What the letters actually stand for
Customer Relationship Management. The name is unhelpful, because it describes an activity that existed long before software did. Every business that sold to the same person twice was managing a customer relationship. The software is only the filing system.
So the useful question is not *what is a CRM* but *what does the filing system have to do*. There are three jobs, and any tool calling itself a CRM does all three.
The three jobs
1. Hold a record of every person and company
Names, roles, email addresses, phone numbers, which company they work for, and which of your people owns the relationship. This sounds trivially easy and is where most systems fail, because the record has to stay true as people change jobs and companies get acquired.
2. Hold the history
Every message sent, every call logged, every note written, every meeting booked. The point is not archival. The point is that the next person to talk to that customer — including you, in four months — can see what was already said without asking anyone.
3. Hold the state of each deal
Which stage it is at, what it is worth, when it is expected to close, and what has to happen next. This is what turns a list of contacts into a pipeline, and a pipeline into a forecast.
If a tool does the first two but not the third, it is an address book. If it does the third but not the first two, it is a spreadsheet with colours.
What a CRM replaces
Usually four things at once, which is why the switch feels bigger than it is:
- A spreadsheet holding the deal list, owned by whoever built it.
- Individual inboxes holding the actual conversations, visible only to the person who sent them.
- A notes app or a notebook holding what was said on calls.
- Somebody's memory holding the rest — which is the one that leaves when they do.
The fourth is the expensive one. A team can rebuild a spreadsheet in an afternoon. It cannot rebuild the reason a customer went quiet in March.
When you actually need one
Not on day one. A founder selling to their first ten customers genuinely does know all of them, and a spreadsheet is faster than any CRM for ten rows. The threshold is not a number of contacts — it is the first time one of these is true:
- 1More than one person needs to know what happened with the same customer.
- 2You have missed a follow-up you meant to make, and only noticed later.
- 3Someone asked what is in the pipeline and the honest answer took a day to assemble.
- 4A person left and took context with them.
Any one of those means the knowledge is already distributed and unrecorded. That is the problem a CRM solves.
What to ignore while choosing one
Feature lists, mostly. Every CRM on the market stores contacts and deals; nobody loses a deal because their CRM could not hold a phone number. The differences that matter in daily use are narrower:
- Does it fit how your data is actually shaped? If your columns cannot change type when the way you work changes, you will end up with a
Notesfield holding six different kinds of information. - Where does email live? If sending is a separate product on a separate bill, the history in your CRM will always be partial, because nobody copies messages into a second tool by hand.
- How much typing does it demand? A CRM that only stays accurate through discipline will not stay accurate. The record has to update as a side effect of doing the work.
The honest summary
A CRM is not a growth strategy and it will not make anyone sell better. It removes a specific, boring failure: the deal that dies because nobody remembered to send the second email. That failure is common enough, and cheap enough to fix, that the tool pays for itself long before anything more ambitious happens.
If you are weighing this against the spreadsheet you already have, when a spreadsheet stops being enough works through the specific point where one stops scaling.
