What They Told The Board
“We need better data to make better decisions.”
This is the rallying cry of every SaaS revenue leader who doesn’t know what’s actually broken. And it sounds brilliant in a board meeting. Who could argue against better data? So the board nodded, approved the budget, and the VP of Sales went on a shopping spree that would make a procurement team weep.
Gong for call intelligence. Clari for forecasting. Outreach for sequences. 6sense for intent data. ZoomInfo for enrichment. LeanData for routing. Salesforce underneath it all, configured by a contractor who’d left eighteen months ago and documented nothing.
Every tool had a champion. Every champion had a dashboard. Every dashboard told a different story. And the one story nobody could tell with any confidence was the only one that mattered: which deals are going to close this quarter and why?
What I Actually Saw
Let me describe the first Monday forecast call I sat in on.
The VP of Sales opened Clari. Pulled up the forecast view. It showed $4.2M in commit, $2.8M in best case, and $6.1M in pipeline. The VP asked each rep to “walk their top deals.” One rep pulled up Gong to reference a call recording. Another cited an Outreach sequence engagement rate. A third mentioned a 6sense intent signal that showed the prospect was “in-market.”
Forty-five minutes of data theater. Not once did anyone reference what the buyer had actually agreed to. Not once did anyone ask whether the economic buyer had confirmed the cost of doing nothing. Not once did anyone question whether a deal sitting in Stage 3 for six weeks with no buyer-side movement was a real deal or a pleasant fiction decorated with software.
They had data everywhere and understanding nowhere.
I spent the next week pulling actual numbers. Here’s what the tools wouldn’t tell them:
Of the $4.2M in commit, $2.6M had no documented access to an economic buyer. Reps had talked to users, champions, sometimes technical evaluators. People who loved the product but couldn’t sign a check. The CRM said Stage 4. Reality said Stage 1.
The average Outreach sequence had a 23% reply rate. The team celebrated this. What they didn’t track was that 80% of those replies were some version of “not now” or “send me information” – which is prospect-speak for “I’m being polite while I delete your next email.” The sequences were generating activity metrics, not pipeline.
6sense showed 340 accounts “in-market.” The team was treating intent data like a lead list. They were blasting sequences at every account that crossed a score threshold, regardless of whether the company matched their ICP, had a problem they could solve, or was experiencing the kind of pain that creates urgency. Intent data tells you someone is researching a category. It doesn’t tell you they’ll buy your product. That distinction was completely lost.
Gong had recorded 1,200+ calls. Nobody was reviewing them systematically. The “call intelligence” was being used to settle disputes about what a prospect said in a meeting – essentially an expensive tape recorder. The coaching insights, talk-ratio analysis, and competitive intelligence sitting in that platform were untouched. Three hundred dollars a seat for a replay button.
The Diagnosis
The tech stack wasn’t the problem. It was the symptom of a much older problem that nobody wanted to name.
This SaaS company had no shared definition of a qualified deal. Not a vague one. Not a bad one. None at all. Ask five reps what qualified a deal to be in pipeline and you’d get five different answers. One said “they took a second meeting.” One said “they requested pricing.” One said “my gut says it’s real.” I’m not making that up. A $28M ARR SaaS company was running qualification on gut feel and hoping the tools would sort it out.
When you have no qualification framework, no shared language for what makes a deal real, and no mechanism to test whether buyer commitment is progressing or decaying – you do what every overwhelmed SaaS team does. You buy software. You buy software because software feels like progress. Software has onboarding calls and implementation timelines and QBRs. It feels like you’re building something. What you’re actually doing is layering measurement on top of a process that doesn’t exist.
Gong can’t coach your reps if nobody has defined what a good call looks like. It can tell you the talk ratio was 70/30 seller. It can flag that the rep didn’t ask about next steps. But it can’t tell you that the rep spent forty minutes on a features demo when they should have spent forty minutes quantifying the cost of inaction. That requires a methodology. They didn’t have one.
Clari can’t forecast if the stage definitions are meaningless. Garbage stages in, garbage forecast out. Clari was faithfully aggregating CRM data that bore no relationship to deal reality. The AI-powered forecast was predicting outcomes based on a pipeline where stage progression meant “the seller did something” rather than “the buyer committed to something.” It was a sophisticated machine learning model trained on lies.
6sense can’t prioritize if you don’t know your own ICP. Their ideal customer profile was “mid-market SaaS companies.” That described about forty thousand companies. Intent data against a profile that broad is just noise with a subscription fee.
The total cost of this stack was $380K a year. The total value was whatever it cost to generate a really impressive set of dashboards for board meetings. As an actual revenue-generating system, it was producing nothing that a whiteboard and honest conversation couldn’t have produced for free.
What We Installed
The instinct was to rip everything out. I didn’t do that. These were good tools. Every single one of them. They were just operating in a vacuum where none of them could do what they were designed to do.
We built the operating system first. Before touching a single tool, we defined qualification criteria that were specific, measurable, and binary. A deal was either qualified or it wasn’t. There was no “kind of.” Qualification required documented evidence across four dimensions – not self-reported rep confidence, not tool-generated scores, actual evidence of buyer commitment that could be reviewed by a manager and verified.
We rewired the CRM to reflect buyer agreements, not seller activities. Stage definitions got rewritten. Stage 2 no longer meant “demo completed.” It meant the buyer had confirmed the scope of their problem and granted access to the person who could authorize a purchase. If the rep couldn’t document that, the deal stayed in Stage 1 regardless of how many demos they’d done. This immediately broke the forecasting model Clari was running – which was the point. You can’t fix a forecast built on bad data by improving the forecast model. You fix the data.
Then we turned the tools back on – but with purpose. Gong became a coaching platform, not a recording platform. We built scorecards aligned to the new methodology. Were reps opening with diagnosis or with demos? Were they quantifying consequences or pitching features? Were they asking for access to power or accepting whatever contact they’d been given? The recordings suddenly had something to be measured against.
Outreach sequences got gutted. We replaced the spray-and-pray cadences with targeted, research-backed sequences that led with a hypothesis about the prospect’s problem. Volume dropped by 60%. Response quality increased by 400%. Actual meetings booked went up.
6sense got scoped to a tight ICP – specific company size, specific tech stack signals, specific hiring patterns that indicated the pain this company actually solved. The “in-market” account list went from 340 to 40. Those 40 were real.
The Numbers, 90 Days Later
Tech stack spend went from $380K to $290K. We cut two tools entirely – one enrichment platform that duplicated ZoomInfo and a “conversation intelligence” add-on that nobody used. The savings weren’t the point, but they made the CFO happy.
Forecast accuracy went from 52% to 79%. This was the number that changed everything downstream. When the forecast is real, the board trusts the revenue leader. When the board trusts the revenue leader, they stop micromanaging pipeline reviews. When they stop micromanaging, the VP of Sales has time to actually lead.
Pipeline shrank by 35%. Again – not a problem. The pipeline that disappeared was never going to close. It was sitting in the CRM making everyone feel better about a quarter that was going to miss regardless. Removing it was like draining an infection. Unpleasant to look at, essential for healing.
Win rate on qualified pipeline went from 14% to 29%. Not because the reps got better overnight. Because they stopped wasting cycles on deals that were never real and started spending time on deals where the buyer had actual urgency, actual authority, and an actual cost of doing nothing.
The rep who had the “gut feel” qualification standard? He either adapted or he didn’t. He didn’t. He was managed out in month two. The four remaining reps outperformed the original five by month four.
The Principle
SaaS companies buy tools to avoid making decisions. A new platform feels like progress. An implementation timeline feels like a plan. A dashboard feels like visibility. None of it is any of those things if the underlying process is broken.
The question isn’t whether your tech stack is good enough. It almost certainly is. The question is whether you’ve given those tools something real to work with – a qualification framework that means something, stage definitions tied to buyer behavior, and a coaching methodology that tells your reps what “good” looks like before you start measuring how far they are from it.
Every SaaS tool in the revenue stack is an amplifier. Plug an amplifier into a signal and you get music. Plug an amplifier into noise and you get louder noise. Most SaaS revenue teams are spending six figures a year on louder noise and wondering why the forecast still misses.
Fix the signal. The tools will take care of themselves.
I help B2B companies fix the revenue systems that legacy methodologies broke. If something in this post made you uncomfortable, it was probably the part that's true. Stop the bleeding.
