METHOD · WORKED EXAMPLE · UPDATED
How do you keep delivery-date history when the next report overwrites it?
Save each report separately, label it with its snapshot date, and match the same order lines across reports. Compare the planned ETA values in chronological order. Keep the source file and row for every observation, and leave missing records visibly unknown.
This method is for buyers who already retain periodic CSV or Excel reports. It does not recover an overwritten file or establish when goods actually arrived.
A three-report example
These are illustrative values, not customer results. All dates are YYYY-MM-DD.
| Report saved | Order / line | Planned ETA | Observed movement |
|---|---|---|---|
| 2026-09-01 | A-100 / 001 | 2026-09-20 | First observation |
| 2026-09-08 | A-100 / 001 | 2026-09-27 | +7 calendar days |
| 2026-09-15 | A-100 / 001 | 2026-10-04 | +7 calendar days |
Result: 2 observed transitions and a net shift of +14 calendar days. The first observed ETA is 20 September, not necessarily the supplier's original promise. The exact dates on which someone edited the ETA are unknown.
Run this example in Date Trail
The minimum input
- At least two retained snapshots with distinct report dates.
- A stable identity: for example purchase order + line ID. For split shipments, include a stable shipment ID if the source has one.
- A planned ETA column. Map renamed columns explicitly; choose the same identity components in the same order.
- Headers in the first row. Use ISO dates or explicitly choose day/month/year or month/day/year.
A SKU alone may not identify an order line. Duplicate identities stop the review so that unrelated lines are not silently merged. Leading zeroes matter: preserve identifiers as text when exporting from Excel.
Which date belongs on each report?
The report date is when the snapshot was saved, not the planned delivery date in its rows. Date Trail compares 2–20 reports and accepts one snapshot per calendar date. If you saved a morning and an afternoon copy on the same day, choose the one that represents your review cut-off and retain the other outside this comparison. Do not invent different dates to make both files fit: that would misstate when the observations were made.
Reports are sorted by their supplied report dates, so upload order does not determine the history. Two files carrying the same report date stop the review. A planned ETA remains a separate field and may be the same in several reports.
What is calculated?
- Observed change count: how often consecutive available ETA observations differ.
- Observed postponement count: how many of those transitions move the ETA later.
- Net calendar-day shift: last observed ETA minus first observed ETA. Positive is later; negative is earlier. This is not working-day delay.
- Absent / blank observations: a line absent from a report or present without an ETA remains unknown. An observed transition across such a gap carries a gap flag.
Missing data example
B-200 / 001 has ETA 18 September in the first report, is absent from the second, and has ETA 25 September in the third. The output shows one observed transition, +7 days net, and one absent report. It does not infer delivery, cancellation, or the number of intervening edits.
Where this approach fits—and where it doesn't
Use this for a retrospective review of files you already have, before a supplier follow-up or internal status meeting. If your purchasing system already records and exports the required history, start with that existing record. Date Trail does not claim to replace procurement platforms.
Excel or Power Query can also maintain history when snapshots are retained and the comparison is configured. A dedicated procurement system may offer broader scorecards, collaboration and live integrations. Date Trail focuses on a small file-based workflow without changing the source system.
Why this workflow exists
A public Excel question from December 2024 asks how to track changing delivery dates across repeated reports. A Microsoft community discussion from May 2022 describes daily order and ETA updates and the need to preserve earlier values. These are problem examples, not endorsements or proof of demand for this product.
Files and privacy
CSV/TSV and XLSX parsing runs locally. No AI service processes the reports. Excel formula results are read as stored; formulas are not recalculated. Export confirmed values if your workbook relies on recalculation or macros.
Usage events exclude file contents, order IDs and filenames. Feedback is sent only when you submit it. Privacy details.
Your workflow may need a different answer
If you need promised-date baselines, actual receipts, supplier groups or split-shipment handling, describe the exact task in our feedback form. Please leave out confidential order data. We review requests before promising development.