The First 90 Days Inside a Broken Revenue Engine

When I walk into a company with declining revenue, everyone has already decided what the problem is. The CEO thinks it's the market. The board thinks it's the VP of Sales. The VP of Sales thinks it's the product. The product team thinks it's pricing.

They're all partially right and almost entirely wrong.

The real problem is almost never what anyone says it is in week one. The real problem is what nobody's willing to say out loud — and it usually lives in the gap between what the CRM shows and what actually happened.

Here's what I've learned from doing this more times than I'd like to count.

The Misdiagnosis Is Part of the Problem

The first thing that breaks in a declining revenue engine isn't the pipeline. It's the narrative.

Leadership starts protecting itself before they even realize they're doing it. The VP of Sales starts reporting "activity metrics" because closed-won is uncomfortable. Marketing starts calling brand spend "demand generation." Finance starts reporting bookings instead of recognized revenue. Each team builds a local story that makes them look functional in isolation.

By the time an operator arrives, there are three or four competing versions of reality. None of them are fully fabricated — but none of them are the truth either. The instinct is to call a "data review." Wrong. Data review sessions become theater. Everyone defends their numbers. Nothing changes.

The first thing I actually do is sit in on sales calls. Not to evaluate the salespeople — to hear how customers describe the problem. Customers will tell you exactly what's broken if you let them talk.

What Actually Breaks in Practice

Here's the pattern I keep seeing, across sectors, across deal sizes:

Pipeline distortion. The CRM is full of opportunities that haven't moved in 90 days. Nobody closes them, nobody kills them. They're zombies. The VP of Sales is protecting them because removing them would expose a coverage gap. The forecast is essentially fiction built on those zombies.

Founder dependence that nobody documented. The company grew because the founder could close deals through sheer force of personality and relationship depth. That era ended 18 months ago, but the playbook was never written down. The sales team is executing moves that made sense when the founder was in the room. They don't make sense anymore.

Pricing drift. The list price means nothing. Discounting is happening deal by deal, with no floor and no logic. Two customers in the same segment are paying 40% different prices for the same thing. Neither knows it, but both have heard rumors. Your best customers are now your most negotiable customers because they know they can push you.

Channel conflict nobody will name. Two reps covering the same accounts in slightly different ways because the segmentation was never enforced. Or a partner channel that's cannibalizing direct while the partnership team celebrates "ecosystem growth." The internal politics around fixing this are so bad that three previous leaders tried and gave up.

Quota as fiction. Sales leadership set quotas based on the board's growth expectation, not bottom-up capacity analysis. So the number was always going to get missed. When it got missed, the explanation was market conditions. This happens for two or three years until someone does the math on rep productivity.

What I Do First

Not a strategy session. Not a reorganization. Not a new comp plan.

I run a pipeline autopsy in the first two weeks. Every deal that's been in stage 3 or later for more than 60 days — I want the full history. When was the last real buyer engagement? Not email opens. Not "sent deck." Actual two-way conversation where a decision-maker engaged with a real commercial question.

This typically eliminates 35-50% of what's being called "active pipeline." That's uncomfortable. It's supposed to be. You can't fix a coverage problem if you're pretending it doesn't exist.

Then I do the same thing to the last 12 months of closed-lost. Not to assign blame. To find the actual pattern. Most teams don't do this. They mark deals lost, move on, repeat the mistake. In one engagement, the closed-lost analysis revealed that the same three objections were killing deals in the final stage — and all three were answerable with product functionality that already existed. The sales team just didn't know about it.

That's a training problem. It's also a GTM sequencing problem. It's not a headcount problem.

What to Avoid

Don't hire a new VP of Sales in the first 60 days. Everyone wants to do this. It feels decisive. It's almost always the wrong move.

Here's why: you don't know yet what you need. You hire someone into a broken process and they inherit the same dysfunction, except now they're spending their first 90 days doing onboarding instead of fixing the core problem. Two years from now, you'll be replacing them too.

Don't launch a new CRM or "data transformation" initiative as a response to revenue problems. I've seen this kill more turnarounds than almost anything else. The team spends six months on system migration while the pipeline continues to deteriorate. The new CRM then inherits the same bad data habits from the old one.

Don't let the VP of Sales run the diagnosis. They're too close to it. Not because they're dishonest — because they built the thing that's broken. Their frame is inherently defensive.

What Good Looks Like After the Reset

By month three, you should have:

A pipeline that you can actually defend. Smaller than before, but real. Every deal has a next action, a contact name, and a date. Not "follow up in Q1." An actual meeting or call scheduled.

A forecast methodology that doesn't make you wince. Bottoms-up, weighted by stage probability based on actual close rates — not wishful thinking.

A pricing floor with defined exceptions and an approval path. The exceptions will still happen. At least now they're visible.

A conversation with the founder about what they're actually going to close themselves versus what they need to hand off. This conversation is uncomfortable. Have it anyway.

And most importantly: one clear thing that's getting better every week. Not five things that are sort of moving. One thing that everyone can point to as evidence that the engine is being rebuilt.

Revenue problems are almost always structural, historical, and compounded by narrative protection. The fix is diagnosis first, sequencing second, execution third.

The 90-day clock starts when you walk in the door. Don't spend the first 30 of them in PowerPoint.