5 signs your CRM needs an audit

Your CRM needs an audit when it has stopped being the place the number comes from. Five signs give it away: each team reports a different figure, opportunities move forward without evidence, the forecast lives in a separate spreadsheet, new joiners learn by asking, and nobody knows why a deal was lost. Switching tools fixes none of them.

1. Everyone has their own version of the truth

If marketing, sales and finance report different numbers for the same metric, you don't have a tool problem. You have a data model problem.

What you'll see. Two people pull the report for the same month and get two different numbers. The meeting is spent reconciling instead of deciding. Often the same company is in the CRM several times, and the count of active accounts is inflated without anyone knowing by how much.

Why it happens. Opportunity, customer and closed-won don't mean the same thing to everyone. Quite simply, nobody wrote the definition down. And without a unique key per company, duplicates aren't an oversight. They're inevitable.

What it breaks. As long as the definitions don't match, conversion, loss-reason breakdowns and churn are calculated on different bases, and any analysis on top — including the one you ask AI to do — inherits the error.

2. Opportunities move to the next stage without the information they need

When an opportunity can change stage without evidence that the buyer has done something, the forecast is built on gaps.

What you'll see. An opportunity moves stage because the rep says so, with nothing to back it up. Required fields get filled with whatever lets you save and move on. There are opportunities that have sat in the same stage for months and nobody has closed them as lost.

Why it happens. Almost always, because the stages describe what the seller does ("demo done", "proposal sent"), not what the buyer has done. And since optimism always pulls the same way, the pipeline isn't wrong at random. It's wrong on the high side.

What it breaks. A field filled in just to be able to save is worse than an empty one, because the gap is visible and the false value isn't. Stage probability, close date and the forecast are built on top of it: a funnel that measures activity and presents it as progress.

3. The real reporting lives in a parallel spreadsheet

If someone has to export data and cross-reference it by hand before every pipeline review, the CRM has already stopped being the source of truth.

What you'll see. The real forecast lives in a spreadsheet that only the person who built it understands. The CRM gets updated on Friday afternoon, for reporting rather than for working.

Why it happens. The spreadsheet exists because it answers a question the system doesn't. It's proof that the stages or the definitions don't support the decision that has to be made.

What it breaks. It creates a second truth that drifts away from the first and concentrates the knowledge in a single person. The forecast meeting argues about which number is right, not about what to do, and when that person leaves, the forecast leaves with them. The way out isn't to ban the spreadsheet. It's to bring its question into the system and, in the meantime, document it.

4. New reps take months to become productive

When onboarding someone means having them shadow your existing reps, the problem isn't the new person. It's that the process isn't written down.

What you'll see. New joiners learn by watching the founder sell and asking on Slack. The ICP is only partly defined and isn't written down, and each new person interprets it their own way.

Why it happens. How you sell isn't written down anywhere. The decisions already made — who not to sell to, what to ask in the first meeting, how to answer an objection, what gets recorded and when — live in the memory of one or two people.

What it breaks. Each hire creates their own version of the process, and the versions diverge without anyone noticing, because there's no document to check them against. And the knowledge is lost as soon as someone leaves.

5. Nobody knows why the deals you lost were lost

If you can't reconstruct why a deal was lost, you can't learn from it or get the next one right either.

What you'll see. The loss reason is written as free text, if it's written at all, and nobody categorises it afterwards. The conversations that moved the deal happened outside the CRM. Stage history isn't kept, so past pipeline gets rebuilt from today's snapshot.

Why it happens. Logging takes effort and gives nothing back to the person doing it. If the loss reason is free text, it can't be added up; if the stage history isn't saved, nobody finds out until the day they try to use it.

What it breaks. The real reason for a loss expires within days: memory rewrites it towards price, which is the comfortable explanation for everyone. Without loss reasons that can be grouped, there's no way to know whether you're losing to a competitor or to a buyer who decides to do nothing, and each needs a different fix.

What a CRM audit really reviews

A useful CRM audit reviews the process and the data model behind the tool, not just its configuration: counting empty fields and duplicates finds symptoms, not their cause. This is what to look at:

• The data model. What an account, a contact, an opportunity and a customer are; whether there is a unique key per company; whether each data point arrives with its source.
• The stages and their exit criteria. Whether they describe buyer states, and whether two different people would give the same answer about which stage a deal is in.
• The mandatory fields. Which ones are used in a specific decision and which ones are filled in just to be able to save.
• The history. Whether the created date, the date each stage was entered and the close date are kept, along with the loss reason.
• The automations and integrations. What each one does, who owns it and which ones fail silently.
• How it's actually used. When it's updated, what is done outside it and which parallel spreadsheets exist.

A technical review alone won't find the gap between what the founder, the prospecting team and the person who owns reporting each tell you. So a serious audit includes interviews, and that gap is usually the most useful finding.

And then there's the order. The five signs aren't independent: badly defined stages produce fields filled in out of obligation, and that produces the parallel spreadsheet. Fixing the reporting before the stages means measuring a pipeline whose stages mean nothing.

Where to start

Start by counting how many of these signs you recognise. If it's several, diagnose your data before you buy a tool or a project.

If you want a free first check, the online self-assessment (in Spanish) gathers these and other statements about how your company sells: you tick the ones you recognise and it tells you where they come from and where to start. It works from your own perception, so it can't see what isn't visible from the inside.

The RevOps & GTM Diagnostic applies the same method to your CRM. We review the CRM, the data model and the automations as they're actually used, interview the people who sell, prospect and report, go through the seventeen components of the sales system and deliver the confirmed symptoms with evidence, where each one comes from, what it breaks and the priority order, with the price of each piece. One rule is explicit: if the process is broken, you don't start by pushing more leads into the top of the funnel. It takes three weeks and costs €1,500 plus VAT at a fixed price, credited in full against the next project. If you decide to fix it in-house, the plan is yours.

What comes next depends on what it finds; often it's a CRM, data and forecasting project within the RevOps & GTM Projects. For background on the discipline, read what RevOps is and why you need it.

Frequently asked questions

How long does a CRM audit take?

Our RevOps & GTM Diagnostic takes three weeks and has a fixed price. It covers the whole sales process and interviews with the team, because many symptoms that show up in the CRM originate outside it.

What if the audit concludes that nothing needs doing?

We'll tell you anyway. The report isn't conditional on selling the next project, and that's precisely why we charge for it.

Is it better to change CRM or fix the one we have?

Almost always, fix the process first. A new CRM on top of badly defined stages reproduces the problem. Switching only makes sense when your current tool blocks something specific, and once the migration cost is on the table.