One of the hardest questions in a sales meeting is a simple one:
"How much will we sell this quarter?"
In many companies the answer still rests on the names of a few big deals, the sales team’s gut feel, and the manager’s experience.
"Things are going well."
"We have three big proposals out."
"The customer will get back to us this week."
These are valuable signals from the field. But they are not a forecast.
A realistic sales forecast requires reading the value, maturity, momentum and historical conversion behavior of pipeline opportunities together.
1. Don’t confuse pipeline value with the forecast
The pipeline total shows the value of every open opportunity that could close at some point in the future.
A forecast, on the other hand, tries to estimate the amount of revenue that will reasonably materialize in a given period.
These two are not the same thing.
For example, having 20 million TL of opportunities in the pipeline does not mean 20 million TL of sales at the end of the quarter.
Every opportunity has a different probability and timing of closing.
That’s why the forecast is a more realistic subset of the pipeline, not the pipeline itself.
2. Derive stage-based probabilities from historical data
In CRM systems, opportunity stages can usually be assigned predefined probability percentages. For example:
20%.
50%.
80%.
The problem: these percentages are not reliable if they are not based on the company’s actual sales behavior.
Suppose only 28% of opportunities at the proposal stage converted to sales over the last two years, while the system assumes a 70% probability for that stage. In this representative calculation, the forecast is artificially inflated.
The right approach is to look at historical conversion rates wherever possible.
Each stage’s real close probability should be measured over time, and the forecast model updated accordingly.
3. Account for opportunity age and your sales cycle
Looking only at which stage an opportunity is in isn’t enough.
For example, if your normal sales cycle is 60 days and an opportunity has been sitting in the same stage for 180 days, the theoretical stage probability may no longer reflect reality.
Aging opportunities are one of the biggest sources of forecast error.
If an opportunity hasn’t moved in a long time, these questions should be asked:
- Is it genuinely still active?
- Does the customer have a next step?
- Has the purchase date been pushed back?
- Why hasn’t the opportunity’s close date changed yet?
A forecast is only as reliable as the hygiene of the pipeline behind it.
4. Don’t separate the rep’s opinion from the data — but don’t rely on it alone either
The rep is the person closest to the customer.
That’s why dismissing field intuition entirely isn’t the right call.
But "the customer will definitely buy" needs to be read alongside the data.
Did the customer attend the last meeting?
Has the technical evaluation been completed?
Is the decision-maker involved in the process?
Has the budget been confirmed?
Is the next step known?
A good forecasting system doesn’t eliminate the rep’s judgment.
It supports it with measurable signals.
5. Track forecast accuracy as a KPI too
A company might be producing a forecast.
But if it isn’t measuring how accurate that forecast is, it can’t tell whether the system is actually improving.
For example, if TL 10 million was forecast last quarter and 6 million TL was realized (a representative example), the important question is not “why did we miss the target?”
First:
Why did the forecast miss by 40%?
— that’s the question to ask.
Were the close dates wrong?
Was the pipeline inflated?
Were the reps too optimistic?
Were more opportunities lost at a particular stage than expected?
This analysis makes the forecasting system more reliable every quarter.
The point of a forecast isn’t to know the future — it’s to reduce uncertainty
No forecast is perfect.
But a good forecast lets a company make healthier decisions.
Stock, cash flow, capacity, hiring, investment and sales targets can all be managed against a more reliable revenue expectation.
That’s why the sign of a successful forecast isn’t saying "our prediction was right" — it’s the margin of error shrinking over time.
You can’t see the future with certainty. But you can reduce uncertainty systematically.