Ask three studios what a CRM costs and you will get three answers far apart. They are usually all telling the truth about different things, because "CRM" is a category, not a specification.
Here is what actually moves the number, in order.
1. History you are bringing with you
This is the one people forget, and it is routinely a quarter of the project. Ten years of deals in a boxed product, or worse, in four spreadsheets and an inbox. Importing it is not a script — it is a series of decisions:
- Which of the three records for the same client is the real one?
- What does a deal with no owner become?
- Do you keep the fields nobody has filled in since 2021?
Every one of those is a conversation with someone in your company, and the import is only as good as those answers. A quote that does not mention data migration is not cheaper. It is incomplete.
2. Integrations, and whose schedule they run on
Accounting, telephony, the warehouse, email. Each integration is a small project with someone else's timeline attached — their API, their sandbox, their support queue. Two are normal. Five is a different project.
3. Rules, not screens
Ten stages with no logic is a cheap CRM. Four stages where the transition depends on amount, on role, on whether the client is in arrears, and on someone's approval, is an expensive one. When we ask "what has to be true before a deal can move here", the length of that answer is the estimate.
4. How many kinds of people use it
One sales team is one interface. Sales, plus a manager who needs a different view, plus finance who needs a third, plus a director who wants a number on a phone, is four. Roles multiply the surface far faster than features do.
Our number, and what is inside it
A first version is $20,000 and ten weeks. What that buys: your pipeline as your business actually runs it, your pricing rules, roles and access, your history imported, and one integration. What it deliberately does not buy: every report you can imagine, a mobile app, and the second and third integrations — those wait for phase two, and they are written down as waiting rather than quietly omitted.
Ten weeks rather than four is mostly items one and three above. The pipeline that did not fit a box goes in first, because it is the reason you are not buying a box.
Getting a comparable quote
If you want three quotes you can actually compare, give all three studios the same document: your process as it runs today with the workarounds included, the list of systems it must talk to, and a rough count of the records coming with you. Then ask each of them what they would leave for phase two.
The answer to that last question tells you more than the price does. A studio that leaves nothing out has either not understood the project or is planning to discover it later, at your expense. How to brief a studio so the estimate means something goes through the rest of that comparison, and the test for whether you need a custom CRM at all is worth doing before any of it.