What to Actually Put in Your CRM

A CRM will let you record almost anything about a customer. Most advice treats that as an invitation. In a business with nobody administering the system, it is closer to a trap: every field you decide to keep is a small tax paid on every record, forever, by whoever has least time to pay it.

So the useful question is not what you could record. It is the minimum that makes a record worth having.

The four things

A CRM record earns its place if it answers four questions. Nearly everything else is optional.

1. Who they are, and how to reach them

Name, company, email, phone. Obvious, and the only part everyone gets right.

One addition worth making: how you came to know them, in plain words rather than a dropdown. "Referred by Sarah at Whitfield" is worth more than a Source field set to "Referral", because in eighteen months the name is the useful part.

2. What happened last time

The single most valuable field in any CRM, and the one most often left empty because it cannot be automated.

Two or three lines: what they wanted, what you said, what was unresolved. Not a transcript — nobody rereads transcripts, and writing them is the kind of overhead that makes people stop logging altogether.

The test for a good note is whether a version of you who remembers nothing could walk into the next conversation without embarrassment. That is not a hypothetical: in four months, that is exactly who reads it.

3. What happens next

One line, stated as an action with an owner. "Send revised quote" or "They will confirm budget."

The distinction that matters is whether the ball is with you or with them, because those are different situations that look identical in a list. A CRM that cannot tell you which of your open items are actually waiting on you is not helping you prioritise.

4. When

A date, attached to the record.

This is the field that makes the difference between a database and a system that works, because it is the only one that reaches out to you. Everything else waits to be looked up; the date arrives.

If you record nothing else, record this. It is also the thing a spreadsheet fundamentally cannot do — the difference is covered here.

What to leave out, and why

The pattern is the same for all of these: they cost a little on every record and pay back only if something downstream consumes them.

Deal value, unless it changes what you do. If you would chase a large opportunity harder than a small one, record it. If every enquiry gets the same treatment, you are inventing numbers that make a pipeline total look authoritative while being fiction. A wrong-looking total is one of the fastest routes to nobody trusting the CRM.

Probability percentages. Almost universally guessed, almost never revisited, and they produce weighted forecasts with an air of precision they have not earned. At small-business volumes, a stage is more honest than a percentage.

Fields you cannot fill from the conversation. Company size, industry, budget authority. If getting the value means research, it will be skipped or guessed. A field that is empty on most records is worse than no field: it makes every record look incomplete.

Anything you will not filter by. This is the cleanest test for a custom field. If the answer only ever gets read while looking at that one record, it belongs in the notes — free, flexible, and nothing to maintain. A field is only worth its cost if you would sort or filter on it.

Two things worth recording that people skip

Why you lost. Not just that a deal was lost — one line on why. Price, timing, went internal, never replied. Four of those in a row is the most actionable information a small business CRM will ever produce, and it costs a sentence.

The human detail. They are on holiday for three weeks in August. They hate phone calls. Their colleague makes the actual decision. This is the kind of thing that makes the next conversation good and that nobody thinks to write down because it feels unbusinesslike.

The note that ages well

Compare these two records of the same call.

"Spoke to client re: proposal. Follow up next week."

"Talked through the proposal. Happy with the approach, worried about the timeline clashing with their year end in October. Wants a version that starts in November. Anya has to sign it off and she is back on the 14th. I said I'd send the revised version by Friday."

The second took forty seconds longer and is still useful in a year. The first is useless within a fortnight, and it is what "keeping the CRM updated" produces when logging is treated as a chore to be discharged rather than a note to a future colleague.

Keep the shape small

The fewer fields a record has, the more likely it is to be complete — and a complete small record beats a partial detailed one every time, because partial records are what teach people the system cannot be trusted.

This is also an argument for products that do not offer much to configure. Zoho Bigin's free plan gives you one pipeline and 500 records, which is a structural nudge toward keeping it simple, and Less Annoying CRM's single tier means there is no more elaborate version sitting there tempting you. Both are covered in the shortlist for businesses with nobody to manage it.

Once the shape is right, keeping it true is a short weekly habit — what maintaining a CRM actually involves.