Building a first real sales system for an AI company selling into enterprise

The problem

The play

1. 1
   Interactive growth model to pressure-test the plan

   Built a live, interactive growth-model app pre-loaded with the company's own numbers and market benchmarks (baseline, stretch and conservative scenarios) and ran it in a working session to pressure-test the plan. The plan-versus-capacity view surfaced under-resourcing on both account executives and customer success.

2. 2
   CRM audit before touching anything

   Read-only connection plus a data-quality baseline of the existing portal: subscription and seat entitlements, marketing-contact usage, the single active workflow in the account, and super-admin access.

3. 3
   GTM lifecycle design, signed off before build
4. 4
   Built directly in production
5. 5
   Rep and executive cockpits plus inbound routing

   Scoped clean rep and executive dashboards, a redesigned website contact-sales form (enforce corporate email, drop a low-value seats field), and chat-based inbound lead alerts with round-robin routing to replace an executive triaging inbound by hand.

6. 6
   Product signal to PQLClay

   Scoped the product-signal-to-PQL pipeline out of the data warehouse so product-qualified leads route to the right motion instead of sitting in product analytics, plus lead-to-account matching in Clay and penetration reporting across parent-and-subsidiary account hierarchies.

What it moved

From a LeanScale case studyRead the full case study →