B2B/B2C SaaS Private Information Cards#

---

Card 1 — Round 1#

Title: AI Feature Costs Erode Margins & Third-Party Infrastructure Expense Escalates

Card Type: Operational Intelligence

Reveal Timing: Round 1 — distribute face-down at start of decision preparation phase

Classification: Restricted

Source: Internal finance analysis, cloud infrastructure billing data, and engineering cost modeling

Shared Intelligence (also received by Big Tech — framed differently):

The Intelligence:

Your internal cost models for AI feature delivery are tracking significantly above projections. The AI features you have deployed (copilots, predictive analytics, intelligent search, automation workflows) are consuming more cloud compute, more third-party API calls, and more engineering maintenance than anticipated. Gross margin on AI-powered features is running 10-15 points below your legacy product margin — and the gap is widening as adoption scales and inference volume grows.

Specific cost pressures:

• Third-party API costs (OpenAI, Anthropic, Cohere) are the largest line item. Your negotiated volume pricing is better than retail, but at scale the cost is material — approximately $2-3 per active user per month for AI features, against a blended subscription price of $25-30 per user per month. As AI feature adoption increases across your customer base, this cost grows linearly while subscription revenue grows only if you can upsell.
• Internal fine-tuning and model optimization work is consuming 30% of your AI engineering capacity, leaving less bandwidth for new feature development. The "build once, deploy everywhere" assumption has not held — each vertical and customer segment requires model customization.
• Vector database and embedding storage costs are a new and growing expense category that was not in your original infrastructure budget.

Competitive intelligence suggests all major SaaS companies are experiencing similar margin pressure from AI features. The companies managing costs most effectively are those with disciplined make-vs-buy strategies — leveraging open-source models for commodity tasks and reserving proprietary development for high-differentiation use cases.

Decision Tension:

Do you absorb the margin compression as an investment in competitive positioning (betting that AI feature adoption will drive retention and upsell that offsets the cost)? Or do you aggressively optimize AI infrastructure costs — switching to open-source models, reducing third-party API dependence, and limiting AI feature scope — to protect margins now?

Questions to Consider:

• What is the true gross margin on your AI features at current scale? At projected scale in 12 months?
• Which AI features can be delivered using open-source models without meaningful quality degradation? Which require proprietary APIs?
• Can you pass any AI infrastructure costs to customers through pricing changes, or will the market not bear it?
• What is the competitive consequence of slowing AI feature development to protect margins? How quickly do AI-native competitors gain ground?
• What is your break-even point — the adoption level at which AI feature revenue (upsell, retention) exceeds AI infrastructure cost?

---

Card 2 — Round 2#

Title: Enterprise Customers Demand Free AI & Competitive Displacement Accelerates

Card Type: Competitive Intelligence

Reveal Timing: Round 2 — distribute face-down at start of decision preparation phase

Classification: Confidential

Source: Enterprise sales team intelligence, customer renewal negotiations, and competitive win/loss analysis

The Intelligence:

Your top 20 enterprise customers have made their position explicit in contract renewal discussions: they expect AI features to be bundled into their existing subscriptions at no additional cost. The framing from customers is consistent — "AI is table-stakes, like search was a decade ago. We shouldn't pay extra for it." Willingness to pay a premium for AI capabilities is minimal across your customer base: only 5% of accounts are willing to pay 10%+ more for AI-powered features.

The situation is compounding:

• Deal slippage: Your enterprise sales cycle has extended by 30-40%. Prospects are evaluating AI-native alternatives in parallel with your platform. Three deals representing $45M in annual recurring revenue are stalled as customers compare your AI capabilities against specialized competitors.

• Churn signals: Three of your top 20 accounts (representing 8-10% of total revenue) have initiated vendor evaluation processes. Two are evaluating AI-native CRM/ERP competitors; one is building internal AI tools on open-source models. Your customer success team assesses these as "high risk" renewals.

• Mid-market displacement: Below your top accounts, AI-native startups are winning new customer acquisition in your mid-market segment. Five accounts that would typically have been your wins in the past quarter chose AI-native alternatives, citing faster AI capabilities, lower cost, and simpler implementation. Your win rate in competitive deals has dropped from 45% to 32% over the past two quarters.

• Pricing contagion: Customers who have seen competitors offer AI features at no additional cost are now demanding the same from you. Your pricing team estimates that bundling AI features into base subscriptions would reduce blended revenue per user by 8-12%.

Decision Tension:

Do you bundle AI features into base subscriptions to defend market position and reduce churn risk (accepting margin compression)? Or do you hold premium pricing for AI features and invest in differentiation to justify the premium (risking accelerated churn and competitive displacement)?

Questions to Consider:

• What is the revenue impact of bundling AI into base subscriptions vs. the revenue impact of losing 8-10% of customers to competitors?
• Can you create a tiered AI offering — basic AI bundled, advanced AI as premium — that satisfies most customers while preserving some pricing power?
• How do you respond to the three high-risk top-20 account renewals? What retention investment is justified?
• What is your competitive win rate trend, and at what point does declining win rate become an existential threat?
• If you bundle AI, how do you communicate this to investors without triggering concerns about margin erosion?

---

Card 3 — Round 3#

Title: AI Talent Cost Spiral & Roadmap Execution Constraints

Card Type: Workforce Intelligence

Reveal Timing: Round 3 — distribute face-down at start of decision preparation phase

Classification: Restricted

Source: HR department talent market analysis, engineering leadership assessment, and compensation benchmarking

The Intelligence:

AI engineering talent costs are spiraling beyond your compensation model's capacity. The situation has moved from "competitive pressure" to "structural constraint on execution."

Specific signals:

• Salary inflation: AI engineering compensation (base + equity) has increased 25-30% year-over-year. Your budget assumed 10-15% increases. Senior AI engineers (5+ years experience in ML/AI) are commanding $400K-$600K total compensation packages. Your compensation structure — designed for enterprise software engineers — is not competitive without case-by-case exceptions that create internal equity problems.

• Attrition acceleration: You have lost 8 AI engineers in the past two quarters (out of a team of 45). Four went to well-funded AI startups; two went to Big Tech; two went to financial services firms. Each departure creates a 3-6 month productivity gap and a 6-12 month replacement timeline. Your AI feature roadmap is now 2-3 months behind plan, with the gap widening.

• Offer acceptance decline: Your offer acceptance rate for senior AI engineering roles has fallen from 65% to 40% over the past year. Candidates are receiving multiple competing offers and choosing based on equity upside (startups), problem quality (Big Tech/research labs), or compensation (financial services). Your enterprise software brand is less attractive to AI-native talent than consumer tech or pure-AI companies.

• Downstream effects: The engineering capacity constraint is rippling through your product organization. Feature launches are being delayed. Quality issues are emerging as remaining engineers are stretched across too many projects. Customer-facing bugs in AI features are increasing, damaging the perception of AI capability that you are trying to build.

• Services workforce transition: Separately, your customer success and professional services teams (1,500+ employees) are seeing role automation from AI tools. Voluntary attrition in these teams is rising as employees sense their roles are at risk. You face a dual challenge: acquiring expensive AI talent you cannot easily afford while managing the transition of a large existing workforce whose roles are changing.

Decision Tension:

Do you dramatically increase AI talent compensation (breaking your compensation model and creating internal equity issues) to retain and recruit competitively? Or do you accept slower AI roadmap execution and focus on retaining your current team through non-compensation levers (interesting problems, career development, stability) while selectively using M&A and contracting to fill gaps?

Questions to Consider:

• What is the revenue impact of a 3-6 month AI roadmap delay? How does it affect competitive positioning and customer retention?
• Can you restructure compensation specifically for AI engineering roles without creating morale problems across the rest of the engineering organization?
• What is the true cost of replacing an experienced AI engineer (recruitment, ramp-up, lost productivity)? Is retention investment cheaper?
• Can M&A (acqui-hiring AI startups) solve the talent problem faster than organic recruitment? What are the integration risks?
• How do you manage the services workforce transition? What is the cost of proactive retraining vs. reactive restructuring?
• Is there a talent strategy that addresses both sides — acquiring AI talent and retaining/transitioning existing talent — without breaking the budget?