Sales forecasting
How to Improve Sales Forecast Accuracy by Testing the Evidence
Test the evidence behind commit deals, separate facts from assumptions and coach the weakest dimension before trusting the forecast.
By OnDen · Published · 4 min read
Stage, amount and close date describe what the seller expects. They do not show what the customer has confirmed. A forecast call that checks only those fields can repeat a claim without testing whether the evidence supports it.
Sales forecast improvement starts with the important deals carried as commit. Ask what must be true for each customer to buy, then make the seller show where each answer came from. Keep the CRM as the system of record; use the review to challenge what is recorded there.
Why stage, amount and close date are not evidence
A late stage can tell you that meetings have happened or a proposal has been sent. It cannot, by itself, tell you that finance accepts the business case or that the approval path fits the date. Activity and customer commitment are different things.
Ask what the customer said, did or agreed that supports the forecast category. If the answer is seller confidence, inspect the gap rather than negotiating a more comfortable number. The purpose is a defensible account of the deal, not a more optimistic percentage.
Test the evidence behind each commit deal
Start with the conditions that would let the customer make the decision in the period stated. Ask for the source and date behind each answer, not just a summary of how the deal feels.
- Urgency: what happens if the customer does nothing, and who confirmed the consequence?
- Value: which business-case numbers came from the customer, and has the budget owner validated them?
- Access: who controls the money and the decision, and have their views been heard directly?
- Decision process: what approvals remain, who owns each one and what supports the dates?
- Competition: why might the customer choose a rival, an internal option, delay or doing nothing?
- Next step: what has the customer agreed to do, with which owner, date and outcome?
Separate facts, assumptions and unknowns
Mark a customer-confirmed answer with its source. Keep the seller's interpretation labelled as an assumption. Where nobody knows the answer, record an unknown rather than filling the space with a plausible claim.
This makes discovery specific. An assumed approval date becomes a question for the person who owns the process. An estimated saving becomes a baseline to validate with finance. A gap is useful when it produces a clear request for evidence.
Coach the weakest dimension, not the number
OnDen describes deal readiness as a 0 to 100 diagnostic across six weighted dimensions: customer strategy and urgency; value and business case; stakeholder access and relationships; decision process and close credibility; competitive position and differentiation; and proof, execution and win plan. The breakdown points to where the pursuit needs work.
Readiness is not win probability or a forecast. Use the weakest dimension to agree an action with an owner, a date and the evidence it should produce. If the gap cannot credibly be closed in the forecast period, revisit the category rather than adjusting the diagnostic to fit it.
How AI deal intelligence supports the review
AI deal intelligence works on one live opportunity: customer context, problems, the value case, the buying group and the evidence gaps. OnDen keeps assumptions visible and connects the analysis to red flags, winning moves and owned actions.
That gives the sales leader a more specific starting point for questions. It does not replace customer discovery, the CRM or commercial judgement, and it does not guarantee an outcome. Revisit the same deal to see whether the evidence has changed, not merely whether activity has increased.
Key takeaways
- Stage, amount and close date are claims until customer evidence supports them.
- Test urgency, value, access, decision process, competition and the agreed next step.
- Keep facts, assumptions and unknowns distinct, with sources for confirmed answers.
- Coach the weakest readiness dimension; do not treat a diagnostic as a forecast.
Try this on a deal you are working now
OnDen turns customer research, value discovery and readiness gaps into a practical win plan. You can start with incomplete information.
Analyze a live dealContinue reading
The Forecast Truth Gap
Your rep says 90%. Your CRM agrees. Then you lose the deal. Why sales forecasts inherit human bias, and how an evidence diagnostic challenges them.
Forecast commit criteria
Define forecast Commit on evidence, not seller confidence. What a complex B2B deal must prove — urgency, value, access, process — before it counts.
10 forecast call questions that test the evidence
Ten questions sales leaders can ask in a forecast call to separate customer-confirmed evidence from seller assumptions on a live deal.
Mutual action plan for B2B sales
Build a mutual action plan with customer-owned steps, named owners and agreed dates. Test whether the path to signature is shared or seller-only.