, all are specified by the portfolio strategy, all are the operating context in which each project’s inner-loop and stage-gate work runs. The conceptual framework is fixed; the operational instantiation is configurable per portfolio.
Contribution 5 — RPP / Company Fit alignment
The fifth contribution sits at the interface between the methodology stack as a whole and the company that will run it. The Company Fit / Resources, Processes, and Priorities (RPP) alignment work — drawing on the framework Christensen and Kaufman articulated in Assessing Your Organization’s Capabilities: Resources, Processes, and Priorities (2006, HBR) — addresses the hidden problem from the fourth observation in the opening: the methodology that worked in one company stalls or fails in another, even when the strategy is sound and the team is capable.
The contribution is to operationalize RPP into stage-gate criteria, into methodology selection and configuration, into portfolio aggregation patterns, and into the cumulative-impact failure-pattern detection that makes the interlock between methodology and host organization diagnosable before it becomes terminal. RPP is one of the most important diagnostic frameworks in innovation work, and it is also one of the most consistently under-applied. The framework is widely known. The operational discipline of applying it during strategy work, evaluating it at each gate, configuring methodology to it at the portfolio level, and watching for cumulative-impact symptom patterns when small misalignments accumulate into compounded execution friction — that operational discipline is rare in practice, and it is what separates the methodology adoption that lands clean on the slide and collapses in the work from the methodology adoption that actually changes how the organization operates.
The cumulative impact pattern is worth naming explicitly.1 Small RPP misalignments accumulate into the compounded friction that makes a project stall on issues nobody can articulate, when the evidence is solid and the team is capable but the project repeatedly runs into small roadblocks or delays when working with other parts of the organization. Each individual misalignment is small, but the compounded effect of many small misalignments is what ultimately kills the project, either by slowing it to the point of non-competitiveness or by forcing numerous strategy compromises that erode the value or competitiveness. Few practitioners diagnose the pattern; fewer still know how to design around it. Operationalizing RPP into the integrated system — at the project level via the Implementation Approach dimension, at the methodology level via configurable stage-gate criteria, at the portfolio level via portfolio-strategy alignment with the host organization’s actual resourcing and decision-making norms — is what makes the interface between methodology and host organization a first-class design dimension rather than the afterthought it usually is.
The RPP integration discipline is not a methodology fix; it is a strategy diagnostic and governance discipline that the integrated system carries from project work all the way up through portfolio governance. It is also the contribution, of the five, that most practitioners will recognize as the answer to a question they have asked many times before without finding the operational discipline that resolves it.
6. Operationalizing the Integrated System — Consulting Delivery and Growth Forge® Software
The integrated system is a methodology. Like any methodology, it has to be operationalized to do real work in real organizations. BRI delivers the integrated system in two coordinated modes that complement each other: through the consulting practice that introduces, customizes, and operates the methodology with client teams; and through Growth Forge® Software, the optimization layer that makes consistent application of the integrated system practical at scale.
Consulting delivery — the foundational mode
BRI’s consulting practice delivers the integrated system through documentation, templates, training, and direct coaching with client teams. The consulting mode is the foundational delivery mode, and it remains the right mode for a wide range of engagements: early customizations of the methodology to a specific client’s portfolio configuration; deep work alongside leadership teams designing the alternative governance the portfolio’s bets require; hands-on coaching of project teams running the inner-loop and stage-gate work for the first time; high-fidelity Strategy Foundations work at the start of a major new-business commitment. The consulting work is what introduces the integrated system, calibrates it to the client’s actual operating context, and gets it running in the practical discipline of the client’s teams.
What consulting delivery is honestly difficult to do at scale, however, is operating the integrated system consistently across many projects, across many teams, across many decision cycles, across many years. The integration discipline depends on persistent objects that hold the strategy hypothesis across iterations, on collaborative access that lets multiple team members work on the same hypothesis without forking, on evaluable structure that resolves the dimensions of the hypothesis into concrete dimensional evidence, and on cross-team-cross-time coherence that ties project work to portfolio governance. Maintaining all of that manually — through document handoffs, template revisions, training refresh cycles, and coaching engagements — is possible but labor-intensive, prone to drift, and difficult to scale without dedicated methodology specialists embedded in every team.
Growth Forge® Software is the optimization layer that addresses this. It does not replace the consulting work; it makes the result of consulting work cost effective and easier to operate consistently at scale.
What structured workflow enables
Growth Forge supports the integrated system by replacing the default state of innovation work in a corporate environment — the fragmented patchwork of PowerPoint templates and Excel models that each team builds for itself — with a structured workflow embedded in a structured data model. The strategy hypothesis is a persistent object held in the data model, not a deck the team rebuilds for each review. The data model carries the six dimensions, the underlying assumptions classification (Choice / Assertion / Uncertainty), the DFV evaluation lens, the fidelity level, the stage progression, the evaluation criteria, the evidence attached to each assumption. The team works on the strategy hypothesis object; the object is what evolves; the artifact for any given review is a view of the object at that moment.
The structured workflow does six things that consulting delivery alone has to maintain manually. First, it persists the strategy hypothesis across teams and across time — the object that a Stage 1 ideation pass produced is the same object that the Stage 4 launch decision evaluates, not a separate document with the same name. Second, it supports complete dimensional coverage — the workflow guides every dimension being addressed, every assumption being classified, every Critical Assumption having prioritized evidence-gathering, every gate decision resting on dimension-specific DFV evidence. Third, it enables collaboration across the team — multiple project members work on the same object simultaneously, without forking or merging document copies. Fourth, it makes the dimensions evaluable — the DFV evaluation criteria run as functions of the dimensional content, not as opinions about the document. Fifth, it surfaces inconsistencies across the strategy hypothesis — assumption conflicts, evidence gaps, dimensional contradictions get flagged at the data-model level. Sixth, it makes cross-project comparison and portfolio aggregation operationally tractable — the portfolio view is a roll-up of the same data-model structure that every project is working in, with no manual translation layer needed.
These are not replacements for what consulting delivery does; they are the operational gains that come from putting the integrated system on a structured workflow. The consulting practice still introduces the system, calibrates it, coaches teams through it, and works alongside leadership on the alternative-governance design. Growth Forge makes the day-to-day operation of the system consistent across teams and time without requiring a methodology specialist on every project.
The Growth Forge® tools as the operational surface
Growth Forge organizes the operational surface across three coordinated tool sets.
Strategy Modeling Tools are where the strategy hypothesis is developed and refined dimension by dimension. Strategy Aspiration and Strategy Builder establish the strategic intent and the choices and key beliefs supporting it. Jobs to Be Done, Requirements, and Value Propositions develop the demand-side discipline at progressively higher fidelity, anchoring Target Markets and Unmet Need work in solution-agnostic, evaluable customer-need articulation. Solution Definition, Implementation Planning, and Company Fit Risks develop the supply-side and execution dimensions, connecting the proposed whole solution to performer-level Implementation Approach choices and to internal-alignment risk assessment. Competitive Analysis closes the Competitive Differentiation dimension against the four-bucket competing-alternatives discipline. Market Segmentation, Market Sizing, Unit Economics, Expense Planning, and Back-of-the-Envelope P&L close the Financial Logic dimension at progressively higher fidelity.
Project Management Tools are where the project’s stage-gated work runs. The Project Dashboard provides project-level visibility into status, milestones, and outstanding decisions. Strategy Evaluation provides the complete summary that anchors the gate decision. Assumptions Manager organizes and prioritizes all assumptions in the project based on level of uncertainty and impact to success, and surfaces potential inconsistencies between assumptions. Evidence Gathering Manager defines and manages experiments and other evidence-gathering activity targeted at Critical Assumptions. Decision Making frames the recommendation, supporting assumptions and evidence, and the specific decisions for the next stage or decision point.
Portfolio Management Tools are where the portfolio strategy lives operationally. Portfolio Modeling models assumptions about investments and portfolio pipeline performance — to determine either how much option value could be generated based on a specific investment fund size, or the size of investment fund required to achieve a specific growth objective. The Portfolio Dashboard provides the cross-project view: project status relative to the methodology, stage and resource-consumption status, cumulative and comparative analysis using evaluation criteria organized in the DFV framework, and tracking of external projects not modeled in Growth Forge. The dashboard is the no-translation-layer portfolio view — the executive team can review the portfolio view with confidence that it reflects the project-level evidence, and can interrogate any specific project’s evidence directly when a particular bet is the question of the moment.
The tools share a common underlying data model, which is what makes their integration work. Production units and sales-model structure defined in Solution Definition are immediately available in Market Sizing. JTBD work surfaces in Requirements work. Implementation Planning’s performer choices flow into Company Fit Risks. The integration is built into the data model rather than coordinated by user discipline. Users are not required to complete tools in a fixed sequence — inputs defined in one tool are automatically available in all related tools.
The configurable-workflow analogy
Growth Forge is described in shorthand as TurboTax for new-business strategy. The analogy is intentional. TurboTax did not replace the tax code or replace the work of preparing a tax return; it replaced the patchwork of forms with a structured workflow that adapts to each filer’s specific situation. The underlying tax code is fixed; the workflow surfaces the right questions for that filer’s complexity, validates inputs, and produces a return that is structurally consistent with the underlying code while being calibrated to the filer.
Growth Forge does the same for innovation work. The underlying methodology is fixed — the dimensional structure, the four-step inner loop, the four canonical stage-gate outcomes, the DFV evaluation lens, the Strategy Hypothesis Model. The workflow surfaces the right tools at the right fidelity for each project’s stage and class, validates assumptions and surfaces inconsistencies, and produces strategy and evaluation artifacts that are structurally consistent with the underlying methodology while being calibrated to the portfolio’s configuration. The methodology framework is fixed; the operational instantiation is configurable per portfolio.
What about AI?
Generative AI makes using Growth Forge faster and easier. Growth Forge integrates AI capabilities at discrete points in the structured workflow. Each integration point preserves the structural integration — AI assistants help populate and analyze the strategy hypothesis model, for human decisions; the persistent, collaborative, evaluable model remains the source of truth.
Three integrations are deployed today. The AI Industry Analyst Agent, when enabled, continuously monitors news and data sources relevant to a project’s industry and presents strategic implications for the team’s consideration rather than making autonomous decisions. AI Discovery Assistants generate starting points for key elements of the strategy hypothesis — Jobs to Be Done, Requirements, Value Propositions, Solution Definition, Value Network build-out, Competitive performance against requirements, Implementation Approach — for users to evaluate, modify, and build upon. AI Summary Assistants organize and distill complex data sets into clear, structured information for the team to use.
Two additional integrations are currently in development. AI Analysis Assistants, deployed in the Analysis tab of each tool, will provide widget-specific and chart-specific interpretation as well as aggregate summary analysis across the whole tab or tool. The Strategy Evaluation AI Summary & Analysis will summarize the key choices and assumptions for each strategy dimension and perform SWOT analysis for medium-fidelity and high-fidelity strategy hypotheses.
The design principle behind all AI features in Growth Forge is that technology should serve innovation teams, not lead them. AI features are user-initiated, transparent in their processes, and require explicit user approval before AI-generated results become part of a project. Project data remains isolated and secure. AI capabilities in Growth Forge are evolving rapidly and progressively rather than as a single launch event — capabilities are added and refined continuously based on customer use and evolving model capability.
What Growth Forge optimizes that consulting delivery does manually
The shorthand for what the software adds to the consulting delivery is: cross-project consistency, scaled operation, and reduced methodology-specialist dependency.
Cross-project consistency comes from the shared data model. Without the software, ten projects across a portfolio carry ten slightly different document structures, ten slightly different assumption taxonomies, ten slightly different evaluation rubrics. Differences accumulate quietly across engagements and across reviewer handoffs. With the software, every project works on the same underlying structure, evaluated against the same configurable criteria, with the same Critical Assumption discipline. Cross-project comparison becomes a structural property of the system rather than an exercise in heroic synthesis.
Scaled operation comes from the persistent data model. Without the software, the strategy hypothesis lives as a deck that gets rebuilt for each review cycle, with the team carrying the cumulative state in their working memory and their archived decks. With the software, the hypothesis lives as a persistent object that the next iteration evolves directly, the next gate evaluates directly, and the next portfolio review rolls up directly. The cumulative work across iterations compounds rather than getting periodically translated into and out of slide form.
Reduced methodology-specialist dependency comes from the structured workflow. Without the software, applying the integrated system rigorously across many projects requires methodology-trained practitioners embedded in every team. With the software, the workflow itself supports the dimensional discipline, the assumptions classification, the DFV evaluation, and the portfolio aggregation — letting smaller teams of practitioners cover more projects without losing rigor.
None of these gains replaces what consulting delivers. They complement it: consulting introduces the system, calibrates it, designs the governance, coaches the teams; the software makes the day-to-day operation of the system consistent, scalable, and operable without requiring a methodology specialist on every project.
7. AI as a New Piece of the Puzzle
The puzzle has a recent addition. The integration question, having been answered for the borrowed frameworks of the past several decades, now extends to a new capability set the field is actively absorbing: generative AI — large language models, chatbots, agents, and AI-enabled applications and services. The frameworks of the past several decades did not anticipate this capability set, because the capability set did not exist when the frameworks were developed.
The shape of the question is familiar. Where does the new capability fit in the integrated system? At which discrete points in the methodology, with what cautions, integrated into Growth Forge® or operating alongside it? What does the integration discipline that produced the rest of the integrated system say about how a new piece of the puzzle gets absorbed without breaking the structural integration of the system?
Where AI fits in the integrated system
AI capability accelerates work at every layer of the integrated system, at specific points in the methodology, in ways that preserve the human decisions the system is built to support.
At Layer 1, the strategy formation work, AI assists the drafting of strategy hypothesis content across many dimensions. AI Discovery Assistants in Growth Forge generate starting points for Jobs to Be Done, Requirements, Solution Definition, Value Network build-out, Competitive performance against requirements, and Implementation Approach — and the AI Analysis Assistants in development will provide widget-level and aggregate analysis interpretation across all of those. The strategy hypothesis schema remains the source of truth; the six dimensions remain the structural backbone. AI is the drafting and interpretive accelerant that gets the hypothesis to a reviewable working version faster than a blank canvas would.
At Layer 2, the nested-loop work, AI assists primarily at the gate-decision summary point. The Strategy Evaluation tool today produces LoFi-level evaluation summaries that feed Step 4 decisions, and the in-development Strategy Evaluation AI Summary & Analysis will extend that capability with dimension-by-dimension summarization of choices and assumptions plus SWOT analysis for MidFi and HiFi hypotheses. The scientific-method discipline remains the source of truth; the four-step cycle remains the operational rhythm; the team’s judgment about what evidence carries the most weight remains the act of evaluation. AI accelerates the synthesis step at the gate; it does not replace the experimental design, the evidence interpretation, or the gate decision itself.
At Layer 3, the portfolio governance work, AI integration is a likely future direction rather than a current capability. The combination of portfolio aggregation, cross-project pattern detection, and allocation-scenario simulation is a natural fit for AI assistance, and the structural data model that Growth Forge already maintains is the right anchor for AI work to operate against. AI capability at this layer is evolving rapidly and progressively, and specific Portfolio Management AI integrations are likely candidates for future Growth Forge releases as the underlying capabilities mature.
Inside Growth Forge at the integration points described in the prior section, AI capability is integrated into the structured workflow at user-initiated, transparent, explicit-approval points. Each integration operates inside the persistent, collaborative, evaluable model rather than as a freestanding alternative to it. The integration preserves the structural backbone of the integrated system — the strategy hypothesis schema, the configurable evaluation criteria, the portfolio data model — while accelerating the work the team does on top of that backbone.
The cautions
The integration discipline that produced the rest of the integrated system applies to AI as well. The discipline asks a specific question of every framework or capability: where does it fit in the integrated system, and what are the structural requirements the integration cannot surrender?
Current general-purpose AI tools — Claude Cowork, Microsoft Copilot, custom assistants built on the major model APIs — partially address the persistence, collaboration, and aggregation issues that an isolated AI chat session does not. A practitioner working with a well-configured assistant can persist context across sessions, attach reference material, and accumulate work across iterations. The point worth naming is therefore not that AI tools cannot persist or aggregate; it is that doing the integrated-system work natively in AI still requires the system’s structure to exist somewhere — either as Growth Forge’s data model, or as context and skills recreated inside the AI assistant for each project (and then maintained alongside the work as it evolves). The user effectively has to bring the integrated system with them as the operating substrate, which is non-trivial to set up and non-trivial to maintain consistently across many projects.
Two specific risks come along with AI work that structured-data-model work does not carry. The first is non-determinism. The same prompts can produce different outputs from one run to the next, which makes AI work harder to make reproducible and harder to evaluate against a stable rubric. The second is hallucination — AI generates plausible-but-wrong content, and the practitioner has to maintain active verification discipline against the underlying evidence. Both risks are manageable when AI is integrated into a structured workflow that anchors AI outputs against the data model and surfaces them for human evaluation; both risks compound when AI is operating as the primary substrate without a structured anchor.
These observations are not predictions about AI’s fundamental limits. They are observations about where current AI capability is well-suited and where it benefits from anchoring, relative to the structural requirements the integrated system carries. The capabilities are evolving rapidly and progressively, not as a single launch event; the boundary between AI as accelerant inside the integrated system and AI as something more substantial will continue to shift as the technology matures and as the integration discipline produces new ways of absorbing the shifting capability.
The integration-not-substitution posture
The integrated system’s posture on AI is integration, not substitution. AI is integrated into the integrated system at the points where it adds productivity without breaking the structural integration the system requires. The principle is durable; the specific integration points evolve as AI capability evolves. Readers who arrive at the article a year from now will find the principle unchanged and the specific integration list updated. That continuity is the design.
The reason the principle is durable while the specifics evolve is that the integration discipline is what produces the integration, not the technology itself. As AI capability shifts, the question — where does the new capability fit in the integrated system, and what structural requirements does the integration preserve — will continue to produce useful answers. A capability that today is an accelerant at the drafting layer may, in some plausible future, be a more substantial participant at the methodology layer or even at the governance layer. When that shift happens, the integration discipline absorbs it the same way the discipline absorbed the borrowed frameworks of the past several decades: by treating the new capability as a piece of the puzzle to be designed in, rather than as a replacement for the puzzle.
Closing
The puzzle is the challenge of applying the strategy and innovation field’s myriad frameworks in an integrated fashion. The integrated system is BRI’s ongoing effort to synthesize and optimize them through decades of practitioner work. Consulting delivery introduces the system, calibrates it, and operates it with client teams. Growth Forge® Software is the optimization layer that makes consistent application practical at scale. AI is the recent addition the integrated system is now absorbing, at discrete points, with the integration discipline that produced the rest of the system. The next piece of the puzzle, when it arrives, will be absorbed the same way.
The integrated system will continue to evolve as the practice continues, as Growth Forge develops, and as new capabilities emerge. The work is ongoing; that is the design. We hope the article is useful to readers working on the same puzzle, and we welcome the conversation that the integrated system makes possible.
Cited Works
Bain & Company / Zook, Chris. Profit from the Core: A Return to the Basics of Business Growth. Harvard Business School Press, 2001.
Blank, Steven Gary. The Four Steps to the Epiphany: Successful Strategies for Products That Win. Cafepress, 2005.
Brown, Tim, and IDEO. Change by Design: How Design Thinking Transforms Organizations and Inspires Innovation. HarperBusiness, 2009.
Christensen, Clayton M. The Innovator’s Dilemma: When New Technologies Cause Great Firms to Fail. Harvard Business School Press, 1997.
Christensen, Clayton M., and Stephen P. Kaufman. “Assessing Your Organization’s Capabilities: Resources, Processes, and Priorities.” Harvard Business Review, 2006.
Christensen, Clayton M., and Michael E. Raynor. The Innovator’s Solution: Creating and Sustaining Successful Growth. Harvard Business School Press, 2003.
Christensen, Clayton M., Taddy Hall, Karen Dillon, and David S. Duncan. Competing Against Luck: The Story of Innovation and Customer Choice. HarperBusiness, 2016.
Collins, James C., and Jerry I. Porras. Built to Last: Successful Habits of Visionary Companies. HarperBusiness, 1994.
Cooper, Robert G. Winning at New Products: Creating Value Through Innovation. First edition Addison-Wesley, 1986; subsequent editions through 2017.
Hambrick, Donald C., and James W. Frederickson. “Are You Sure You Have a Strategy?” Academy of Management Executive, 15(4), 2001.
Kim, W. Chan, and Renée Mauborgne. Blue Ocean Strategy: How to Create Uncontested Market Space and Make the Competition Irrelevant. Harvard Business Review Press, 2005.
McGrath, Rita Gunther, and Ian C. MacMillan. “Discovery-Driven Planning.” Harvard Business Review, July–August 1995.
Moore, Geoffrey A. Crossing the Chasm: Marketing and Selling High-Tech Products to Mainstream Customers. HarperBusiness, 1991.
Moore, Geoffrey A. Zone to Win: Organizing to Compete in an Age of Disruption. Diversion Books, 2015.
Nagji, Bansi, and Geoff Tuff. “Managing Your Innovation Portfolio.” Harvard Business Review, May 2012.
Osterwalder, Alexander, and Yves Pigneur. Business Model Generation: A Handbook for Visionaries, Game Changers, and Challengers. John Wiley & Sons, 2010.
Porter, Michael E. Competitive Strategy: Techniques for Analyzing Industries and Competitors. Free Press, 1980.
Ries, Eric. The Lean Startup: How Today’s Entrepreneurs Use Continuous Innovation to Create Radically Successful Businesses. Crown Business, 2011.
Tushman, Michael L., and Charles A. O’Reilly III. Winning Through Innovation: A Practical Guide to Leading Organizational Change and Renewal. Harvard Business School Press, 1997.
Ulwick, Anthony W. What Customers Want: Using Outcome-Driven Innovation to Create Breakthrough Products and Services. McGraw-Hill, 2005 (foundational Outcome-Driven Innovation work originating 1991).
About the Authors
Jim Bodio and Stephen Rich are co-founders of BRI Associates and creators of Growth Forge® Software. Both spent decades inside Intel’s New Business Incubator (NBI) — an early pioneer of VC-style investment discipline applied to internal innovation — where they engaged directly with Burgelman (Stanford) and Christensen (Harvard) on the foundational work that the integrated system builds on. BRI Associates is a boutique consulting practice serving Heads of Innovation, Strategy and Corporate Development leaders, and founder-CEOs of mature enterprises pursuing new product and business growth. Practitioners, not pundits or academics.
Notes
- For BRI’s thought-leadership on the cumulative-impact pattern, see the Scaling Transformation series (Bodio and Rich, 2024): Part 1 at bri-associates.com/post/scaling-transformation-part-1, Part 2 at bri-associates.com/post/scaling-transformation-part-2, Part 3 at bri-associates.com/post/scaling-transformation-part-3, Part 4 at bri-associates.com/post/scaling-transformation-part-4. The When Digital Transformation Collides with Operational Norms series (2018): Part 1 at bri-associates.com/post/digital-transformation-part-1-of-3, Part 2 and Part 3 at the corresponding bri-associates.com/post/digital-transformation-part-2-of-3 and bri-associates.com/post/digital-transformation-part-3-of-3 URLs. Is Your Innovation Project a “Zombie”? (2018, republished 2024) at bri-associates.com/post/is-your-innovation-project-a-zombie. Operational depth on the RPP framework is treated at bri-associates.com/supporting/rpp. ↩
