Meridian — AI Document Processing

The scenario

Meridian is an enterprise contract management platform. Competitors now offer native AI that reads, extracts, and flags contract clauses automatically. Meridian has no AI capability. The sales team is consistently losing deals at the final stage. The board is alarmed.

Constraint

You must present a recommendation to the board next week with a clear rationale.

Your task

Recommend build vs buy vs partner and defend the trade-offs.

1. Clarify: What do you need to know before evaluating the options?

Pick the 3 questions you'd ask first.

Model approach (strong picks)

• Strong pick

  What specific AI capabilities are customers asking for — clause extraction, risk flagging, plain-language summaries?
• Strong pick

  What is the time-to-market for each option — when can we actually ship something customers can see?
• Strong pick

  Is AI document processing a temporary differentiator or a permanent core capability — will this be table stakes in 2 years?

2. Hypothesize: What are the key trade-offs between the three options?

Pick the 3 most critical factors to weigh.

Model approach (strong picks)

• Strong pick

  Build gives full control and differentiation but 18–24 months to ship — we keep losing deals in the meantime.
• Strong pick

  Partner is fastest to market (weeks not months) but creates dependency on a vendor and margin compression at $3M/yr.
• Strong pick

  Buy (acquire) gives capability + team fast but $8M is a significant capital event and integration risk is high.

3. Investigate: What do you nail down before the board meeting?

Pick the 3 highest-leverage preparations.

Model approach (strong picks)

• Strong pick

  Get a time-to-demo estimate for each option: how fast can a prospect see a working AI feature?
• Strong pick

  Talk to 5 lost prospects: is AI the deciding factor, or was price or fit the real issue?
• Strong pick

  Model the revenue impact: at current deal loss rate, what does 12-month delay cost in ARR vs. each option's cost?

4. Propose: What is your recommendation to the board?

Pick the one strategic path you'd recommend.

Model approach (strong picks)

• Strong pick

  Partner now, build later — sign a 12-month partnership to stop deal losses immediately, while a small internal AI team begins building the proprietary capability in parallel for year 2.
• Strong pick

  Acquire — $8M for a specialist team and proven IP is faster and safer than building from scratch, and cheaper than 18 months of lost deals.

Related questions

• How would you decide whether to enter the European market?
• How would you build the case for a multi-year product investment that won't show ROI for 18 months?
• How do you set product strategy when the CEO keeps changing their mind?
• How would you approach building a product in a regulated industry (fintech, health, etc.)?

Solve this case study