The total value of every open opportunity can look reassuring, but it is not a revenue forecast. A pipeline usually mixes new enquiries, active negotiations and deals that have not moved for months. Adding them together without distinction produces a measure of ambition, not a credible view of what may close.
A useful forecast does not require an elaborate CRM. It does require shared definitions, a defined period, an owner for every opportunity and a record that can later be compared with reality. Its purpose is not to predict the future perfectly. It should show a reasonable range of outcomes and expose the gap between the plan and current evidence.
Separate the pipeline from the forecast
The pipeline contains all active opportunities. The forecast is a narrower view of revenue that may close within a particular period based on evidence available today. This is why common CRM systems distinguish deal stages from forecast categories such as pipeline, best case, commit and closed won.
For each opportunity, record at least its amount, current stage, expected close date, last meaningful contact, agreed next step and accountable owner. A deal with no next step, or a close date that moves every week, should not carry the same confidence as an active negotiation with a documented decision process.
Make every stage an observable claim
Labels such as “negotiation” or “almost done” invite inconsistent interpretation. Give each stage an entry or exit rule. “Proposal reviewed”, for example, might mean the buyer has discussed price and scope, the relevant decision-maker has participated, outstanding questions are recorded and a date for the next decision is agreed.
Do not assign probability only from a salesperson’s confidence. Start with your own history: of the opportunities that entered a stage, how many eventually closed? Compare like with like—similar offer types, deal sizes and sales cycles. If the sample is small, treat the result as provisional and use a range instead of pretending that a precise percentage is reliable.
Illustrative example: a weighted forecast
Illustrative example: three opportunities could close this month. A €10,000 proposal is in a stage with a 45% historical win rate, a €6,000 negotiation is in a 70% stage, and an early €3,000 opportunity is in a 20% stage. The weighted estimate is €4,500 + €4,200 + €600, or €9,300. It is neither a promise nor an invoice; it is an expected value derived from stated assumptions.
Keep three scenarios beside the weighted number. Commit should include only deals supported by strong evidence. A realistic scenario combines revenue already won with the calibrated share of active opportunities. A broader upside view shows potential rather than expectation. One unexplained number is easily mistaken for certainty.
Use time and slippage as evidence
Two deals in the same stage can have very different quality. One may have entered three days ago with the next meeting booked; another may have had no response for six weeks. Track time in stage and the number of close-date changes. An old opportunity need not be deleted automatically, but it deserves lower confidence or an explicit decision about whether it is still active.
Calibrate the system every week
In a short weekly review, do not ask only whether a deal will close. Check the evidence for its stage, the next step, the timing and the main obstacle. At the end of the period, compare the saved forecast with the outcome. Record both the size and direction of error: was the team consistently optimistic, too cautious, or mainly wrong about timing? Field research on organisational sales forecasts has found recurring bias and inefficiency in judgmental estimates, which is why preserving the original forecast matters.
A 30-day implementation checklist
- Agree four to six stages and write an observable criterion for each one.
- Clean the open pipeline by adding amount, owner, next step and expected close date.
- Use completed opportunities to estimate stage conversion, separating materially different sales motions.
- Build commit, realistic and upside scenarios for one defined period.
- Save the first forecast and compare it with actual outcomes after the period without rewriting the original numbers.
The goal is not to produce perfect probabilities in one day. It is to replace an unstructured opinion with a repeatable decision process. When a company knows what evidence advances a deal and where its forecast is repeatedly wrong, the forecast becomes useful for capacity, cash planning and sales priorities.
Put these ideas to work in your business.
Tell us about your situation. Together, we can choose a sensible first step and agree the scope of support.