Home/Blog/Build vs Buy
Build vs Buy

Rebuilding a Bespoke CRM as the System of Revenue Truth: What We'd Do Differently

Direct answer · What did you learn from rebuilding a bespoke CRM?

Three things. Sequencing the data-model work before the interface work was correct and we would do it again. Running the parallel-run validation for two full cycles rather than one was the decision that de-risked the migration. And we would argue harder, earlier, for defining stage entry and exit criteria before any build began — we treated it as a workshop input when it was really the engagement's critical path.

Summary

Publishing your own second-guessing is rare in consulting. It is also the only way a case study tells you anything useful.

Why a retrospective rather than a case study

A conventional case study describes a problem, an intervention and a result, and every one of them is written to make the firm look competent. They are not lies, but they are selected — and a reader learns very little from a story in which nothing went sideways.

This is the same engagement written differently: what we would keep, what we would sequence differently, and the one thing we should have pushed harder on. The client is anonymised under confidentiality, as all our case work is.

The situation

A specialty lending business running high-velocity deal flow through a CRM that had drifted a long way from how deals actually moved. Underwriting, sales and servicing each kept their own records. Pipeline reporting required manual assembly. Stage data was unreliable enough that leadership had no defensible view of conversion or capacity.

Deal velocity is the business model in specialty lending. Every hour of manual reconciliation and every mis-staged deal was constraining funded volume directly.

What we would keep

  • Data model before interface. We rebuilt the deal record and its lifecycle before designing a single screen. It felt slow to the client for three weeks and saved months later — every screen built afterwards was building on something settled.
  • Two parallel-run cycles, not one. We ran the new system alongside the old for two full deal cycles before cutover. One cycle would have looked sufficient and would have missed the edge cases that only appear at month-end.
  • Joint design with underwriting and sales in the same room. Sequential design — sales first, underwriting later — was the obvious efficiency and would have been a mistake. The handoff between them was where the actual friction lived.
  • Migration validation gates. No batch moved without a reconciliation check that had to pass. Slower, and the reason nobody spent the following quarter hunting for missing records.
The stage definitions were the critical path. We treated them as a workshop input. That is the thing we would change.

What we would sequence differently

We ran the stage-definition work as a workshop during discovery — a good workshop, with the right people, that produced a clear document. Then we started building.

In hindsight the definitions were not an input to the build; they were the build's critical path. Two of the three schedule slips traced back to a stage whose exit criteria turned out to mean different things to underwriting and to sales, discovered when the automation encoding it behaved 'wrong' for one of them.

What we would do now: treat stage definitions as a deliverable with its own sign-off gate, and have two leaders independently write the criteria before comparing. Where the two lists diverge is the real specification, and it is cheaper to find in week two than in week nine.

The decision we would argue harder about

The client wanted to migrate seven years of historical deals. We raised the cost, they weighed it, and we migrated it. That was a defensible call on the information available.

What we know now is that almost none of it has been queried since. The reporting that matters looks at eighteen months. The historical data added meaningful time to the migration, meaningful risk to the validation, and produced very little.

We would now push a specific alternative rather than simply flagging cost: archive everything older than two years to cold storage, migrate the rest, and agree a defined path to retrieve archived records if anyone ever needs them. Nobody has, on any engagement since, in our experience.

What actually changed for the business

Outcome categories observed, in the format we publish all client results — directional, not invented:

  • One deal record and one lifecycle shared across sales, underwriting and servicing
  • Automated handoffs replacing email-and-spreadsheet coordination between the three functions
  • Executive pipeline reporting produced by the system rather than assembled around it
  • Stage data leadership was willing to make capacity decisions from

Where exact figures are not approved for publication we do not invent them. That constraint is why the retrospective format is more useful than the numbers would have been anyway.

The transferable lesson

If you take one thing: the expensive part of a CRM rebuild is almost never the software. It is the organisational agreement the software forces you to reach. Most of what looked like project risk on this engagement was actually unresolved process disagreement surfacing under deadline pressure.

Which means the cheapest thing you can do before any rebuild — bespoke or platform — is get two leaders to write the stage criteria independently and compare. It costs a morning. It tells you whether you have a systems problem or an alignment problem, and those need very different projects.

Frequently asked

In this case yes — the deal workflow was genuinely unlike anything a platform models, and that is the narrow test that justifies bespoke. Building to save on licence fees is almost always a mistake; building because your core workflow is a competitive advantage is not.

Longer than the original estimate, and the overrun traced almost entirely to the stage-definition issue described above rather than to engineering. That is the honest answer and it is why we now gate that work separately.

Two things. Export the data and count duplicates and dormant records — that percentage predicts migration cost better than anything else. And have two leaders independently write the stage criteria. Both are free and both are diagnostic.

No. Client names are withheld under confidentiality on every engagement. The sector and scale are published so the context is real; the identity is not ours to give.

CE
Corelynx Editorial · Corelynx · info@corelynx.com
Keep reading
Salesforce & Agentforce August 28, 2026 10 min read

Salesforce Optimisation vs Replatforming: How to Tell Which You Need

Most mid-market companies are not buying their first CRM. They are deciding whether to fix the one they have or leave it.

Read the article
Salesforce & Agentforce August 24, 2026 13 min read

Agentforce Readiness: Six Evidence Tests, and What Failing Each One Costs

The demo is always impressive. What determines whether it survives production is six things, and none of them are the agent.

Read the article
AI Architecture August 21, 2026 9 min read

Most AI Pilots Fail Before the Model Is Chosen

An argument against buying AI tooling, from a firm that sells AI transformation.

Read the article
AI Architecture August 17, 2026 12 min read

The AI Architecture Decision That Determines Whether You Can Ever Switch Providers

Most AI lock-in is not contractual. It is architectural, self-inflicted, and decided in week two of the pilot.

Read the article
Technical Leadership August 14, 2026 8 min read

Why We Tell Some Founders They Don't Need a Fractional CTO

A meaningful share of founders who ask us for a fractional CTO do not need one. Telling them costs us a deal and buys something worth more.

Read the article
Technical Leadership August 10, 2026 12 min read

The 12 Decisions a Founder Shouldn't Make Without Technical Leadership

Most technical debt is not written in code. It is decided in a meeting, by someone without the context to know what they are choosing.

Read the article
Build vs Buy August 3, 2026 11 min read

What a Production-Grade MVP Actually Includes

Two quotes for the same MVP, $20,000 apart. The gap is almost never features — it is the eight things that do not demo.

Read the article
Adoption & Change July 29, 2026 9 min read

Your CRM Will Not Fail on Technology. It Will Fail on Adoption.

The post-mortem on a failed CRM almost never blames the software, and almost always blames "adoption". But adoption is an outcome, not a cause. Here is what produces it.

Read the article
CRM Strategy July 22, 2026 12 min read

How to Develop a CRM Strategy Before You Buy Anything

Most CRM strategies are written after the platform is chosen, which makes them implementation plans wearing a strategy label. Here are the four decisions that have to come first — and the order they go in.

Read the article
Revenue Operations July 15, 2026 11 min read

Is Your CRM Delivering? Six Tests It Should Pass

Most CRM reviews measure adoption — logins, records created, fields filled. Those tell you the system is being used, not that it is working. Here are six tests that measure whether it earns its cost.

Read the article
Build vs Buy July 8, 2026 10 min read

Why an Off-the-Shelf CRM Is Not Enough for Your Company

Packaged CRM is the right answer more often than custom-software firms like to admit — and the wrong answer more often than buyers realise. Here is the decision framework, including the cases where you should not build.

Read the article
Field Operations June 30, 2026 9 min read

Mobile CRM: What Field Teams Actually Need From It

Every CRM has a mobile app and most field teams do not use it. The reason is a design decision, not a technology limit — mobile gets treated as a smaller desktop rather than a different job.

Read the article
Revenue Operations June 24, 2026 16 min read

What Is Revenue Intelligence? The 2026 Executive Guide

Vendors use the term for everything from call recording to dashboards. Underneath the noise is a real discipline — here's the plain-language version, with a maturity model and a starting sequence.

Read the article
AI Architecture June 18, 2026 17 min read

Public vs. Private LLMs: An AI Architecture That Protects Your Data

You don't have to choose between AI capability and data privacy — and you definitely don't have to marry one vendor. The architecture that solves both, explained in plain language.

Read the article
Salesforce & Agentforce June 10, 2026 14 min read

Salesforce Agentforce Implementation Cost in 2026: A Transparent Breakdown

Agentforce ARR is growing 205% year over year, and every Salesforce AE has quota pressure to sell it. Here's what implementation actually costs — and the readiness question to answer before spending anything.

Read the article

Talk this through with a practitioner.

The first conversation is about context and fit — nothing more.

Book a Strategy Session

Keep exploring.

See everything under Blog.

Browse Blog
Book a Strategy Session