JAS Diamonds: a memo risk and sales intelligence layer on a legacy ERP

The problem we took

JAS Diamonds is a family-owned wholesale diamond business in downtown Chicago, founded in 1983. It had about 2,130 open memos worth over $20M outstanding at any time and no reliable way to track them. The ERP was a legacy, vendor-hosted system with no API. Reps worked from memory and instinct; nobody knew which memo was old, which account was slow to pay, or which retailer to call next. Follow-up was manual email, one at a time. Credit risk lived in a third-party report nobody cross-checked. After months on a mainstream CRM, fewer than 1 in 2,000 company records had a valid owner.

$2M sat past due across 234 accounts. Reps spent the first hour of each day deciding who to call with no data behind the choice. Memo status chases took days of staff time each week. Slow-pay accounts aged silently until the stone was gone or the balance was stale.

What the client already had

The data, locked inside the ERP, and a sales team that knew its accounts. The client keeps the final say on every outbound email: a staff member approves each batch before anything sends.

What we built

An API service that reads the ERP’s internal data and serves it through 28 clean, versioned endpoints, since the vendor offers none. A scoring service that rates every open memo 0 to 100 for risk every 15 minutes on age, value and credit data. A follow-up engine that drafts status requests for aged memos and past-due invoices, with Gmail send, inbound reply detection and HubSpot contact and owner sync. A sales workbench that gives each of 16 reps a ranked daily call queue with full account context. A natural-language assistant on Claude that answers sales, margin, memo and inventory questions and exports spreadsheets. And a read-only MCP server, with query allowlists, scrubbed credentials and per-user audit logs, so the client’s own claude.ai sessions can safely query live ERP data; an Agent Skill teaches Claude the schema.

Before the model choice, 3Pillars ran a 30-prompt bake-off across four vendors: Claude Haiku 4.5 and Claude Sonnet 4.6 beat two open-weight alternatives on tool accuracy and speed. Daily RapNet market prices (about 283,000 listings at about 95% coverage) feed buying intelligence, and a 181-test automated suite guards the follow-up engine before every deploy.

What it returned

The client’s team sees the whole memo book, ranked by risk, every morning. The system manages more than $20M of memo exposure and chases $2M in past-due balances on a fixed twice-monthly cadence; faster collection on even 1% of past-due balances offsets the platform’s full annual cost. Reps open the day with a reasoned call list instead of a blank slate, slow-pay accounts surface instead of aging silently, and every outbound customer email passes human review.

Figures come from the platform’s own databases and sync audits, April to August 2026, compiled by 3Pillars. Model choice for the assistant: Claude Haiku by default, accessed through an API gateway.