An order export is incomplete even though the file opens successfully

Readability is not completeness

A CSV that opens cleanly proves only that the file is well-formed. Completeness is a claim about coverage: every order in the intended scope appears at least once. Those are independent properties, and the first says nothing about the second. Before any comparison, write down what the export was supposed to cover: the date range, the order statuses, the channel or location, and the purpose the file will serve.

Preserve the file exactly as received, along with its export settings and the time it was taken. Work from a copy. If the file is later found incomplete, the original settings are the evidence that explains why, and an edited file destroys them. Note also when the export was taken relative to the work it will drive; an export from this morning cannot contain this afternoon's orders, and that is a timing gap, not a platform failure.

Check the scope the export actually applied

Shopify's order export lets the person exporting choose the scope: which orders, by date range and by the filters in effect, and whether to include transaction history. An export taken from a filtered view carries that view's filters with it. A file described as all orders can therefore silently be last month's paid orders from one channel. Record the actual selections, not the intention.

Check the boundaries that commonly cut a scope: a date range that ends at midnight in the wrong timezone, a status filter that excludes pending or archived orders, and a range that starts after the oldest order the task needs. If the export was produced in chunks, for example one file per month, check the seams: an order on a boundary date can fall between two chunks or appear in both.

Also check the file's row structure before counting. In a Shopify order export, an order with several line items occupies several rows, with some order-level fields blank on the continuation rows. A naive row count therefore overstates orders, while a deduplication that keys on a blank order field can understate them. Count distinct order identifiers, not rows.

Reconcile distinct references against the source

Extract the distinct order identifiers from the file and count them. Then obtain the source system's own count for the identical scope: the same filters, statuses and date boundaries, taken as close in time as practical. A matching count supports, but does not prove, completeness; a mismatch proves a gap and its size. Record both numbers and the exact scope used for each.

When counts differ, localize before re-exporting. Compare the file's identifiers against the source list for a narrow slice, such as a single day, to find which segment is missing. A missing segment points to a chunk or page that failed; orders present in the source but absent everywhere in the file point to a filter; orders absent from the source count too point to a scope misunderstanding rather than an export defect.

Watch for the opposite defect while reconciling: duplicated identifiers from overlapping chunks. Duplicates inflate a work list just as omissions shrink it, and both must be resolved before the file drives picking, purchasing or customer communication. Document the final verified set of identifiers as the authoritative work list.

Re-pull the gap and name an owner for the recurrence

Fill a confirmed gap by re-exporting the missing segment from the source with corrected settings, then re-run the same distinct-identifier reconciliation on the combined set. Do not type missing orders into the file from memory or from a screenshot; a reconstructed row is not a source record and will not survive a later audit. If the source itself cannot produce the segment, escalate that as a platform or access question with the specific scope and counts attached.

If the same export feeds a recurring process, assign an owner to its definition: the saved filters, the date logic, the chunking and the verification step. A completeness check that runs every time is cheaper than reconciling a bad work list after the warehouse has acted on it. Record who confirmed the corrected export and when.

If the investigation shows the gap comes from an integration or a custom report nobody owns, that maintenance question is a reasonable topic for a Prism consultation. Describe the platform, the export's purpose and the measured gap in ordinary language; confirm scope, responsibilities, fees and terms before work. Keep the export itself and any customer data out of the public inquiry form.

Export completeness reconciliation

Use one sheet per suspect export. Compare requested scope, applied settings, distinct identifiers and source counts, and record where the gap localized. The verified identifier set is the only authoritative work list. Never reconstruct missing rows by hand; re-pull them from the source.

Worksheet entries are not submitted by Prism’s worksheet and are not saved by the site. Use record types, availability, anonymized observations, or match/mismatch results. Do not enter government identifiers, customer names or addresses, customer messages, receipt-access links, card or bank details, passwords, or keys. Send sensitive documents only through the provider’s verified secure channel.

These are temporary notes. Leaving or reloading this page may clear them. Worksheet entries are not sent automatically. If you copy notes into the consultation message and submit the form, Prism receives them as part of your request.