We build custom CRM systems. We also talk a good share of the people who ask for one out of buying it from us. That is not modesty — a custom system built for a company that did not need one is the most expensive kind of software there is, because it has all the cost of being bespoke and none of the advantage.

Here is the test we use.

The test: remove the CRM and see what changes

Take your sales process as it actually runs today, including the workarounds. Now imagine your competitor has exactly the same CRM as you. Does anything about your position change?

If the answer is no, buy the box. Your advantage is somewhere else — your people, your product, your prices — and a custom system will not move it. Boxed CRMs are genuinely good, they are cheap per seat, and somebody else pays to keep them running.

If the answer is yes, and you can say specifically what changes, you have found the part worth building. Usually it is one of three things:

  • Your pipeline is not a pipeline. Deals move sideways and backwards, approvals branch by amount and by who is asking, and half your stages exist because of a rule that is genuinely yours.
  • Your pricing is a calculation, not a number. It depends on volume, on distance, on grade, on last year's agreement. Everyone with a spreadsheet is a single point of failure.
  • Your history is the asset. Ten years of what was quoted, to whom, at what, and why it was lost. It has to come with you, in a shape that can be queried.

The cost of being wrong, in each direction

Wrong towards buying: you pay per seat forever, you contort your process into somebody else's shape, and the part that made you money becomes a text field somebody fills in inconsistently. It is a slow, survivable loss.

Wrong towards building: you pay once, up front, and then keep paying to maintain something you did not need. A first version from us is $20,000 and ten weeks. Spent on a process that a $30-a-seat product would have handled, that is a bad quarter, and it is the failure mode we work hardest to prevent — which is why the first conversation is about talking you out of it.

The shape that usually wins

It is rarely all or nothing, and the answer that works most often is neither: keep the boxed product for the commodity parts, build one small system for the part that is genuinely yours, and connect them.

Your reps keep the interface they know. The pricing calculator, the approval rules, the thing your business actually runs on, becomes a small system that speaks to it over an API. Far cheaper than replacing everything, and the custom work sits exactly where it earns its cost.

That small system is the size we build first: one journey working end to end, in weeks rather than quarters.

What to bring to the conversation

Write down your process as it runs today — with the workarounds included, because the workarounds are the requirements. Then get a demo of two boxed products using your data, not theirs. Then ask a studio to price the custom build against the same document.

If the boxed products cover eighty percent and the missing twenty is not where you make your money, buy. If the missing twenty is exactly where you make your money, build that part and integrate the rest. And the article on what a custom operations system costs has the same test applied to a bigger class of system, if yours is more than sales.