When your business actually needs a CRM

Every growing business is told it needs a CRM. Most are told too early, buy something expensive, use 5% of it, and quietly go back to the spreadsheet.

A CRM is not a maturity badge. It's a solution to specific problems. If you don't have those problems yet, a well-maintained spreadsheet is faster, cheaper and easier to change.

The signals that actually matter

More than one person touches a deal. The moment two people need the same context, shared memory becomes a system requirement rather than a convenience.

You can't answer "what's in the pipeline?" in ten seconds. If the answer requires reconstruction from an inbox, you're managing sales from memory.

Leads are being forgotten. Not lost to competitors — forgotten. This is the clearest signal, and the most costly.

You have more than roughly 30 open opportunities at once. Around there, human tracking reliably fails.

Follow-up depends on someone remembering. Systems should hold the schedule, not people.

You need history. When a client returns after eight months, what was discussed matters — and it's usually in a departed employee's inbox.

If fewer than three of these are true, stay where you are and improve your process instead.

What a CRM must do for a small business

Only four things, honestly:

• Hold every lead and customer in one place, with source
• Show pipeline stage at a glance
• Record the history of contact
• Trigger the next action, with a date and an owner

Everything else — forecasting models, territory management, elaborate scoring — is for later or never.

Choosing without regret

Prefer the simplest tool your team will actually use daily. Adoption beats capability every time; an unloved CRM produces worse data than a spreadsheet, because people half-fill it and everyone stops trusting it.

Three practical filters: does it capture leads from your real channels without custom work, can a new team member understand it in fifteen minutes, and can you export everything if you leave.

Migrating in a day, not a quarter

1. 1Define your stages first — five or six, each with an unambiguous entry condition. This is 80% of the value and takes an afternoon of clear thinking.
2. 2Import only active and recent contacts. Old dead leads add noise, not history.
3. 3Set required fields to the bare minimum: source, stage, owner, next action date.
4. 4Move all lead capture to the CRM on day one, so nothing arrives outside it.
5. 5Delete the spreadsheet. Two systems means no system.

Making it stick

One rule keeps a CRM alive: every open deal has a next action with a date and an owner, or it isn't open.

Run a fifteen-minute weekly pipeline review against that rule. Deals with no next action either get one or get closed. Within a month the data becomes trustworthy, and once it's trustworthy people use it without being asked.