How early is too early for a CRM

A spreadsheet is the correct CRM for a company where one person sells and remembers everything.

That stops being true at a point which has very little to do with how many people work there.

Headcount is the wrong trigger

The advice usually arrives as a number. Ten employees. Twenty. Your first sales hire. Numbers are easy to write down, and they describe nothing that actually happens inside a company.

We have opened portals at companies easily big enough to justify one and found a tool nobody had a reason to use yet, bought because it felt like the responsible thing to do. We have also watched very small teams lose work because nobody could remember what had been promised, to whom, in which month. Size was not the deciding variable in either case.

What actually means you need a CRM

The question is not how big you are. It is whether more than one person needs the same answer about a customer, and whether anybody still remembers it. Some of the things that mean yes:

• Two people are talking to the same customer and neither can see what the other said.
• A deal runs longer than you can hold in your head. Long cycles put the facts somewhere outside memory or they stop being facts.
• Someone asks what happened with an account, and answering takes a search through two inboxes and a question in chat.
• A follow-up depends on a date, and the date lives in one person’s calendar.
• You have promised work to a future version of yourself: a renewal, a check-in, the thing you said you would come back to in March.

None of those is about volume. Each one is about whether what the company knows survives somebody being on holiday.

When a spreadsheet is still the right answer

One person selling, a short cycle, a volume you can read through in a sitting, no handoffs. A spreadsheet handles that, and it handles it better than a CRM nobody has a reason to open.

What matters at that stage is the discipline rather than the software: one row per customer, a column saying who owns it, a column with the next step and a date. A sheet kept that way beats an expensive system three people fill in three different ways. The reason you rarely hear this answer is that the spreadsheet does not bill.

What it costs to start too early

Buying before you know your own process means adopting somebody else’s. The default pipeline stages that ship with every CRM are a vendor’s reasonable guess at how a sale works. They are not how yours works, and the gap stays invisible until the reporting starts contradicting what the team already knows.

Then you pay twice. Once for seats and tiers doing nothing yet, and again for the rebuild when the stages stop matching reality. The rebuild is the expensive half, because by then there are automations, reports and habits attached to the wrong shape. A portal that was half-built and left usually needs cleaning up before it needs anything new, and the field list is where that starts: a properties audit on a young portal takes an afternoon and saves a year.

What it costs to leave it too long

The records import. The reasons do not.

Names, companies, amounts and dates all move across without much trouble. What does not move is why that deal stalled in the spring, what the client said they needed before they would sign, who made the introduction, which competitor came up on the second call. That context is the only thing a CRM holds which a spreadsheet never did, and it is the one part nobody can backfill.

Leaving it also tends to mean the first CRM arrives in a hurry, usually just after something went wrong, which is a poor condition for making decisions you will live inside for years. Leads going missing between people is the version of this that gets noticed, because it shows up in the number.

Starting with the smallest thing that works

Start with the smallest system that answers the question you actually have. Free tiers are genuinely enough for a long time, and a free CRM with five fields people fill in is worth more than a paid one with fifty they do not.

Add reporting when there is something to report on. Add automation when the manual version is a known repeated step rather than a guess at one. Discounted startup pricing is worth taking if you were going to buy that tier anyway, but a discount on a tier you do not need yet is not a saving, it is a smaller bill for the same unused thing. Whether the plan or the tool is the problem is a separate question and we took it apart in is HubSpot too expensive; which platform to pick at this size is in HubSpot or Salesforce for a smaller team.

When the time does come, the switch is harder than the demo makes it look, for reasons that have nothing to do with the software. We wrote that up in moving off spreadsheets. If the tool you are already sending client email from is doubling as the contact database, there is a separate trap in which path that email goes down. And if you would rather have somebody senior make the call with you, that is the first hour of HubSpot management work.

The question is rarely whether the company is big enough for a CRM. It is whether anything you know about your customers lives anywhere other than in somebody’s memory.

Common questions