Key Takeaways
- Per SiriusDecisions, 79% of sales organizations miss their forecast by more than 10% — most forecasting, even at established companies, is closer to hoping than predicting.
- A defensible forecast is built from three known inputs: current qualified pipeline, historical stage-to-stage conversion rates, and active channel output — not a target worked backward into a plan.
- Forecasting in a range, not a single number, is more honest and more useful — a single number implies false precision that historical conversion data rarely supports.
- The goal isn't perfect accuracy. It's a forecast built from evidence instead of hope, which is what actually lets you act on it.
Why Most B2B Forecasts Fail
Per a SiriusDecisions study, 79% of sales organizations miss their forecast by more than 10%. That's not a fringe statistic about badly-run businesses — it's the norm, which suggests the way most businesses forecast is structurally unreliable, not just occasionally unlucky.
The common failure pattern: start with a revenue target (often set by ambition, not evidence), then work backward into activity levels that would theoretically produce it. That's planning, not forecasting — it tells you what would need to be true, not what's actually likely given what's currently happening.
The Three Inputs a Real Forecast Uses
A defensible forecast runs forward from evidence, not backward from a target:
Current qualified pipeline. Not total leads — qualified opportunities specifically, since qualified conversations vs. booked calls covers exactly why the unqualified version of this number is misleading.
Historical stage-to-stage conversion rates. What percentage of qualified opportunities have historically become signed clients, and how long has that typically taken. From interest to signed client covers the stages this conversion data should be tracked against.
Active channel output. What each currently-running channel has actually been producing recently, not what it produced during an unusually good month being treated as the new baseline.
Multiply these together with a realistic time lag, and you get a forecast built from what's actually happening — not from what you'd like to happen.
Why a Range Beats a Single Number
A single-number forecast implies a precision that conversion data rarely supports — real pipelines have variance, and presenting one number hides that variance rather than accounting for it. A range (built from historical best-case and worst-case conversion rates applied to current pipeline) is both more honest and more useful, because it tells you what to actually expect, including how much things could reasonably swing.
Per Gartner, fewer than 50% of sales leaders have high confidence in their own forecasts — a range, built from real historical data rather than a single confident-sounding number, is part of what closes that confidence gap, because it's honest about uncertainty instead of hiding it.
What Forecasting Well Actually Enables
The point of a real forecast isn't a satisfying number to report upward — it's being able to act on it. A forecast that shows pipeline trending below what's needed next quarter is useful specifically because it's early enough to do something about: add capacity to a channel, revisit qualification criteria, or have an honest conversation about a hiring decision, months before the shortfall actually shows up in signed revenue.
The true cost of an unpredictable pipeline covers what happens to businesses that only find out about a shortfall once it's already too late to act on it. Predictable growth metrics covers the specific rates and ratios worth tracking alongside this forecast, once it exists.
Frequently Asked Questions
How often should a B2B forecast be updated? Monthly is a reasonable default for most established service businesses — frequent enough to catch a meaningful shift in pipeline or conversion trends, infrequent enough to avoid reacting to normal week-to-week noise.
What if there isn't enough historical data yet to build reliable conversion rates? Start with a wider, more honestly uncertain range, and narrow it as real data accumulates. An early-stage forecast being less precise is fine — what matters is that it's built from actual pipeline data rather than pure aspiration, even if the range is wide.
Should marketing and sales forecast separately or together? Together, ideally against the same pipeline data — marketing and sales as one engine covers why forecasting split across two disconnected functions tends to produce two different, hard-to-reconcile numbers instead of one defensible one. Predictable growth covers how this forecasting discipline fits the rest of the system.
Related reading
Predictable Growth: Why It Comes From Systems, Not Better Campaigns
Predictable growth isn't a bigger marketing budget or a better sales pitch — it's a marketing system and a sales system operating as one engine. Here's what that actually looks like.
Predictable Growth Doesn't Come From Better Campaigns — It Comes From Better Systems
A better campaign is still a campaign — bounded, temporary, and destined to end. Here's why no amount of campaign optimization can produce what only a system can.
The True Cost of an Unpredictable Pipeline
The cost of an unpredictable pipeline isn't just the bad quarters. It's every decision made worse by not knowing, in advance, that a bad quarter was coming.