When Airtable’s CEO Howie Liu announced in June 2024 that his company was “refounding” itself around AI, he chose his words carefully. This wasn’t a pivot—correcting strategic mistakes or exploring adjacent markets. This was something more significant: a comprehensive reimagining of the company’s value proposition, technical architecture, and organisational structure in response to a paradigm shift that threatened to render traditional business models obsolete.
Within months, Handshake, Opendoor, and MoneyGram announced similar transformations. A pattern emerged. Established startups with significant traction faced a choice: add AI features incrementally and risk becoming “legacy companies with AI bolted on,” or refound themselves as AI-native organisations capable of competing with purpose-built AI startups.
This comprehensive guide explores the refounding phenomenon—what it means, why it’s happening now, and how technical leaders can evaluate whether comprehensive transformation or incremental adoption makes strategic sense for their organisations. We’ve organised this resource around seven deep-dive articles that address distinct aspects of the refounding journey, from definitional clarity through strategic decision-making to technical implementation and organisational change management.
Navigation: Your Refounding Resource Library
Understanding the Fundamentals:
- What Startup Refounding Actually Means and Why It Is Not Just Another Pivot — Definitional clarity using Yale research on institutional drift, historical context, and frameworks distinguishing refounding from pivots and incremental change
- Why AI-Native Startups Win Against Legacy Companies and What It Means Now — Competitive dynamics analysis showing structural advantages of AI-native companies, data flywheel economics, and urgency for transformation
Strategic Decision-Making:
- How to Decide Whether Your Company Should Refound or Add AI Features Incrementally — Comprehensive decision frameworks from Bain and Yale, board communication strategies, and evaluation criteria for technical leaders
Economic and Technical Foundations:
- Outcomes-Based Pricing and AI-First SaaS Gross Margin Economics Explained — Business model economics, pricing strategy evolution, margin compression analysis, and improvement playbooks with specific examples
- Agentic AI Architecture and the Semantic Gap Challenge in Data-Centric Systems — Technical deep-dive on AI architecture requirements, data-centric system design, semantic gap challenges, and infrastructure optimisation strategies
Implementation and Execution:
- Managing Organisational Transformation During Startup Refounding and Cultural Change — Practical frameworks for team restructuring, cultural transformation, communication strategies, and change leadership during refounding
- Startup Refounding Case Studies from Airtable Handshake Opendoor and MoneyGram — Detailed examination of four companies’ refounding approaches with specific metrics, comparative analysis, and lessons learned
What Is Startup Refounding and How Does It Differ from Pivoting?
Refounding represents comprehensive business model transformations where mature startups fundamentally reimagine their value propositions and business foundations in response to technological disruption, particularly AI integration. Unlike pivoting, which corrects strategic mistakes or explores adjacent markets, refounding retains the core mission while transforming how it’s achieved. The term emerged in 2024 when companies like Airtable and Handshake publicly announced existential transformations to compete with AI-native startups.
The distinction matters. Pivots acknowledge strategic errors—incorrect market assumptions, product-market fit, or go-to-market approach. Refounding acknowledges that external forces—AI’s emergence as a transformative technology—have altered what “right” looks like. As Howie Liu explained when announcing Airtable’s transformation, they used “the language of founding because the stakes feel the same.” The company hadn’t failed; the market had shifted.
Yale research on institutional drift provides the academic framework underpinning this distinction. Over time, organisations accumulate decisions that distance them from foundational identity—not through failure, but through incremental choices optimised for yesterday’s environment. Refounding addresses drift that occurs despite success, requiring a comprehensive reassessment of a company’s goals, culture, and operational frameworks.
This is why companies choosing to refound face different challenges than those pivoting. Pivots involve changing direction. Refounding involves transforming execution while maintaining strategic continuity. The former addresses internal mistakes; the latter responds to paradigm shifts that render existing business models potentially obsolete.
For a comprehensive exploration of definitional boundaries and what distinguishes refounding from related concepts, read What Startup Refounding Actually Means and Why It Is Not Just Another Pivot.
Why Are Companies Choosing to Refound Instead of Adding AI Features Incrementally?
Companies refound because AI-native startups demonstrate advantages that incremental AI feature additions cannot match. Agentic AI enables entirely new business models—particularly outcomes-based pricing where customers pay for results rather than software access. Incremental approaches leave technical debt intact, fail to realign organisational culture around AI-first thinking, and position companies as “legacy with AI features” rather than AI-native competitors. Refounding signals to investors, customers, and employees that the transformation is existential, not cosmetic.
The competitive urgency stems from structural advantages AI-native startups enjoy. They build data flywheels from inception—self-reinforcing cycles where user interactions generate data improving AI performance, attracting more users generating more data. Legacy companies must re-engineer entire operations to achieve similar flywheel effects, a transformation requiring comprehensive architectural changes rather than incremental feature additions.
Business model implications drive refounding decisions. Agentic AI—autonomous systems completing complex jobs independently—enables outcomes-based pricing that differs from traditional subscription or seat-based models. Sierra, Bret Taylor’s customer service AI company, exemplifies this: customers pay pre-negotiated rates when AI resolves issues autonomously, nothing if human escalation is necessary. This pricing model requires architecture designed around autonomous outcome delivery, not tools requiring human operation.
Organisational alignment drives cultural transformation. Refounding creates commitment that incremental initiatives often lack—signaling to investors, customers, and employees that transformation is existential rather than optional enhancement.
Explore the detailed decision frameworks and evaluation criteria in How to Decide Whether Your Company Should Refound or Add AI Features Incrementally. For competitive context, see Why AI-Native Startups Win Against Legacy Companies and What It Means Now.
What Does Agentic AI Mean and Why Does It Drive Refounding Decisions?
Agentic AI refers to autonomous systems that perceive environments, make decisions, act independently toward goals, and adapt strategies based on new information—going beyond generative AI or simple automation to proactively coordinate complex workflows. This capability enables new business models where software delivers outcomes rather than providing tools for humans to achieve outcomes. Companies refound when they recognise that agentic AI isn’t just a feature enhancement but a transformation requiring new architecture, data infrastructure, and commercial models to capture its full value.
At its core, agentic AI exhibits four defining characteristics: autonomy (operates without constant human intervention), proactivity (initiates actions rather than waiting for prompts), goal-orientation (works toward defined objectives), and adaptability (learns and adjusts strategies based on feedback). These characteristics differentiate agentic systems from generative AI tools like ChatGPT, which respond to prompts but don’t independently pursue complex objectives across multi-step workflows.
The business model implications are significant. Traditional software provides tools—features users operate to achieve outcomes. Agentic AI delivers outcomes directly, autonomously coordinating the complex workflows previously requiring human expertise. This shift enables outcomes-based pricing models where customers pay for results delivered rather than software capabilities provided. The value proposition changes: from “here’s a powerful tool” to “here’s the result you wanted.”
Why this triggers refounding becomes clear when examining technical requirements. Agentic AI demands data-centric architectures where data quality, availability, and structure are primary architectural concerns. Legacy systems optimised for human-driven workflows rarely have the data infrastructure, integration patterns, or real-time capabilities agentic AI requires. Comprehensive re-architecting becomes necessary—the scale of transformation that merits refounding terminology.
Industry recognition confirms agentic AI as the primary refounding catalyst. Bain research identifies agentic systems as the fourth SaaS disruption scenario—the most comprehensive transformation tier requiring business model changes. Sequoia analysis through Bret Taylor positions agentic AI in the “applied AI/agents” market tier as the most attractive competitive positioning for enterprise value creation.
For detailed technical exploration of agentic AI architecture, data-centric system requirements, and implementation challenges, read Agentic AI Architecture and the Semantic Gap Challenge in Data-Centric Systems.
How Does Refounding Change a Company’s Business Model and Pricing Strategy?
Refounding typically shifts companies from access-based revenue models (subscriptions, seats) to outcomes-based pricing where customers pay for results achieved rather than software capabilities provided. This transformation requires re-architecting products around agentic AI that can autonomously deliver outcomes, changing gross margin economics, customer success metrics, and go-to-market strategies. Companies also often shift from product-centric to network-centric models, as exemplified by MoneyGram’s transformation from remittances business to payments network.
The pricing model evolution follows a predictable path: seat-based → subscription → usage-based → outcomes-based. Each transition reflects changing value delivery mechanisms. Outcomes-based pricing represents the most significant shift because it aligns payment directly with value realisation rather than access or consumption. Customers pay for jobs completed, problems solved, or outcomes achieved—requiring AI systems capable of autonomous delivery without human operation.
Gross margin implications demand careful analysis. AI-first products exhibit different unit economics than traditional SaaS. Traditional SaaS companies achieve 80-90% gross margins because software distribution costs approach zero. AI-first companies typically see 50-65% margins due to inference costs, model development expenses, and infrastructure requirements. GitHub Copilot famously lost $20 per user monthly while charging $10 subscriptions—canonical margin compression requiring pricing model innovation.
Margin improvement becomes a strategic priority. Companies employ multiple strategies including infrastructure optimisation, intelligent routing, and scale efficiencies—detailed in the economics cluster article.
Network effects represent alternative strategic positioning. Some refounding companies pivot from individual product value to network or platform value. MoneyGram exemplifies this shift from transaction business to network business, changing competitive moats and value capture mechanisms.
For comprehensive economic analysis including specific margin ranges, pricing model comparisons, and improvement strategies, explore Outcomes-Based Pricing and AI-First SaaS Gross Margin Economics Explained.
What Are the Main Organisational and Cultural Changes During Refounding?
Refounding demands organisational transformation back to startup-like intensity: return to office mandates, longer hours, faster decision cycles, and cultural reset around urgency. Companies often restructure teams using patterns like Airtable’s approach where some teams ship AI features weekly while others make long-term infrastructure investments. Workforce reductions (typically 10-15%) accompany refounding, reallocating resources toward AI priorities. Leadership behaviour changes dramatically—founders and CEOs often return to hands-on building to signal transformation depth.
Cultural intensity represents a highly visible transformation dimension. Handshake implemented mandatory five-day office weeks with explicit expectations for employees to operate “with a pace and number of hours that is meaningful and will help us hit goals”. This cultural shift from mature startup flexibility to founding-era intensity creates significant employee impact, yet Handshake’s leadership frames it as necessary to compete with AI-native startups built around this intensity from inception.
Team restructuring patterns vary by company but share common themes. Airtable implemented dual-speed structures: some teams optimise for rapid AI feature delivery (weekly releases), others for foundational architecture investments (multi-month projects). This organisational design acknowledges that refounding requires both immediate market presence and long-term technical transformation—different time horizons demanding different team structures and success metrics.
Workforce realignment accompanies most refounding transformations. Handshake reduced headcount by 15% (approximately 100 employees from 650 U.S. staff) while simultaneously growing its AI division from 15 to 150 employees. This pattern—strategic reductions in legacy operations funding investments in AI-native capabilities—characterises refounding workforce changes. The reductions aren’t cost-cutting; they’re resource reallocation toward transformation priorities.
Leadership engagement signals transformation seriousness. Howie Liu positioned himself as “IC CEO”—individual contributor CEO—returning to hands-on technical work rather than purely strategic oversight. This leadership behaviour change signals to organisations that refounding represents existential transformation requiring founder-level engagement, not delegated initiative management.
For practical frameworks on team restructuring, cultural transformation, communication strategies, and change management during refounding, read Managing Organisational Transformation During Startup Refounding and Cultural Change.
Which Companies Have Successfully Refounded and What Can We Learn from Them?
Airtable, Handshake, MoneyGram, and Opendoor represent prominent refounding examples with distinct approaches. Airtable pioneered the term and implemented comprehensive product transformation. Handshake emphasised cultural intensity with workforce reduction and built a $100M ARR AI division in 8 months. MoneyGram transformed from remittances to payments network after going private. Opendoor brought in new leadership to drive AI-era transformation. Each demonstrates different refounding patterns: product-centric, organisational, business model-centric, and leadership-driven approaches.
Airtable’s approach established the refounding playbook. In June 2024, the company announced comprehensive transformation from no-code collaboration tool to “AI-native app platform,” adding AI assistant “Omni” as standard feature and revamping product structure and pricing. The transformation goes beyond feature additions to product repositioning—changing how customers perceive and engage with the platform.
Handshake’s intensity demonstrates the most dramatic cultural shift. The company implemented 15% workforce reduction (100 of 650 employees) while building its AI division from 15 to 150 employees, achieving $100M ARR in just 8 months. Current combined ARR reaches $200M with projected year-end combined ARR of $300M. The AI business is expected to surpass core recruiting operations by year-end. Handshake leveraged its network of 500,000 PhDs and 3 million Master’s degree holders for post-training data labelling—a strategic asset enabling rapid AI development.
MoneyGram’s model represents business model refounding in a traditional industry. Since going private in 2023, the company evolved from remittances-focused player to fintech built around global payments network. Cross-border volume increased approximately 8%, digital transaction share rose by over 20 percentage points, with digital now representing one-third of total volume. CEO Anthony Soohoo’s framing captures the strategic shift: “Remittances was the old way to think about the business, now the network is our business”—platform thinking replacing transaction business thinking.
Opendoor’s leadership demonstrates the CEO transition approach to refounding. The real estate tech company appointed new CEO Kaz Necatian and repositioned as “software and AI company” rather than real estate technology provider. This leadership-driven transformation addresses the question of whether external CEOs with AI expertise or existing founders should lead refounding—Opendoor chose fresh AI-era leadership.
For detailed examination of each company’s approach, specific metrics, comparative analysis, and lessons learned, explore Startup Refounding Case Studies from Airtable Handshake Opendoor and MoneyGram.
When Should a Company Consider Refounding Versus Incremental AI Adoption?
Consider refounding when: (1) AI-native competitors demonstrate advantages you cannot match incrementally, (2) agentic AI enables entirely new business models that obsolete your current approach, (3) technical debt prevents meaningful AI integration without comprehensive re-architecting, (4) you have the financial runway and investor support for multi-year transformation, and (5) leadership commits to organisational intensity required for success. Consider incremental adoption when competitive urgency is lower, legacy systems remain viable, or transformation risks outweigh competitive threats.
Competitive pressure indicators signal when refounding becomes strategically necessary. When AI-native startups capture significant market share, customers demand outcomes-based pricing, or investors question long-term viability of current models, incremental approaches risk positioning your company as legacy player regardless of AI feature additions. The question becomes whether you can compete effectively as “traditional company with AI features” or must become “AI-native company with established market presence.”
Technical feasibility assessment determines transformation scope. Can existing architecture support agentic AI, or does technical debt require rebuilding? What scale of data infrastructure changes are necessary? If the answer involves comprehensive re-architecting—data-centric architecture, real-time capabilities, integration patterns for autonomous workflows—refounding-scale transformation becomes necessary rather than incremental enhancement.
Financial readiness establishes feasibility constraints. Refounding requires 2-3 year runway minimum, potential short-term revenue disruption, and significant investment in AI capabilities. Do you have capital and investor alignment? The transformation timeline and investment requirements exceed typical feature development, requiring board-level strategic commitment rather than product roadmap decisions.
Organisational capacity influences execution outcomes. The human dimension critically influences refounding outcomes alongside technical and financial factors—cultural transformation requires consistent leadership commitment and organisational resilience.
Market timing influences competitive positioning. Are you early enough to establish AI-native positioning, or too late such that refounding won’t close competitive gaps? Handshake CEO Garrett Lord framed this urgency: “Winners and losers are being defined right now.” The window for refounding closes as AI-native competitors establish market presence and customer relationships.
For comprehensive evaluation criteria, risk assessment frameworks, and strategic decision-making processes, read How to Decide Whether Your Company Should Refound or Add AI Features Incrementally.
What Are the Primary Technical Challenges in Refounding for AI-First Architecture?
The semantic gap—disconnect between user intent and machine execution—represents a technical challenge requiring attention. AI-first architectures must bridge the gap between what users want to accomplish (high-level goals) and what data and systems can actually deliver (technical capabilities). This requires data-centric system design where data quality, availability, and structure are primary architectural concerns. Legacy systems optimised for human-driven workflows rarely have the data infrastructure, integration patterns, or real-time capabilities agentic AI demands.
The semantic gap causes AI systems to misunderstand instructions and context. Traditional approaches using structured languages or rule-based systems fail to address this challenge adequately. Agentic AI must understand intent despite natural language ambiguity and business context complexity.
Data-centric architecture shift represents a re-engineering challenge. Traditional software prioritises code and features; AI-first systems prioritise data quality, accessibility, real-time availability, and continuous learning loops. Enterprise knowledge scattered across incompatible systems—CRM platforms, ERP systems, document repositories—prevents comprehensive AI reasoning. Different departments using incompatible terminology for identical concepts creates additional complexity: when sales systems refer to ‘customers’ and finance systems call them ‘clients,’ AI must recognise these as the same entity rather than treating them as distinct.
Integration complexity compounds architectural challenges. Agentic AI must coordinate across multiple systems, data sources, and workflows—requiring robust integration architecture and API strategies. The systems must support real-time and batch data processing, structured, semi-structured, and unstructured data, automated machine learning pipelines, and enterprise-grade governance and security. Most legacy architectures weren’t designed for these requirements.
Technical debt migration often means comprehensive rebuilding rather than incremental enhancement. Legacy code, data models, and architectural patterns directly conflict with AI-native requirements. The scale of required changes—data-centric architecture, semantic layer implementation, integration patterns for autonomous workflows—explains why companies choose refounding terminology rather than framing transformation as incremental improvement.
For detailed technical exploration including architecture patterns, semantic gap mitigation strategies, and implementation approaches, explore Agentic AI Architecture and the Semantic Gap Challenge in Data-Centric Systems.
How Do Companies Communicate Refounding to Investors and Customers?
Refounding announcements typically frame transformation as proactive response to market opportunity rather than defensive reaction to competition. Companies emphasise AI-enabled business model innovation, new market adjacencies, and growth potential rather than risk mitigation. To investors, refounding signals renewed growth trajectory and commitment to market leadership. To customers, it demonstrates innovation commitment and future-proofing. Public announcements generate media attention and market repositioning.
Investor messaging frames refounding as offensive strategy—capturing AI-era opportunities—rather than defensive posture avoiding obsolescence. Companies emphasise total addressable market expansion enabled by new business models, improved unit economics as AI systems achieve scale, and competitive positioning as AI-native player rather than legacy incumbent. Board communications typically include multi-year transformation roadmaps with clear milestones, investment requirements and expected returns, competitive analysis showing urgency, and risk mitigation strategies addressing execution challenges.
Customer communication requires balancing reassurance with excitement. Product-focused refounding announcements (like Airtable’s) emphasise backward compatibility and gradual enhancement—existing customers continue using familiar tools while gaining AI-native capabilities over time. Business model refounding (outcomes-based pricing) requires more significant customer education—explaining how new pricing aligns payment with value realisation and potentially reduces customer costs by eliminating underutilised seats or features.
Public positioning uses “refounding” terminology deliberately to signal transformation magnitude. The term generates media coverage (TechCrunch, The New York Times, and industry publications covered the trend extensively in 2024-2025), differentiates from routine product updates or feature additions, and positions company as AI-era innovator rather than legacy player adding AI features. Airtable’s June 2024 announcement exemplifies this approach—the company could have announced “new AI features” but chose “refounding” to communicate existential transformation.
Employee alignment represents the most critical communication audience. Internal messaging must be authentic about intensity expectations—longer hours, faster pace, higher performance standards—while inspiring conviction in mission and opportunity. Handshake CEO Garrett Lord’s internal communication exemplified this balance: acknowledging the demanding nature of refounding while framing it as necessary to compete with AI-native startups defining the next decade of their industry.
What Timeline Should Technical Leaders Expect for Refounding Transformation?
Refounding transformations typically require 2-3 years minimum for meaningful results, with initial organisational changes in months 0-6, technical foundation building in months 6-18, and new product or business model rollouts in months 12-24. Handshake demonstrated unusually rapid execution with a $100M ARR AI division in 8 months, but comprehensive business model transformation requires longer horizons. Technical leaders should plan for multi-year commitment with clear phase gates rather than expecting quick wins.
Phase 1 (Months 0-6): Organisational Foundation involves restructuring teams for AI-first development, implementing cultural intensity shifts (office mandates, pace expectations), strategic workforce reductions and AI talent acquisitions, and leadership behaviour changes signaling transformation commitment. This phase establishes the organisational capacity for transformation—companies that skip or rush this foundation often struggle with execution in later phases.
Phase 2 (Months 6-18): Technical Transformation focuses on data-centric architecture implementation, AI capability development (model selection, training, integration), technical debt migration from legacy systems, and initial product experimentation with AI-native features. This represents the most technically complex phase, where architectural decisions have long-term implications. Companies often underestimate this phase’s timeline, leading to compressed schedules and quality compromises.
Phase 3 (Months 12-24): Market Rollout includes launching AI-native product capabilities, transitioning to new business models (outcomes-based pricing), customer migration from legacy to AI-native offerings, and market positioning as AI-first company. This phase overlaps with Phase 2—some teams ship customer-facing AI features while others continue foundational work. The dual-speed team structure Airtable implemented addresses this reality.
Phase 4 (Months 18-36): Scale and Refinement involves scaling AI-native products based on market feedback, refining business models and unit economics, demonstrating sustainable competitive positioning, and achieving financial targets justifying transformation investment. Success metrics shift from technical milestones to business outcomes—revenue from AI-native products, customer retention and satisfaction, competitive win rates, and gross margin improvements.
Accelerated execution is possible but rare. Handshake’s $100M ARR in 8 months represents exceptional execution enabled by unique advantages: access to 500,000 PhDs and 3 million Master’s degree holders for data labelling, existing customer relationships for rapid AI product distribution, strong financial position allowing aggressive AI investment, and leadership willingness to make difficult cultural and workforce decisions quickly. Most companies face longer timelines with more typical constraints.
Timeline risks include underestimating technical complexity leading to delays, employee attrition disrupting continuity, competitor advances while refounding is underway, and investor impatience if short-term metrics decline. Managing these risks requires explicit contingency planning, selective retention strategies for critical talent, clear communication of multi-year transformation expectations, and milestone-based progress tracking showing forward momentum even when financial metrics lag.
What Risks Should Technical Leaders Consider Before Refounding?
Key risks include: (1) employee attrition from cultural intensity and workforce reductions, (2) customer disruption if products change dramatically or pricing models shift, (3) execution failure if technical complexity exceeds organisational capability, (4) competitive timing risk if AI-native startups move faster than refounding progress, and (5) investor pressure if short-term metrics decline during transformation. Successful refounding requires explicit risk mitigation strategies, clear success metrics, and realistic timelines (typically 2-3 years minimum).
Talent retention represents an immediate risk. Refounding intensity—office mandates, longer hours, startup pace—causes attrition. Plan for 15-25% turnover and ensure retention of AI-critical talent through selective compensation adjustments and ensuring cultural intensity doesn’t disproportionately affect key contributors. Handshake CEO Garrett Lord acknowledged the human cost during refounding.
Customer impact varies by refounding scope. Product-focused refounding (like Airtable) typically maintains backward compatibility while adding AI-native capabilities, allowing gradual customer migration. Business model refounding (outcomes-based pricing) requires contract renegotiations and customer success strategy changes. Companies usually grandfather existing customers under legacy terms while offering new models to new customers, then gradually migrating the customer base.
Technical execution risk stems from agentic AI and data-centric architecture complexity. Underestimating technical challenges leads to delays and cost overruns. The semantic gap challenge, data infrastructure requirements, and integration complexity frequently exceed initial estimates. Organisations must honestly assess whether engineering capability matches transformation ambition, or if execution risk outweighs competitive threat.
Competitive dynamics create timing risk. Refounding takes years; AI-native startups continue advancing. The question becomes whether you can close competitive gaps or if they’ll maintain insurmountable leads. Handshake’s rapid execution—$100M ARR in 8 months—represents unusually successful execution; most transformations require longer timeframes with uncertain outcomes.
Financial pressure during transformation tests organisational resilience. Short-term metrics often decline as companies invest in transformation while maintaining legacy operations. Gross margins compress as AI inference costs increase before scale efficiencies emerge. Revenue growth may slow as organisations focus on product transformation rather than sales execution. Managing investor expectations around multi-year transformation timelines becomes a leadership challenge.
📚 Startup Refounding Resource Library
Understanding Refounding Fundamentals
What Startup Refounding Actually Means and Why It Is Not Just Another Pivot
Definitional clarity, historical context, and frameworks distinguishing refounding from pivots, transformations, and strategic shifts. Uses Yale research on institutional drift for academic legitimacy and provides clear decision criteria for which terminology applies to your situation.
Why AI-Native Startups Win Against Legacy Companies and What It Means Now
Competitive dynamics analysis showing structural advantages of AI-native companies—data flywheel economics, organisational agility metrics, and historical precedents (Salesforce vs Siebel). Addresses “are we too late?” concerns with evidence-based perspective on transformation timing.
Making the Refounding Decision
How to Decide Whether Your Company Should Refound or Add AI Features Incrementally
Comprehensive decision framework with evaluation criteria from Bain and Yale, risk assessment tools, and strategic considerations. Use this article to build your business case for or against refounding, including board communication strategies and Bret Taylor’s three AI markets framework for competitive positioning.
Technical and Economic Implementation
Agentic AI Architecture and the Semantic Gap Challenge in Data-Centric Systems
Technical deep-dive on AI architecture requirements, semantic gap challenges (the disconnect between user intent and machine execution), data-centric system design patterns, and infrastructure optimisation strategies. Balances technical depth with strategic context, explaining why vertical specialisation addresses reliability challenges.
Outcomes-Based Pricing and AI-First SaaS Gross Margin Economics Explained
Economic analysis of new business models, pricing strategies (subscription vs usage-based vs outcomes-based), and unit economics for AI-first products. Includes specific margin ranges (50-65% vs 80-90% traditional SaaS), GitHub Copilot’s margin compression example, and margin improvement playbooks.
Organisational Execution
Managing Organisational Transformation During Startup Refounding and Cultural Change
Practical frameworks for team restructuring, cultural transformation, workforce management, and change leadership. Addresses communication strategies for announcing refounding, severance and equity handling during workforce reductions, and shifting from specialist hierarchies to generalist agility. Uses Handshake as detailed organisational case study.
Real-World Case Studies
Startup Refounding Case Studies from Airtable Handshake Opendoor and MoneyGram
Detailed examination of four companies’ refounding approaches with specific metrics, comparative analysis, and lessons learned. Includes Handshake’s $100M ARR in 8 months, MoneyGram’s business model transformation, Airtable’s product repositioning, and Opendoor’s leadership-driven approach. Serves as evidence hub for board presentations and strategic planning.
Frequently Asked Questions About Startup Refounding
Is refounding just a Silicon Valley buzzword or a meaningful strategic framework?
Refounding represents a strategic response to AI-driven market transformation, not marketing terminology. The term gained legitimacy when multiple established startups (Airtable, Handshake, Opendoor, MoneyGram) independently announced comprehensive transformations in 2024-2025, each emphasising AI-driven business model changes rather than incremental product updates. While “refounding” is newer terminology, the underlying phenomenon—business model transformation in response to technological paradigm shifts—has historical precedent in previous platform transitions.
Learn more: What Startup Refounding Actually Means and Why It Is Not Just Another Pivot
Can bootstrapped or private-equity-backed companies refound, or is it only for VC-backed startups?
Refounding is feasible across funding structures, though execution differs. VC-backed companies often refound to justify continued growth narratives and prevent valuation stagnation. Private-equity-backed companies may refound as value-creation strategy during ownership period—MoneyGram exemplifies this after Vista Equity acquisition. Bootstrapped companies have freedom to refound without investor pressure but face capital constraints for AI investments. The decision criteria (competitive threat, business model obsolescence, AI opportunity) apply regardless of funding structure.
What happens to existing customers during a refounding transformation?
Customer impact varies by refounding scope. Product-focused refounding (like Airtable) typically maintains backward compatibility while adding AI-native capabilities, allowing gradual customer migration. Business model refounding (outcomes-based pricing) requires contract renegotiations and customer success strategy changes. Companies usually grandfather existing customers under legacy terms while offering new models to new customers, then gradually migrating the base. Customer communication emphasising enhanced value delivery and future-proofing is critical to retention during transformation.
How do you measure whether refounding is succeeding or failing?
Success metrics evolve across refounding phases. Early indicators (Months 0-12): talent retention of AI-critical roles, technical milestone achievement (architecture rebuilds, AI capability launches), organisational culture shifts measured through engagement surveys and behaviour changes. Mid-term metrics (Months 12-24): new product adoption rates, customer willingness to adopt new pricing models, AI feature usage statistics, product velocity improvements. Long-term success (Months 24-36): revenue from AI-native products as percentage of total revenue, competitive positioning vs AI-native startups, gross margin trends, market share changes. Refounding is multi-year transformation. Impatience with short-term metrics often leads to premature abandonment.
Learn more: How to Decide Whether Your Company Should Refound or Add AI Features Incrementally
What percentage of employees typically leave during refounding transformations?
Workforce dynamics vary but patterns emerge from case studies. Handshake implemented 15% strategic workforce reduction as part of refounding announcement. Additional voluntary attrition occurs due to cultural intensity shifts (office mandates, longer hours, startup pace)—expect 10-20% voluntary turnover in first year beyond planned reductions. Critical variable is retention of AI-critical talent—if senior engineers, AI specialists, and technical leaders depart, refounding risks failure. Successful refounding requires selective retention strategies, often including compensation adjustments for critical roles and ensuring cultural intensity doesn’t disproportionately affect key contributors.
Learn more: Managing Organisational Transformation During Startup Refounding and Cultural Change
Do companies need to bring in new leadership to refound successfully?
Leadership approaches vary. Some companies bring in external CEOs with AI expertise (Opendoor with Kaz Necatian) to drive transformation. Others rely on existing founders returning to hands-on building (“IC CEO” model like Airtable’s Howie Liu). Success depends less on leadership source and more on: (1) CEO technical conviction and AI understanding, (2) willingness to embrace organisational intensity, (3) board alignment on transformation necessity, and (4) ability to attract and retain AI talent. External CEOs bring fresh perspective and AI experience; founder CEOs bring organisational credibility and mission continuity. Both models work with right capabilities and commitment.
How does refounding affect company valuation in the short and long term?
Valuation impact is complex and timing-dependent. Short-term (first 12-18 months): valuations often stagnate or decline as refounding depresses current metrics (revenue slowdowns, margin compression from AI investments, customer churn from product or pricing changes). Investors may mark down valuations during transformation execution risk periods. Long-term (24-36 months): successful refounding can dramatically increase valuations by expanding total addressable market, improving unit economics, and positioning companies as AI leaders rather than legacy players. Failed refounding leads to sustained valuation declines and potential acqui-hires. Companies typically communicate refounding explicitly to manage investor expectations around multi-year transformation timelines rather than quarterly performance.
Can companies refound in phases or must it be “big bang” transformation?
Phased refounding is possible but requires careful sequencing. Common pattern: Phase 1 (Organisational)—restructure teams and establish cultural intensity while maintaining existing products. Phase 2 (Technical)—build AI-native architecture and capabilities parallel to legacy systems. Phase 3 (Commercial)—introduce new pricing models and business models for new customers while supporting legacy customers. Advantage of phased approach: reduces execution risk and customer disruption. Disadvantage: prolongs time in hybrid state where company has neither legacy efficiency nor AI-native advantages. Most successful refounding involves rapid organisational transformation (6-12 months) followed by more deliberate technical and commercial evolution (18-24 months).
Conclusion: Navigating Your Refounding Decision
The refounding trend reflects a significant market transition. AI doesn’t just enable new features; it enables new business models where software delivers outcomes rather than tools. Companies face a choice: comprehensive transformation positioning them as AI-native competitors, or incremental enhancement risking “legacy with AI bolted on” positioning.
The decision criteria are straightforward even if execution is complex. Refound when AI-native competitors demonstrate advantages you cannot match incrementally, when agentic AI enables business models that obsolete your current approach, when technical debt prevents meaningful integration without re-architecting, and when you have financial runway and leadership commitment for multi-year transformation.
The seven articles in this resource library provide the frameworks, economic analysis, technical guidance, organisational playbooks, and case study evidence to navigate this decision thoughtfully. Start with definitional clarity and competitive context, proceed through strategic decision-making, then dive into technical and organisational implementation details as needed.
Handshake CEO Garrett Lord’s framing captures the urgency: “Winners and losers are being defined right now.” Whether that urgency justifies refounding-scale transformation depends on your competitive position, technical capabilities, financial resources, and organisational capacity. This guide provides the analytical tools to make that determination rigorously rather than reactively.
