Back to all articlesForecasting discipline

Handle Missing Forecast Evidence Without Turning Unknown Into No

Alactic Editorial9 min read

Prepared with AI assistance by Alactic Editorial. Examples identified as hypothetical are illustrative, not customer results.

A sales opportunity has an agreed scope, no verified purchasing authority, and a completion date the buyer has explicitly rejected. Treating all three as incomplete information hides a material difference: one claim is supported, one is unresolved, and one is wrong.

To handle missing information in a sales forecast, classify the evidence against each specific claim as unknown, contradicted, or confirmed. Then make three separate decisions: whether the opportunity belongs in a forecast view, whether its contribution needs different weighting, and what uncertainty must accompany the reported number. A blank field alone should neither earn forecast credit nor trigger a definite failure judgment.

The practical aim is not to fill every gap before forecasting. It is to stop gaps from changing meaning as an opportunity moves from a seller's notes to a management summary.

1. Classify claims, not whole opportunities

Evidence states work best at the claim level. An opportunity can have confirmed scope and unknown authority at the same time. Calling the entire deal confirmed would overstate what is known; calling it unqualified would discard useful evidence.

  • Unknown: There is not enough relevant evidence to support or reject the claim. This includes missing information, an unverified assumption, or conflicting accounts that have not been resolved.
  • Contradicted: Relevant evidence conflicts with the claim. The existing assumption should not continue supporting the forecast unchanged.
  • Confirmed: Relevant evidence supports the specific claim, for the scope and period being assessed. It does not confirm every related claim or guarantee an outcome.

Write the claim so another reviewer can judge it. Purchasing authority identified is more useful than stakeholder status. Buyer can complete the purchase within the forecast period is more useful than timing looks good.

Also separate the claim from the record. A blank authority field means the record lacks an answer. It does not establish whether anyone actually verified authority elsewhere. Conversely, a populated field containing an employee's name does not prove that person can approve the purchase.

2. Separate inclusion, weighting, and uncertainty

These decisions often collapse into one probability adjustment. That makes the forecast hard to explain and easy to manipulate.

Inclusion: which view can legitimately contain the opportunity?

Keep an unresolved opportunity visible in the pipeline. Decide separately whether it qualifies for a supported period estimate, a conditional scenario, or neither. Excluding a deal from the supported current-period estimate does not mean declaring it lost.

Use criteria tied to what the view represents. If the supported estimate requires an evidenced route to purchase, unknown authority may prevent inclusion there. If authority is not an inclusion requirement in your established method, retain the opportunity under that method but explicitly expose the unresolved dependency.

Weighting: what contribution is justified?

Do not invent a precise discount because a fact is missing. Unknown authority does not, by itself, justify any particular probability. If you use existing stage weights, retain their established meaning and document where they do not resolve deal-specific uncertainty.

A contradicted assumption may require removing the affected contribution rather than merely lowering its weight. If the only basis for current-period inclusion has been disproved, a small discount leaves unsupported timing inside the estimate.

Uncertainty: what does the reader need to know?

Report which assumption remains unresolved, what part of the estimate depends on it, and what evidence would change the treatment. When a numerical range is not defensible, a conditional scenario is more honest than an invented interval.

Unknown is an evidence state, not a standard weighting rule. Its forecast effect depends on whether the missing fact is necessary for the amount, timing, or purchasing path being estimated.

3. A filled hypothetical evidence worksheet

Fictional worked example: Every organization, amount, date, and event below is hypothetical. The worksheet illustrates a recommended operating method, not a customer result or a product feature.

Harbor Tools is considering a service package from a fictional supplier. The opportunity is recorded at a hypothetical $40,000, with a June purchase target and June implementation completion. The forecast measures signed orders, not implementation completion. The seller's original timing assumption was that implementation had to finish before the buyer would sign.

The worksheet preserves that distinction so an implementation delay is not automatically treated as a delayed order.

Worksheet header

  • Opportunity: Harbor Tools service package.
  • Review date: June 10, hypothetical planning year.
  • Forecast question: Is there sufficient support to include a signed order in June?
  • Recorded amount: Hypothetical $40,000 quote; signed value not yet confirmed.
  • Current treatment: Visible in pipeline; excluded from the supported June estimate; retained as an unquantified conditional June scenario.
  • Reason: Purchasing authority is unresolved, the implementation completion assumption is contradicted, and the relationship between implementation timing and signing remains unverified.

Evidence row: unknown purchasing authority

  • Claim being tested: The current buyer contact can authorize the purchase or has identified the person who can.
  • Available evidence: The operations contact approved the proposed scope. No evidence identifies purchasing authority or an approval path.
  • State: Unknown. Scope approval does not establish purchasing authority.
  • Inclusion treatment: Keep the opportunity in pipeline. Do not include it in this team's supported June estimate, which requires an evidenced purchasing path.
  • Weighting treatment: Do not assign a new probability based on the gap. Preserve any existing stage-weighted calculation as a separate model view, without presenting it as evidence that authority is resolved.
  • Reported uncertainty: A June order depends on an approval path that has not been verified.
  • Owner and next evidence: Opportunity owner to obtain an attributable account of who authorizes the purchase and what approvals remain.
  • Review trigger: New buyer evidence or the next scheduled forecast review, whichever occurs first.

Evidence row: contradicted completion date

  • Claim being tested: Implementation can be completed by June 30.
  • Available evidence: In a hypothetical June 9 email, the buyer's project lead states that the site will not be available until July.
  • State: Contradicted. The June completion assumption is no longer supportable.
  • Inclusion treatment: Remove June completion from the planning assumptions. Because the forecast measures signed orders, separately verify whether signing must wait for implementation. Do not automatically move the order to July.
  • Weighting treatment: Remove any forecast credit that relied solely on June completion. Do not replace it with an arbitrary lower probability or an unsupported July date.
  • Reported uncertainty: June completion is contradicted; the effect on signing timing is unknown.
  • Owner and next evidence: Opportunity owner to verify whether the buyer can sign before site availability.
  • Review trigger: Evidence about the signing dependency, not merely the passage of time.

Evidence row: confirmed scope

  • Claim being tested: The quoted service package matches the buyer's stated operational requirements.
  • Available evidence: The buyer's operations contact confirmed the listed deliverables in a hypothetical June 8 email, referring to the current scope version.
  • State: Confirmed for operational scope only.
  • Inclusion treatment: Retain the agreed scope as supported pipeline evidence. It does not override the unresolved purchasing path or timing dependency.
  • Weighting treatment: Do not add a separate probability bonus if scope agreement is already reflected in the opportunity's stage.
  • Reported uncertainty: Operational scope is confirmed; approval of price, purchasing authority, and signing timing are not established by this evidence.
  • Owner and next evidence: Opportunity owner to check whether subsequent scope changes invalidate the current quote or confirmation.
  • Review trigger: A scope revision or new buyer objection.

4. Explain the resulting forecast without hiding the deal

The worksheet produces a clear reporting decision: this hypothetical opportunity is not part of the supported June estimate, but it remains visible as conditional pipeline. The exclusion is a statement about forecast support, not a claim that the buyer will refuse to purchase.

A useful forecast annotation would say:

Harbor Tools remains outside the supported June order estimate. Operational scope is confirmed. Purchasing authority is unknown. June implementation completion is contradicted, but whether that prevents June signing is unresolved. Reassess when the purchasing path and signing dependency are evidenced.

If a manager asks for upside, show the opportunity as conditional exposure, with its hypothetical quoted amount clearly distinguished from expected contribution. Do not add the full amount to the supported estimate and then describe the total as evidence-backed.

Likewise, an existing weighted pipeline total can remain useful as a model output. Label it separately from the evidence-reviewed estimate. The difference between those views is information to explain, not necessarily an error to eliminate.

5. Avoid double penalties and borrowed confidence

Unknown authority and uncertain signing timing may overlap. If both arise from the same missing approval path, applying separate numerical penalties can count the same uncertainty twice. Record the relationship before changing any model contribution.

The opposite mistake is borrowed confidence: confirmed scope makes reviewers feel comfortable enough to assume authority, budget, and timing. Keep each confirmation inside its boundaries. A person can know exactly what the business needs without being able to authorize spending.

There is a real trade-off in keeping conditional opportunities separate. A stricter supported estimate may omit deals that eventually close. A broader estimate may include more potential orders but require stronger caveats. Choose the view deliberately and use its criteria consistently; do not change the standard just to preserve a preferred total.

When no reliable weighting method exists, report the supported estimate alongside named conditional opportunities. An unquantified dependency is less convenient than a single number, but more useful than false precision.

6. Recover when the evidence changes or conflicts

Evidence classification needs a review trail, not permanent labels. Record what changed, when the forecast treatment changed, and why.

  • Unknown becomes confirmed: Reassess the affected inclusion requirement. Do not automatically promote the opportunity if another critical dependency remains unresolved.
  • Unknown becomes contradicted: Remove support for the specific claim. Check which amounts or periods depended on it.
  • Confirmed becomes stale: Reopen the claim when its conditions change, such as a new scope or approval process. Do not invent a universal expiry period.
  • Evidence conflicts: Record both accounts and mark the claim unresolved until the conflict is understood. Do not choose the more optimistic account merely because it is newer.

For the hypothetical worksheet, a verified purchasing path would resolve only the authority row. A statement that signing can precede implementation would resolve the timing dependency, but would not restore the rejected June completion date. These are separate changes with separate consequences.

7. Run a focused evidence review tomorrow

Start with the opportunities whose unresolved claims could materially change the next reported period. For each, write the claim, evidence reference, state, affected forecast view, treatment, owner, and review trigger. Use the worksheet above rather than trying to complete every blank in the record.

At the end of the review, count three observable outcomes: opportunities whose treatment changed, unresolved claims with an owner and review trigger, and unsupported assumptions removed from the forecast explanation. Also reconcile any change in the supported estimate to the claims that caused it.

The immediate success measure is traceability: another reviewer should be able to see why an opportunity is included, excluded, or conditional without guessing what a blank field meant.

Sales forecastingEvidence qualityPipeline managementForecast uncertainty

Put the next conversation to work.

Explore Alactic Communications for WhatsApp conversations, campaigns and AI-assisted workflows. Choose the product and plan that fit your team.

Explore Communications