Let us start with the unhelpful-but-honest position: most businesses should not build a custom CRM. Salesforce, HubSpot, Pipedrive and Zoho represent decades of accumulated product thinking, and for the majority of sales processes they are better than anything you would commission. Choosing to build instead is usually an expensive way to recreate features you could have rented.
But “usually” is not “always”. There is a real category of business for which platform CRMs are a permanent tax — in licence fees, in workarounds, and in process distortion. This article is about telling the two situations apart.
The four signals you have outgrown a platform
1. Your process does not fit the object model. Every CRM assumes a shape: leads become opportunities, opportunities become closed deals, deals attach to accounts and contacts. If your business runs on something structurally different — multi-party transactions, recurring project cycles, asset-based relationships, regulated case handling — you will spend years bending custom fields around a model that was never meant to hold your data. The tell is a field called something like Custom_Field_14__c that three people understand.
2. Licence cost has outgrown build cost. This is arithmetic, not philosophy. At 60 users on a £120/user/month plan you are spending £86,400 a year. That is not a subscription; that is a development team. The threshold is lower than most people assume — more on the maths below.
3. Your competitive advantage lives inside the workflow. If how you qualify, price, or route work is genuinely differentiated, encoding it in someone else's tool means capping it at what their configuration allows. When the workflow is the business, owning it matters.
4. Integration is where all the money goes. If most of your CRM spend is middleware, connectors, and sync jobs stitching the platform to your real systems of record, you are already paying for custom development — just in the least efficient possible form.
One signal is not enough. Two should prompt a serious evaluation. Three or four and the case usually makes itself.
The break-even maths
The honest comparison is not licence cost versus build cost. It is total cost of ownership over five years, because custom software has ongoing costs that subscriptions bundle invisibly.
| Cost line | Platform CRM | Custom CRM |
|---|---|---|
| Initial build | £0 | £35,000–£120,000 |
| Licences (40 users, 5 yrs) | £144,000 @ £60/user/mo | £0 |
| Configuration / consultants | £10,000–£60,000 | Included in build |
| Integrations | £5,000–£40,000 | Included in build |
| Hosting (5 yrs) | £0 | £3,000–£18,000 |
| Maintenance (15%/yr) | £0 | £26,000–£90,000 |
Run those totals and the crossover typically appears somewhere between 25 and 50 users, arriving sooner if your configuration and integration spend is high. Below roughly 20 users, building is very rarely justified on cost alone. Above 100 users with a genuinely non-standard process, the platform is usually the more expensive option — you simply pay for it in a way that never appears as a capital line.
You can model your own numbers with our build vs buy decision tool and the cost estimator.
What custom CRM development actually costs
| Scope | Typical cost | Timeline |
|---|---|---|
| Focused pipeline tool (contacts, deals, activity, one integration) | £18,000–£35,000 | 8–14 weeks |
| Full CRM (roles, reporting, automation, 2–4 integrations) | £35,000–£80,000 | 3–6 months |
| Sector-specific platform (regulated data, complex workflow, portals) | £80,000–£200,000+ | 6–12 months |
The cost driver that surprises people is not the CRM features — it is reporting. Everyone underestimates how much of a CRM's value is in its dashboards, filters, and exports, and how much engineering that represents when you build it yourself.
The middle path most people miss
The choice is not binary, and the best answer is frequently a hybrid: keep the platform for what it is genuinely good at, and build only the part that does not fit.
In practice that looks like keeping HubSpot or Pipedrive for contact management, email tracking, and standard pipeline reporting — then building a custom application for the specialised workflow, connected by API. You get commodity features at commodity prices and bespoke capability exactly where it differentiates you.
This is usually cheaper than either extreme, and it dramatically reduces risk: if the custom piece disappoints, you still have a functioning CRM. Almost every organisation that tells us they need a full custom CRM turns out to need this instead.
What you lose by building
An honest account of the trade-off:
- The ecosystem. Hundreds of pre-built connectors, an app marketplace, and an army of people who already know the tool.
- Free feature velocity. Platforms ship improvements you did not pay for. Your custom CRM improves only when you fund it.
- Hiring familiarity. New staff arrive already knowing Salesforce. Nobody arrives knowing yours.
- Someone else's uptime problem. Availability, backups, security patching and compliance become yours.
- Speed to value. A platform is running this week. A custom build is months away.
That last point deserves weight. The opportunity cost of six months without improved tooling is real, and it is rarely in the business case.
How to decide, in order
- Document the actual process — the one people follow, not the one on the org chart.
- Try to configure it in a platform, honestly, with someone competent. Many “impossible” requirements are unfamiliarity with the tool.
- Cost the workarounds in hours per week. If people spend six hours a week in spreadsheets because the CRM cannot do something, that is £15,000–£25,000 a year of hidden cost.
- Run five-year TCO both ways, including maintenance on the custom side.
- Consider the hybrid before committing to a full build.
If you complete that sequence and building still wins, you almost certainly should. The businesses that regret building a custom CRM are the ones that skipped step two.