What Is a CRM? A Plain Explanation for Growing Businesses

Every business starts with the relationship data in someone's head. The founder knows which client is annoyed about the last invoice. The first salesperson remembers that the deal at Acme is waiting on their budget cycle in March. It works, and it works well, right up until it doesn't.

It stops working at three predictable moments:

When a second person starts selling. Now there are two heads, and neither knows what the other promised.

When someone leaves. They take the context with them. The handover document covers a fraction of what actually mattered.

When you can't hold it all. Somewhere past a few dozen active relationships, human memory quietly stops being a system and starts being a liability. Follow-ups get missed. Nobody notices, because there's nothing to notice against.

A CRM turns relationship knowledge from something people have into something the company owns.

Strip away the marketing and a CRM is a handful of connected record types:

Contacts — the individual people. Names, roles, contact details, and the history of every interaction with them.

Accounts — the companies those people work for. One account usually has several contacts, and treating them as one relationship rather than four separate ones is most of the point.

Leads — people who might become customers but aren't qualified yet. Keeping them separate from real contacts stops your database from filling with noise.

Deals (or opportunities) — a specific possible sale, with a value, an owner, and a stage. This is the heart of the system.

Activities — calls, emails, meetings, notes, tasks. Each attached to the relevant contact, account, or deal, so the history assembles itself.

Pipeline — all your open deals arranged by stage, so you can see the shape of what's coming rather than just a list.

The connective tissue matters more than any single record. When a contact replies to an email, that reply should appear on their contact record, on their account, and on the open deal — automatically. If it doesn't, you've bought a filing cabinet.

Three common confusions worth clearing up.

A CRM is not a contact list. A spreadsheet of names and emails is a contact list. A CRM tracks state — where each relationship stands, what happens next, and who owns it.

A CRM is not an ERP. A CRM manages revenue and relationships (outward-facing). An ERP manages resources and operations — inventory, procurement, accounting (inward-facing). Different problems entirely. (Longer explanation: CRM vs ERP.)

A CRM is not automatically a sales system. This one's subtle and it's where most CRM disappointment comes from. Many products called CRMs are systems of record — they store what already happened, faithfully, and do nothing else. A CRM earns its cost when it also runs the work: routing incoming leads to the right rep automatically, scoring them so nobody wastes a week on a tyre-kicker, triggering follow-up tasks, generating a quote from a real product catalogue without anyone opening a spreadsheet.

The difference between recording and running is the difference between a CRM your team resents updating and one they'd notice if you took away.

Most CRM implementations don't fail loudly. They fail by decay.

It goes like this. You buy it. Everyone's enthusiastic for three weeks. Then a rep is busy and doesn't log a call. Then two reps are busy. Within a quarter the CRM shows a version of reality that's roughly true but not actually true, and once that's the case, people stop trusting it — which means they stop updating it, which makes it less true. It's a spiral, and it's almost always fatal.

The root cause is nearly always the same: the CRM was set up to serve management reporting rather than to help the person doing the work. If updating it is pure overhead — data entry that benefits someone else's dashboard — it will decay. If it saves the rep time (the email's already there, the quote generates itself, the follow-up task appears without being created), it survives.

So the practical test when evaluating one isn't “does it have the features.” It's: would a busy salesperson at 6pm on a Friday find it faster to use this than to not use it?

Not yet, if:

• One person does all the selling and can hold every open conversation in their head
• Your sales cycle is a single conversation
• You have fewer than a dozen active relationships at a time

A spreadsheet is genuinely fine. Buying software for a problem you don't have is its own kind of waste.

Probably yes, if any of these are true:

• More than one person sells. The moment two people can contradict each other, you need shared state.
• You've lost a deal to a forgotten follow-up. That's not a discipline problem, it's a systems problem, and it will repeat.
• Your forecast is a feeling. If “how's the quarter looking?” is answered with a vibe, you can't plan hiring, cash, or inventory against it.
• Answering “what happened with that account?” requires asking a person. That's the clearest signal. If the knowledge lives in people, it leaves with them.
• Handovers are painful. Rep changes, territory changes, someone goes on leave — every one of these is a data-loss event without a CRM.

Briefly, in priority order:

1. Will people actually use it? Weight this highest. A simple CRM your team maintains beats a feature-heavy one they abandon.

2. Does it run the workflow, or just record it? Ask specifically about lead routing, scoring, and quote generation. If the answer is “you can integrate that,” you're buying two products.

3. Is the pipeline configurable to how you actually sell? Your stages should match your real process, not a vendor's template. (See How to Set Up Sales Pipeline Stages.)

4. Does email connect properly? Two-way sync matters enormously in practice. If reps have to BCC the CRM, they'll forget, and the decay starts.

5. What does reporting give a manager? Win rate, cycle time, source performance, and an honest forecast — without a weekly reconciliation exercise.

6. Can you get your data out? Ask before signing. Export shouldn't be a negotiation.

A CRM is how a business stops depending on individual memory. Buy one when the memory starts failing — and choose the one your team will actually keep alive.