It's week eleven. Your forecast said $4 million. You're heading for $2.8 million. Again.
You're not alone. Gartner found that fewer than half of sales leaders have high confidence in their own forecast. Think about that. Most of the people calling the number don't believe it.
So why does it keep happening? And what actually fixes it?
The forecast isn't the problem. The inputs are.
A forecast is just arithmetic on top of opinions. Each rep looks at each deal and makes a call. Commit. Best case. Pipeline. The spreadsheet adds it up.
The trouble is the question behind every call: "How do you feel about this deal?"
Feelings are optimistic. A buyer who loved the demo feels like a buyer who will sign. A deal that has been "almost there" for six weeks still feels close. And nobody wants to be the rep who kills a deal in front of the team.
So hope goes in. Your forecast comes out. It looks precise. It isn't.
You can't manage a result. You can only manage what creates it.
Revenue is an outcome. By the time you see it, it's history. You can report it, but you can't coach it.
What you can manage are the activities underneath every deal. Those activities are either happening or they're not. No opinion required.
That shift, from asking how a rep feels to checking what has actually happened, is the single biggest lever on forecast accuracy I've seen in 25 years.
The four checks that separate real deals from fiction
Run your top ten deals through these four questions. Be strict. "Sort of" counts as no.
- Is the next meeting booked? A real date and time, in the rep's calendar and the buyer's. No meeting means the deal is drifting, no matter how warm it feels. I call these deals Falldown.
- When did the rep last speak with the buyer? Spoke, not emailed. If the answer is "a couple of weeks ago," the deal is cooling.
- Has the close date held? A close date that has moved more than once is a deal telling you the truth. Listen to it.
- Does the manager know the real status, without opening the CRM? If your manager can't tell you what happens next and when, neither can your forecast.
Most sales leaders who run this exercise find one or two "commit" deals that fail the first check alone. Many find more. Every one of those deals is sitting inside your number right now.
Why better software won't save you
AI forecasting tools are getting sharper every quarter. But they all read the same thing: the data your reps enter. Feed an algorithm hopeful data, and you get a confident wrong answer, faster.
The fix happens upstream, in the Monday pipeline review, where managers decide what counts as real.
What changes when you inspect activities, not opinions
Forecast calls get shorter, because there's less to debate. Weak deals leave the pipeline early, before they cost you a quarter. Your managers stop defending numbers and start coaching the people who make them.
And when your CEO asks, "Are we going to hit it?", you answer with evidence.
The short version
Your forecast is wrong because it's built on feelings. Replace one question, "How do you feel about this deal?", with four checks your managers inspect every week. The number follows.