The user in this Reddit post has run into a problem that every spreadsheet power user eventually faces: data that disappears from its source but still needs to live on in the report. They are trying to append daily CSV updates without losing the history. That is a reasonable goal, and it exposes a fundamental limitation of how most people treat spreadsheets as data storage.
Here is the hard truth: a CSV file is not a database. When your boss overwrites that external CSV each day with only new records, the old records are gone. Power Query, for all its power, reads what is in the file at the moment of refresh. It has no memory of what was there yesterday. So yes, if you load that CSV directly into Power Query today, you will only see today's data. The historical rows you need for the external report will vanish. That is not a bug; it is the consequence of treating a daily snapshot as a live source.
The practical fix is not to fight Power Query's design but to change your data architecture. Instead of pointing your query at the overwritten CSV, point it at a persistent table where you store the appended history. This could be a simple Excel table in a separate workbook that acts as a staging area, or it could be a proper database if your organization has one. Each day, before you refresh your main report, you copy the new CSV rows into that staging table. Then your Power Query reads from the staging table, which already contains all prior days' data plus the new rows. You lose nothing, and you gain a clean, repeatable process.
The deeper lesson here is about ownership of your data. Your boss controls the CSV output, but you control how that data enters your workflow. Relying on a file that gets erased each day is a recipe for lost work and manual re-entry. By inserting a simple persistence layer, whether a hidden sheet, a separate workbook, or a database, you take ownership of the historical record. That is the difference between reacting to data and managing it.