Building a Forecast You Can Bank On: Why Most Revenue Predictions Are Fiction — And How to Fix That
Here is a number worth sitting with: 95% of finance and revenue operations leaders express confidence in their ability to plan using their existing forecasts.
Here is the number that exposes the gap: 98% of those same leaders acknowledge they struggle to actually formulate an accurate one.
That is not a rounding error. That is a systemic illusion — and according to Xactly's 2024 State of Sales Forecasting Benchmark Report, it plays out in real dollars every quarter. Four in five sales and finance leaders missed a quarterly forecast in the past year. More than half missed it two or more times. Only 20% of sales organizations achieved forecasts within 5% of their projections.
The question this raises for every CEO is a hard one: is your revenue forecast a strategic tool — or is it a number your organization has simply agreed to believe?
The War Room Nobody Wants to Be In
Several years into leading a high-growth, publicly traded technology company's sales organization, I thought we had forecasting figured out. We had built the kind of process most companies only talk about: weekly individual contributor forecast calls, manager rollups, my own inspection of the largest deals in the pipeline, and a discipline that had held through sixteen consecutive months of hitting our forecast. We were consistently operating at 93 to 110 percent accuracy. The team had internalized the culture of honesty in the numbers.
Then came the quarter that changed how I inspect a deal forever.
In the final month of that quarter, we had fourteen large deals in the pipeline, all tracking well. Our forecast required only three to close. The cushion felt enormous. The confidence felt earned.
All fourteen fell. Ten dropped in the final week of the quarter. Seven of those fell in the final two days.
The CEO and CRO convened a war room. Every executive in the company — the CFO, finance team, customer success, engineering, product development — was pulled into the room to try to rescue a quarter that should have been routine. I attended virtually, managing my own close push with my leaders, moving in and out of the executive room throughout. My CRO was at the center of it. And I knew with complete clarity that they had every right to expect delivery. We had been the foundation the rest of the organization built its forecast on. When that foundation cracked, the entire structure came down.
The shareholder call that followed was a different conversation than the ones before it. Eight thousand employees felt the miss.
76%Our forecast accuracy that month — down from a sustained 93–110% for sixteen consecutive months. The number told the real story before any post-mortem began.
The diagnosis, once the dust settled, was not a process failure in the traditional sense. The process was running on schedule. The cadence was intact. What had failed was the quality of what was inside it. Deal qualification had silently slipped. The team had happy ears — they wanted to believe the deals were real and found confirming evidence rather than challenging assumptions. The market had shifted. Customer behavior had changed. And because the trailing quarter results had been so strong, no one in the inspection chain pushed hard enough on whether the data matched the reality.
A forecast built on optimism is not a forecast. It is a wish that has been formatted as a number.
Why Forecast Failure Is Almost Never What You Think
If you believe your forecast is underperforming because your team lacks effort or urgency, the research suggests you are looking at the wrong problem.
According to Gartner research, fewer than 50% of sales leaders have high confidence in their own forecasting accuracy. A poll conducted by Challenger Inc. in January 2024 found that less than 20% of sales leaders rated their forecast accuracy as genuinely "predictable." These are not junior leaders or poorly resourced teams. These are experienced sales professionals who run sophisticated commercial organizations — and still cannot count on their numbers.
Xactly's 2024 Benchmark Report identified the root cause clearly: the top barrier to accurate forecasting is that reporting systems cannot access historical CRM or performance data, cited by 66% of leaders. A further 60% said they were not entirely sure where their pipeline data was actually coming from. You cannot build an accurate forecast on a foundation you do not understand.
McKinsey's 2020 research on CFO forecasting practices, drawn from a survey of 130 CFOs, found that 40% reported their forecasts were "not particularly accurate" — and that the process consumed far too much time for what it delivered. The best predictor of satisfaction among those CFOs was not the sophistication of their model. It was whether they used a rolling forecast that adjusted inputs as conditions changed.
The pattern is consistent: most forecast failures trace back to input problems, not output problems. Dirty CRM data. Inconsistent stage definitions. Deal qualification that was never challenged. Optimism that was never converted to evidence. Fix the inputs with genuine discipline and the number takes care of itself. Leave the inputs unchallenged and the number will eventually lie to you — always at the worst possible moment.
What Real Forecast Discipline Looks Like
After the quarter where we missed all 14 of our largest deals, I rebuilt my inspection approach with a harder edge. Not a new framework. Not a new tool. A harder set of questions asked earlier, more frequently, and with less tolerance for the language of hope. Four principles drive forecast integrity in practice.
CRM hygiene is not administrative — it is structural.
Gartner research shows that improving CRM data hygiene alone can increase forecast accuracy by up to 30%. Dirty data — missing close dates, undefined next steps, outdated stage assignments — does not just produce bad reports. It produces false confidence in the manager reviewing them. Every field exists for a reason. Treat it that way.
Stage exit criteria, not stage entry.
Most organizations define what a deal looks like when it enters a stage. High-performing organizations define what must be true for it to leave one. Without exit criteria, deals drift forward on optimism rather than evidence. A deal in late stage with no signed legal review, no confirmed executive sponsor, and no active procurement process is not a late-stage deal — it is a wish wearing the clothing of one.
The difference between a pipeline review and a deal inspection.
A pipeline review tells you what is in the funnel. A deal inspection tells you whether it belongs there. The question I rebuilt my weekly calls around after that quarter was simple: does the rep actually know what is happening inside this account right now? Not what they were told six weeks ago. Not what the champion signaled at the last call. What is happening today. Happy ears live in the gap between what a rep wants to believe and what they have actually confirmed.
Accuracy as a tracked metric — not just a feeling.
The most underused discipline in mid-market sales organizations is tracking forecast accuracy at every level: individual contributor, front-line manager, second-line leader. Not as a punitive exercise, but as a continuous improvement loop. When my team was operating between 93 and 110 percent accuracy month over month, we were not guessing — we were measuring. The moment we fell to 76%, that number told a story that the pipeline count alone never could.
High-performing B2B teams target accuracy within 5% of the final number by the last month of a quarter. The SiriusDecisions research published through Forrester found that only 21% of companies achieve 90% or greater forecast accuracy at thirty days out. The Sales Collective's 2025 survey of over 123,000 U.S. sales leaders found that 51% of organizations cited forecast accuracy as the area that improved most after implementing a structured sales process — the single highest-impact result recorded across all categories. Not pipeline volume. Not win rates. Forecast accuracy.
The CEO's Hidden Cost
The war room I described above was not a sales problem. It was a whole-company problem that sales created.
When a foundational team misses — the team the rest of the organization relies on as its revenue anchor — the cascade consumes leadership time that should be pointing forward, not backward. The CEO was not leading growth strategy that week. The CRO was not building the next quarter's plan. The CFO was not evaluating next quarter's investment capacity. Engineering and customer success had their attention pulled from their own priorities. Every hour in that war room was an hour extracted from the future to pay for a present that should never have arrived as a surprise.
McKinsey's CFO research is instructive here: revenue forecasting, done well, drives every consequential decision in a company — capital reallocation, hiring, strategy, production, and investment. When the forecast is reliable, those decisions compound forward. When it is not, organizations react rather than plan. They over-hire against numbers that do not materialize. They under-invest in periods of genuine growth because they do not trust the signal. They make conservative decisions where confidence was warranted, and confident decisions where caution was needed.
Revenue predictability is not a sales operations concept. It is the foundation on which a CEO builds everything else.
The Lesson That Cost the Most to Learn
We recovered our forecast accuracy. We adjusted the inspection discipline. The team got harder on deal qualification and more honest about what happy ears sounds like from the inside. That miss happened inside a three-year run where the organization averaged 156% year-over-year growth — and recovering the accuracy discipline was what allowed that growth to continue with the organization behind it, not scrambling to catch up to it.
But here is what I want every CEO to understand: the conditions for that miss did not appear overnight. They built slowly and silently, one unchallenged deal conversation at a time. The market had changed. Customer behavior had shifted. And because the numbers had been good for so long, the culture of rigorous questioning had quietly softened. The forecast looked right from the outside. The cadence was intact. What was corrupt was the quality of the data being put into the process — and no one caught it until fourteen deals confirmed it simultaneously in the last week of the quarter.
A forecast you can bank on is not built in the final week of the quarter. It is built in the weekly inspection, in the uncomfortable question the rep cannot answer, in the stage exit criteria that forces honesty, and in the culture that rewards truth over optimism at every level of the organization.
The number on the board is only as reliable as what happened in the room before it got there.
Build the Foundation Before You Expect the Results
If your forecast regularly surprises you — in either direction — the problem is not your people's effort. It is the process they are operating inside. And the process starts with what you ask, how often you ask it, and whether the answers have to be earned or allowed to float on hope.
Most of the CEOs I work with are running $10M to $100M businesses. Many of them have excellent products, proven markets, and talented sellers. What they often lack is a forecasting culture — the habit of measuring accuracy, the discipline of deal inspection, and the infrastructure that separates what is real from what is wished.
That infrastructure is buildable. It is not expensive. And the cost of not building it is a war room you never wanted to convene.
Ready to build a forecasting discipline your whole organization can rely on?
About the Author
Andrew Devlin is a fractional VP of Sales and Sales Xceleration Certified Advisor serving CEOs of $10M–$100M B2B companies. With 25+ years of technology sales leadership experience at Cisco, Cloudflare, Splunk, TELUS, and Contentsquare, he has generated over $2 billion in revenue and developed more than 100 sales professionals to President's Club performance. He teaches B2B sales at Okanagan College and is based in Kelowna, BC, serving clients across North America and internationally.
Sources
Xactly Corporation. 2024 State of Sales Forecasting Benchmark Report. July 2024. xactlycorp.com
McKinsey & Company. "Bringing a Real-World Edge to Forecasting." March 13, 2020. mckinsey.com
The Sales Collective. "Sales Process Statistics: USA 2025." May 2025. thesalescollective.com
SiriusDecisions (Forrester). "The Definitive Way to Measure and Grade Sales Forecast Accuracy." forrester.com
Gartner. Sales forecast confidence and accuracy research. 2020.
Gartner. CRM data hygiene and forecast accuracy research.
Challenger Inc. "Sales Forecast Accuracy: Why You're Getting Sales Projections Wrong." February 2024. challengerinc.com
Forecastio. "Sales Forecasting Accuracy and Analysis: A Complete Guide." January 2026. forecastio.ai
McKinsey & Company. AI-driven forecasting error reduction research. mckinsey.com