Skip down to the research record

Deal score refresh: 4 clocks before the weekly forecast

Quota Forecast Editorial Team · Published · 11 min read

In most cases, the initial score appears within approximately 36 hours, or up to 48 hours during peak processing periods, according to HubSpot.

Key takeaways

  • Note the automatic refresh trigger for stale data. According to HubSpot, if the relevant properties for a deal have not been updated in two weeks, the score will be updated automatically.
  • Use the last change time to judge freshness. If a score has not changed within the stated window, it may reflect outdated property values.
  • Record the score and timestamp for each open deal discussed on the call. This allows you to reference the specific clock applied by your current system.
  • Confirm whether your workflow triggers an overwrite of manually entered confidence values. This determines if rep-entered numbers persist until the next trigger event.
  • Keep the source page open while reviewing scores. This ensures you compare the displayed number against the documented latency limits for that specific product.

Score clocks by publisher in one table

The table below lists the specific timing and logic rules for four clocks from three publishers. Each row isolates a single vendor's documented behavior so you do not apply one company's hours to another system. Use this reference to verify which clock applies to the software deal you are discussing on today's call.

PublisherWhat you check before the callRule copied from that publisherLimit of this rule
HubSpotInitial score arrival timeHubSpot states the initial score appears within approximately 36 hours, though it may take up to 48 hours during peak processing periods.It does not define a fixed deadline for when a rep must manually update the probability field.
HubSpotChanged score update latencyHubSpot notes changes to existing scores may take up to 6 hours to appear after relevant properties have been updated.It does not specify if a small property edit triggers an immediate recalculation outside this window.
MicrosoftNew opportunity display timeMicrosoft explains the model calculates the score every 24 hours, so the application may take up to 24 hours to display the score for new opportunities.It does not state whether the 0 to 100 range includes decimal values or only whole numbers.
CloseStatus-triggered confidence updateClose describes how the tool automatically updates an Opportunity’s Confidence percentage the moment it moves into a specific status.Whenever the workflow trigger fires, it overwrites whatever Confidence is currently set.

Each entry above reflects only what the cited page documents about its own product. The HubSpot rows distinguish between the wait for a first score and the wait for a refresh after an edit. The Microsoft row focuses strictly on the calculation cycle for new records. The Close row addresses the workflow that modifies confidence based on pipeline stage movement. Do not treat these as interchangeable standards across your entire stack. If your firm uses a mix of these platforms, check the specific tool attached to each open deal before quoting its number. A score that is fresh in one system may be stale in another because their underlying refresh mechanisms differ. Keep this table visible while you review your pipeline to ensure you attribute each timestamp to the correct vendor. This prevents the error of assuming a universal six-hour or twenty-four-hour standard exists across all sales software. Your goal is to confirm the age of the data point you are about to present, not to guess at a generic industry average.

When a new deal receives its first score

Before the weekly call, check how long your tool takes to score a brand-new opportunity. If you see no score on a new deal created two days ago, it is still within the documented window for HubSpot.

Microsoft uses a different cycle. Their model calculates the score every 24 hours Microsoft. Consequently, the application may take up to 24 hours to display the score for new opportunities. A new deal entered yesterday morning might not show a score until today’s forecast call if the calculation cycle has not yet run.

Do not apply one vendor’s hours to every tool. The 36-hour standard and 48-hour peak for HubSpot do not define Microsoft’s 24-hour maximum. Keep these clocks separate when you review your pipeline.

For your next forecast call, identify any open software deals created in the last 48 hours. Note which tool holds each deal. If a HubSpot deal is under 48 hours old, the score may still be pending within the documented window. If a Microsoft deal is under 24 hours old, the score may not have appeared yet. Record the creation time and the current score status for each. This prevents you from treating a missing score as a zero value during the discussion.

Check the specific product page for your exact plan if exceptions apply. The quoted rules cover general behavior but may vary by configuration. Verify whether your instance follows the stated cycles before relying on them. For more on the limits, see Deal score limits: 4 checks before turning scores on.

When a changed deal receives a new score

A revenue leader checking an open software deal before the weekly forecast call must verify whether the displayed score reflects recent property changes. According to HubSpot, changes to existing scores may take up to 6 hours to appear after relevant properties have been updated. This latency applies specifically to the product named in that page’s title, "Predict likelihood to close with deal scores." The six-hour window is a documented maximum for this vendor's system and does not define the refresh time for other sales tools. If you are reviewing a deal where a stage or amount changed yesterday, the score on screen might still reflect the prior state if less than six hours have passed since that update.

The system does not refresh the visible number for every minor fluctuation. For existing deals, the score will be updated if the new score is significantly different (+/- 3%). This threshold determines whether a recalculated value replaces the current one. If the computed difference falls within that range, the interface retains the previous figure. This rule prevents visual noise from small variations that do not materially alter the predicted likelihood to close. You should note this boundary when comparing two dates; a change of less than three percent will not trigger a visible update on the record.

When the new computed score was not significantly different from the prior score (+/-3%), the old value remains in place. This behavior means that a stable deal with minor property tweaks will show the same score across multiple days. The documentation specifies this condition for existing deals where the variance stays under the defined limit. It is a distinct logic path from the initial scoring process covered elsewhere. Your checklist for the call should include checking whether the newly computed score differs from the prior score by more than three percent. If it did not, the score you see is likely unchanged by design, not by error.

Illustrative example: a score change

Suppose a deal shows a stored score of 80. Its properties change, and the newly computed score is 95, a difference of 15 points. That is well above the three-percent threshold, so the record updates to 95. If the newly computed score were only one point higher, the difference would fall within the threshold and the record would keep 80. The numbers are round figures for illustration only.

What a hand-set win percent does to the score

In Pipedrive, the stage probability is the likelihood of closing a deal based on the pipeline stage it’s in. Pipedrive automatically calculates the deal’s value based on this provided probability. This calculation makes sales volume predictions easier by tying the numeric value directly to the stage-based likelihood. When you set a probability manually for an open software deal, the system uses that input to derive the weighted amount. The rule applies specifically to the product and plan described on the cited page. Documented examples on that page are illustrations, not your specific pipeline data or required targets. You can verify how a hand-set percentage affects the total by checking the calculated value against the stage probability definition. For the arithmetic, see Weighted deal amount.

In Close, a workflow can automatically update an Opportunity’s Confidence percentage the moment it moves into a specific status. Close states that once this is set up, the automatic update triggers immediately upon the status change. If a rep has previously typed a confidence value into the field, the workflow overwrites whatever Confidence is currently set whenever the trigger fires. This behavior means the hand-set value does not persist if the automated rule activates. The documentation confirms that the overwrite occurs regardless of the existing manual entry. This rule applies only to Close and the specific product configuration stated on the cited page. Before the forecast call, check whether any workflows are configured to override manual entries for the deals you will discuss. If a workflow is active, the number on screen may reflect the last status change rather than the most recent manual adjustment.

Recording scores for today's call

Record the specific values you will discuss. The opportunity scoring model applies a score of 75 for Opportunity A and 55 for Opportunity B, which serves as the documented example for your entries Microsoft. An opportunity with score of 100 has the highest likelihood of converting into a win deal, providing the upper bound for the scale you are reviewing Microsoft. Write the current number you see on the screen next to the deal name. Note the last time that screen says it changed, using the tool you already have open. Label each number by its source; a score on your screen is pipeline data, even when it matches an example. Keep these notes separate from any hand-set probability you may have entered manually. Do not treat the example numbers as a required target or a benchmark for your specific software deal. Record only what is visible in your system at the moment before the call begins. This keeps your discussion grounded in the actual state of the open deal rather than hypothetical scenarios.

Related guides: AI call scorecard checks: 5 rules before the forecast · Check 6 caption settings before the weekly sales call · Check 8 data rules before the weekly sales call.

FAQ: how old a deal score is before the weekly forecast call

How long can a brand-new deal wait before a score appears?

HubSpot states that in most cases, the initial score appears within approximately 36 hours, though it may take up to 48 hours during peak processing periods. Microsoft Learn notes that the model calculates the score every 24 hours, so the application may take up to 24 hours to display the score for new opportunities. These timeframes are specific to each vendor's system and do not apply universally across all sales tools.

Does a tiny change to a deal update the score?

For existing deals, HubSpot updates the score only if the newly computed score is significantly different from the prior score, defined as a difference of plus or minus 3%. Changes smaller than this margin do not trigger an automatic refresh according to that page’s rule. If relevant properties for a deal have not been updated in two weeks, the score will be updated automatically regardless of whether the change meets the significance threshold.

If I type 0 or 100 as the win percent, does the score stop?

Compare the score on the record after typing 0 or 100 with the score you saw before. An opportunity with a score of 100 has the highest likelihood of converting into a win deal, according to Microsoft Learn. The model assigns a score between 0 and 100 for opportunities based on signals from opportunities and related entities such as contact and account. Verify whether your specific configuration treats extreme values as static flags or dynamic inputs.

Will a status change overwrite a confidence percent a rep typed?

Yes, when a workflow trigger fires, it overwrites whatever Confidence value is currently set, according to Close. Once this is set up, Close will automatically update an Opportunity’s Confidence percentage the moment it moves into a specific status. This behavior replaces any manually entered percentage with the workflow’s assigned value at the moment the trigger activates. If you rely on manual overrides, check whether your active workflows are configured to allow exceptions or strictly enforce the automated change.

Which number is the vendor's example rather than my pipeline?

The scoring model applies a score of 75 for Opportunity A and 55 for Opportunity B, as documented by Microsoft Learn. These figures serve as illustrative examples in the documentation and do not represent benchmarks for your actual deals. Do not cite these specific numbers as expected outcomes for your own pipeline during the forecast call. Use them only to understand how the model differentiates between two distinct scenarios in its documentation.

Sources