Most companies should not build their own CRM. HubSpot, Salesforce, Pipedrive and their peers are cheap compared with a custom build, and they are very good at the job they were designed for: tracking leads and deals through a sales pipeline.
The trouble starts when your customer relationship is not a pipeline. When the customer signs up, pays every month, books sessions, owes money, brings their family or team, and the important events happen inside your own product. Then the CRM sees only the outside of the relationship, and staff work in two systems that never quite agree.
This post helps you tell which situation you are in, and what a custom build has to cover if you go that way.
When an off-the-shelf CRM is right
Buy one if most of these are true:
Your main process is sales: leads, calls, proposals, deals won or lost.
Customers buy occasionally, and what happens after the sale lives in another tool.
Your team needs email sequences, call logging and forecasting more than custom workflows.
You have fewer than a few hundred active accounts to manage by hand.
Nobody on your side wants to own software.
Spend the money on setting it up well: clean pipeline stages, required fields, and an integration with your billing tool. That beats any custom build for a sales-led business.
When customer data belongs in your platform
A custom build, or more often CRM features inside the platform you already run, pays off when:
The relationship happens in your product. Usage, bookings, attendance or orders are the signals that matter, and they live in your database.
Money is recurring and has rules. Subscriptions, dues, discounts, instalments, grace periods.
Accounts have structure. A parent with several children, a company with several users, a partner with several sites.
Staff spend their day switching tabs between the CRM and the admin of your app, copying notes and statuses.
Per-seat pricing hurts. Front-desk staff, teachers or field agents all need access, but only to a small part.
In that case, the CRM you need is a set of screens and rules on top of data you already own. It is usually smaller than people expect.
An example: a school's families, dues and progress
Meraki is a web and desktop platform for a children's language school. Its "customers" are families: more than 2,600 students and 780+ parents, with 870 subscriptions. A sales CRM would hold the parents' contact details and little else.
The platform holds the whole relationship instead. Subscriptions, payments, dues and discounts are tracked per student. Parents and teachers follow each child's progress through lessons and voice exercises. The front desk runs a desktop app wired to a time clock, a receipt printer and a TV display, synced with the cloud. When staff need to know who owes what, the answer comes from the same system that holds the subscriptions and payments.
That is the case for building: the important data was already in the product, and the work was to give staff the right views and rules on top of it. The full build is in the Meraki case study.
What a custom CRM has to cover
If you build, these are the parts people forget until they need them:
| Part | Why it matters |
|---|---|
| One record per customer | Duplicates are the first thing that breaks trust in the data |
| History of every status change | "When did they downgrade, and who did it?" needs rows, not overwritten fields |
| Roles | Front desk, finance and managers each see a different slice |
| Reminders and tasks | Follow-ups, renewals and overdue payments without someone remembering |
| Email and messaging | Usually sent through a provider, logged on the customer's record |
| Exports and an API | Accounting, marketing tools and reporting will want the data |
| Tests on the money flows | Dues, discounts and refunds are where silent bugs cost real money |
A hybrid is often the best answer: keep a light sales CRM for prospects, and move a contact into your platform the moment they become a customer.
What moves the cost
The biggest factors are how much of the data already lives in your own system, how complex the billing rules are, how many integrations you need, and whether old records have to be cleaned and imported. Adding customer views to an existing platform costs far less than building a CRM from scratch. Our platform builds start from $15k, and we quote after a short discovery. For how to read that quote, see fixed price vs. time and materials. The same build-or-buy reasoning applies to ERPs, which we cover in Odoo vs. custom.
Next step
Tell us what your customer relationship looks like after the sale, and which tools your team switches between today. On an intro call we will say whether a CRM you can buy covers it, or what belongs in your own platform. Book an intro call.