Case study: replacing a nine-year-old CRM in fourteen weeks
Forty-one undocumented behaviours, 1.8 million records and a sales team that could not stop selling for a weekend. Here is how the sequencing worked.
Key takeaways
- Discovery found 41 behaviours that were not in any document or ticket.
- Capability-by-capability cutover kept the pipeline live throughout.
- Go-live week generated 6 support tickets, none of them severity one.
What was delivered?
A nine-year-old CRM instance holding 1.8 million records was replaced in fourteen weeks with no sales downtime. Discovery documented 41 undocumented behaviours, cutover ran capability by capability, and go-live week produced six support tickets with no severity-one incidents.
- 14 weeks
- Elapsed time
- 1.8m
- Records assessed
- 41
- Undocumented behaviours found
- 0
- Go-live severity-one incidents
Weeks one to three: archaeology
We instrumented the live instance and watched what was genuinely read and written, then interviewed every team with a workflow inside it. The behaviour inventory came back at 41 items, of which 9 were rules that no current employee had authored.
Three of those turned out to be obsolete and were retired rather than rebuilt, which is the cheapest kind of scope reduction available on a migration.
Cutover sequence
| Phase | Capability | Parallel running |
|---|---|---|
| Weeks 6-7 | Accounts and contacts | 2 weeks |
| Weeks 8-9 | Pipeline and opportunities | 2 weeks |
| Weeks 10-11 | Quoting and approvals | 1 week |
| Weeks 12-14 | Reporting and archive access | None |
What we would do differently
Reporting should have moved earlier. Leaving it last meant the analytics team spent three weeks working against two sources, which was avoidable and cost goodwill we did not need to spend.
Everything else held. The behaviour inventory was the artefact that made the estimate survive contact with the codebase, and it is now the first deliverable on every migration we quote.
What the archaeology phase cost and saved
Discovery was three weeks and roughly 11% of the total engagement cost. On the original fixed-price quote the client had received elsewhere, that money would have gone into a contingency line instead.
The return was measurable. Nine of the forty-one behaviours would have been discovered during or after cutover, and three of those touched month-end reporting, which is the worst possible time to find them. Two more were rules that the finance team had been manually working around for years, and retiring them removed a recurring task rather than rebuilding it.
We also used discovery to agree what would not move. Roughly 60% of records were archived rather than migrated, a decision made with the analytics team in week two rather than negotiated under pressure in week twelve. That single scope reduction is the reason fourteen weeks was achievable at all.
FAQFAQ
Frequently asked questions
About the author
Sofia runs delivery across engineering engagements — discovery, estimation, increments and handover. She writes about migrations, pricing models and the undocumented business logic that turns a clean estimate into an overrun.
- Enterprise migrations
- Project estimation
- Delivery management
- CRM and ERP systems
