Check five boundaries before setting a forecast snapshot cutoff: report day, data availability, refresh timing, date filters and database consistency. Stripe groups account activity by calendar day in the time zone selected for the report, from 12:00 AM through 11:59 PM.
Key takeaways
- Confirm the report's time zone before you trust any daily total.
- Capture only after the previous day's complete data appears in the Balance summary tab.
- Read a refresh timestamp in its configured time zone, not in UTC.
- Use fixed dates for the final snapshot, and log the calendar dates behind any relative filter.
- Record the database isolation level and handle SQLSTATE 40001 serialization failures.
| Check | Publisher | Documented Boundary |
|---|---|---|
| Report day time boundary | Stripe | Calendar day in selected time zone, 12:00 AM to 11:59 PM source |
| Data availability deadline | Stripe | Complete daily data available by 12:00 PM next day source |
| Refresh time zone logic | Microsoft Learn | Configured local time determines current date, not UTC source |
| Fixed date selection | Metabase | Specific dates always refer to the same date(s) source |
| Shared statement snapshot | PostgreSQL Global Development Group | Statements use same snapshot and do not see each other's effects source |
Forecast snapshot cutoff: the reporting boundaries to check
Verify five distinct boundaries before capturing a quarter-end forecast record. First, confirm the report day boundary where activity is grouped by the selected time zone. Second, check reporting data availability to ensure complete daily records are present. Third, inspect refresh cutoff timing to see which rows are included or excluded during processing. Fourth, review fixed date selection to distinguish specific dates that remain constant from relative dates that move over time. Fifth, examine database snapshot consistency to understand how transaction visibility and ordering guarantees affect the captured extract. Each boundary operates independently within its own system context, and a single timestamp does not align all five across different tools. Record the actual settings used for each boundary in your audit trail, and keep separate notes for each boundary type. Do not assume one tool's rule applies to another. The RevOps owner must verify each item against the specific tool documentation.
Before capture, check report days and available data
Start by confirming the report-day boundary. This window defines which transactions belong to a given reporting date. You must verify that your selected time zone matches the operational context of your sales team before capturing the forecast record. If the time zone differs from where your deals are closed, the daily grouping will shift accordingly.
Next, determine when complete data becomes available. Complete data for each day appears in the Balance summary tab by 12:00 PM the following day in that time zone. For example, activity from September 16, 2025, is available by 12:00 PM on September 17, 2025, in the selected time zone. Noon the following day is the documented deadline for complete data. Capturing the snapshot before this noon boundary risks including incomplete records for the previous day.
Finally, distinguish these UI boundaries from API notification times. For API integrations, Stripe sends webhook events when new reporting data is available, at 12:00 AM UTC and 12:00 PM UTC. These fixed UTC timestamps operate independently of the report's selected local time zone. If your workflow relies on automated database refreshes triggered by webhooks, note that the UTC notification times may differ from the noon deadline in your selected time zone. Record whether your capture method uses the manual report view or an API-driven process, as the timing concepts differ between them. Verify that your chosen cutoff respects both the local data completeness window and any relevant API event schedules.
During capture, inspect refresh timing and date filters
Check the configured time zone before trusting a refresh timestamp. A refresh occurring at 8:00 PM Pacific Time (US and Canada) with a time zone configured determines the current date and time based on Pacific Time, not UTC, which would return the next day Microsoft Learn. This distinction matters when your report uses local business hours rather than server coordinates. If your schedule runs at 4:00 AM every morning, note that the example treats new rows appearing in the data source table during those four hours between midnight and 4:00 AM as rows you don't want to account for Microsoft Learn. Treat this as an example of boundary logic, not a universal requirement for all refresh schedules. Verify whether your specific tool handles late-arriving data similarly or includes it in the next cycle.
Select fixed dates to lock the snapshot range. Specific dates like November 1, 2010, or June 3 – July 12, 2017, always refer to the same dates Metabase. Use these values to keep the calendar range fixed however often the query is re-run. Relative dates such as “the previous 30 days” or “the current week” change as time passes Metabase. Avoid relative ranges for final quarter-close artifacts, because the dates they refer to keep moving. You could set the range as the “Previous 7 days, starting from 2 days ago” if your workflow requires an offset, but record the actual calendar dates used for the snapshot Metabase. Documenting the specific start and end dates prevents ambiguity about which records were included in the forecast calculation. For a close date that changed, see Close date change log: why the date moved.
Illustrative example: a relative date filter
Suppose a filter reads “the previous 30 days” and you capture on quarter-end day. Five days later, the same label covers a different 30-day window, so a re-run would not match the saved snapshot. Store the calendar dates the filter covered on capture day instead. A fixed range avoids this drift.
During database capture, distinguish snapshot guarantees
When extracting data for the quarter-end forecast record, verify how the reporting database handles concurrent transactions and statement visibility. The PostgreSQL documentation describes specific behaviors for shared snapshots within common table expressions. According to PostgreSQL Global Development Group, all statements in a WITH query are executed with the same snapshot, meaning they cannot see one another's effects on target tables. This rule applies specifically to the execution context of those statements within that query structure.
Check whether your extract relies on updates made earlier in the same transaction. PostgreSQL notes that each query sees the effects of previous updates executed within its own transaction, even if they are not yet committed. This behavior provides a stronger guarantee than required by the SQL standard for this isolation level, preventing phenomena described in Table 13.1 except for serialization anomalies. Confirm if your reporting tool captures these uncommitted changes or waits for final commits.
If your environment uses serializable isolation, plan for potential failures during the capture process. PostgreSQL states that serialization failures always return with an SQLSTATE value of '40001'. The documentation emphasizes the need for a generalized way to handle these failures because it is hard to predict which transactions might contribute to read/write dependencies requiring rollback. Ensure your extraction script can detect this specific error code and retry or log the event appropriately.
Understand the ordering guarantees provided by the strictest isolation level. PostgreSQL defines Serializable as the most strict level, where any concurrent execution of a set of Serializable transactions is guaranteed to produce the same effect as running them one at a time in some order. This definition comes from the standard's description of the isolation level. Record whether your database session uses this level when capturing the forecast snapshot.
Verify that your capture method accounts for these distinct behaviors before saving the record. Check if the extract uses WITH queries, relies on uncommitted transaction visibility, or operates under serializable isolation. Note the presence of any SQLSTATE 40001 errors in your logs. Document the specific isolation level and snapshot behavior used for this quarter's data pull. To check the extract itself, see Forecast extract validation: 5 checks for missing rows.
After capture, retain the cutoff reference table
Record the specific timing rule for each tool in a single audit row so the quarter-end extract remains traceable. The table below maps each publisher’s documented behavior to the exact boundary or guarantee you verified during setup. Use this reference to confirm that the report day, data availability, refresh zone, date filter type, and database snapshot all align with the source documentation before archiving the record.
| Publisher | Timing Concept | Documented Rule or Guarantee |
|---|---|---|
| Stripe | Report day time boundary | Activity groups by calendar day in the selected time zone, from 12:00 AM through 11:59 PM. |
| Stripe | Complete daily data availability | Complete data for each day appears in the Balance summary tab by 12:00 PM the following day. |
| Microsoft Learn | Configured refresh time zone | A refresh at 8:00 PM Pacific Time determines current date and time based on Pacific Time, not UTC. |
| Metabase | Specific dates remain fixed | Dates like November 1, 2010, always refer to the same date(s). |
| PostgreSQL Global Development Group | Shared statement snapshot behavior | Statements execute with the same snapshot, so they cannot see one another's effects on target tables. |
Verify that the time zone selected in your reporting tool matches the zone used for the quarter close. Confirm the data availability deadline falls before the forecast meeting starts. Note whether the date filter uses a fixed specific date or a relative range that moves with time. Record the database isolation level and snapshot behavior if you extracted raw SQL rows. This retention step ensures the audit trail reflects the exact configuration state when the number was captured. To reconcile the captured totals, see Bookings reconciliation: 5 checks for quarter close.
Snapshot record fields to log
Log these fields for each quarter-end capture.
| Field | What to record |
|---|---|
| Report time zone | The time zone selected for the report |
| Date range | The start and end calendar dates used |
| Availability boundary | The time by which the day's complete data is available |
| Refresh schedule | The run time and any exclusion window |
| Date filter type | Whether the filter uses fixed or relative dates |
| Capture method | Manual report view or API-driven process |
| Isolation level | The database isolation level, if raw SQL rows were extracted |
Next step today: open the Balance summary tab for one recent day, confirm the time zone shown in your report settings, and write that time zone and the noon deadline into your audit row.
Related guides: AI call scorecard checks: 5 rules before the forecast · Calendar availability: 4 tools for forecast call timing.
FAQ: forecast cutoff dates and reporting consistency
What should a forecast snapshot cutoff checklist include?
It covers five boundaries: report day, data availability, refresh timing, date filters, and database consistency. Verify each item against your specific tool settings before capturing the quarter-end record. This ensures the snapshot reflects the intended time window without mixing different timing concepts.
Does a relative date filter stay fixed when I reopen my Metabase question?
No, relative dates change as time passes. Options like “the previous 30 days” or “the current week” refer to shifting date ranges Metabase. For audit purposes, record the actual date range used for the snapshot rather than just the relative label.
When is complete daily Stripe reporting data available?
By 12:00 PM the following day, in the selected report time zone Stripe. Check this boundary to ensure your extract includes all activity from the previous calendar day before finalizing the forecast number.
Does the Power BI refresh example use my configured time zone or UTC?
It determines the current date and time based on the configured time zone, not UTC Microsoft Learn. Using UTC would return the next day for certain evening refreshes. Preserve the example's context by checking your specific configured local date rather than assuming a universal cutoff. This distinction prevents rows arriving between midnight and the refresh from being incorrectly included or excluded.
What does PostgreSQL SQLSTATE 40001 mean?
It indicates a serialization failure PostgreSQL. Handle it with a retry or rollback procedure, because predicting which transactions need rollback is hard. Implement rollback procedures to prevent serialization anomalies during database extracts.