Bain's $100B SaaS Opportunity in Cross-System Labor

Key takeaways

1. Date / source. May 2026, bain.com/insights, brief part 2/5 of the "software industry in the age of AI" series.
2. Authors. David Crawford, Chris McLaughlin, Greg Fiore (Bain SaaS partners).
3. Pivot thesis. agentic AI's primary market opportunity is not replacing existing SaaS but automating cross-system coordination labor. ### The market sizing | Geography | Market size | |------------|---------------| | US | ~$100B | | US + Canada + Europe + AU/NZ | ~$200B | | Current capture | $4-6B (10%) | | Still up for grabs | >$90B | Distribution by function (US):
4. Sales: $20B
5. COGS / operations: $26B
6. R&D / engineering: $6-12B
7. Support: $6-12B
8. Finance: $6-12B ### The six automation factors 1. Output verifiability — can the quality of the result be easily verified? 2. Consequence of failure — how severe is an error? 3. Digitized knowledge availability — is the necessary knowledge digitized? 4. Integration complexity — how many systems must be connected? 5. Process variability — is the process standardized or highly variable? 6. Physical world dependency — does it depend on actions in the physical world? → The more verifiable the output + the lower the consequence of error + the more digitized the knowledge + the simpler the integration + the lower the variability + the less physical-world dependency, the higher the automation potential. ### Automation potential by function | Function | % automatable | |----------|-----------------| | Customer support | 40-60% | | R&D | 40-60% | | Finance | 35-45% | | HR | 35-45% | | Sales | 30-40% | | IT | 30-40% | | Legal | 20-30% | ### The strategic shift — the new moat Before: system of record ownership — Salesforce/SAP/Workday own the data, and that is what gives them a moat. Now: cross-workflow decision context — the cross-cutting ability to see and act across multiple integrated systems. Durable moat: "accumulated execution data that grows more valuable over time and becomes harder for competitors to replicate" — every agent execution enriches the case base, which becomes more valuable for the next execution. Classic flywheel effect but on a new substrate. ### Four illustrative examples | Player | Position | |--------|----------| | Sierra | Autonomous customer issue resolution (cross-system) | | Glean | Cross-function employee request coordination | | GitHub Copilot | Extended beyond source control (multi-system dev workflows) | | Cursor | ARR doubled in a quarter, reaching $2B | ### Three-phase strategic playbook | Phase | Activities | |-------|-----------| | 1. Assessment | Identify high-value automatable workflows via the 6 factors; size market opportunity | | 2. Strategic Positioning | Assess data assets; identify adjacent workflow opportunities; map actual operational workflows (not theoretical processes) | | 3. Execution | Close capability gaps (build / buy / partner); restructure organization and incentives; redesign data foundations for agent readiness | ### Dossier connections #### Convergence on "agent readiness" / "data foundations"
9. Bain. redesign data foundations for agent readiness.
10. DORA ROI 2026. (2026-04-21): AI-accessible internal data + healthy data ecosystems + machine-readable documentation quality.
11. Foundation Capital — Context Graphs trillion-dollar opportunity. (2025-12-22): decision traces, new systems of record.
12. Habert PROJ-AI. (2026-05-05): DOCS / IDEAS / DR / OUT / DOCTRINE / AGENT — six zones, doctrine.
13. Seale Semantic Agent. (2026-04-17): (Model+Harness) + (Ontology+Data) — ontology as the only moat.
14. Talisman Ontology Pipeline Refresh. (2026-05-04): governance + AI partnership in the ontology pipeline.
15. → Strong convergence: preparing data for agents is the strategic 2026 project, regardless of the model. #### Convergence on "moat = execution data + cross-workflow context"
16. Bain. accumulated execution data + cross-workflow decision context.
17. Foundation Capital. Context Graphs as the new system of record.
18. Curran/Intercom. (2026-04-16): Skills-Based Plugin Architecture, 153 contributors, 267 skills — accumulation of execution patterns.
19. Stripe Minions. (2026-02-19): Toolshed ~500 MCP tools, blueprints, devboxes, 1300+ PRs/week — agentic operations moat.
20. → Convergence: the 2026 moat is no longer the database but the execution base (traces, decisions, skills). #### Productive tension with MIT NANDA "95% pilots fail"
21. MIT NANDA. (cited in DORA 2026 as a "pessimistic perspective"): 95% of AI pilots fail, shadow AI economy.
22. Bain. 90% of the market still uncaptured.
23. → Correct reading: the two converge — 95% of pilots fail precisely because 90% of the market remains unstructured; the players who succeed in turning the pilot into a product will be those who capture the agentic opportunity. Bain is the strategic framework for turning pilots into products. #### Sectoral convergence "Sierra"
24. Bain. cites Sierra as a reference agentic example.
25. Sierra Iyengar/Asemanfar/Wang. (2026-04-22): AI-native interview Plan/Build/Review.
26. Taylor Sierra. (2026-04-20): engineering hiring overhaul.
27. Sierra Iyengar/Asemanfar/Wang AI-native interview. (2026-04-22).
28. → Sierra is featured in 3 notes in the dossier — an emblematic position for cross-workflow decision context in customer support. ### To be used for
29. SaaS / PE software executive presentations. quantified sizing of the opportunity ($100B US / $200B extended) — a canonical reference for business cases.
30. B2B SaaS product strategy. reframe the product brief in terms of cross-system coordination labor to automate.
31. Investors / VCs. the six-factor grid serves as a quick due diligence tool for evaluating an agentic dossier.
32. CDOs / Data leaders. redesign data foundations for agent readiness becomes a priority, budgetable project.
33. FR / Europe connection. Bain provides the US framework; to be cross-referenced with Wescale (Usine Logicielle Augmentée) for French executive committee presentations.