I thought I was building a PDF-to-Excel converter. One comment changed the product.
I started building ParseToSheet around a simple problem:
Extract data from PDFs and put it into spreadsheets.
But after talking to potential users, I realized I had framed the problem incorrectly.
Most people don’t want another newly generated spreadsheet.
They already have:
- an Excel template
- existing formulas
- fixed columns
- internal formatting
- a workflow their team already understands
What they actually want is:
“Take the data from these PDFs and put it into the spreadsheet I already use.”
That changed how I think about the product.
The PDF extraction is only one step. The real product is preserving the user’s existing spreadsheet workflow.
It also created a second problem: trust.
If AI fills an existing business spreadsheet, users need to know:
- where each value came from
- which page contained it
- which cells are uncertain
- what was changed before export
So I’m now focusing less on “PDF to Excel” and more on:
PDF → your existing Excel template → review uncertain values → export
I’m curious how other founders handle this:
When user feedback reveals that your original category was too broad, do you change the product positioning first—or rebuild the workflow first?
I’m building this at https://parsetosheet.com/
