How We Did the Research (and Where It Broke)
We chose Google Workspace Marketplace as the target platform: 5,300+ public apps, 3 billion+ Workspace users, native pricing support in the listing itself. No separate payment stack needed for v1, and Google's review process, while slow, has clear rules.
The research method was simple: pick a category, compare install count to star rating on the same listing, read the actual review text for specifics. Cross-check against Capterra and Trustpilot where available. Do not mix one product's rating with another product's complaints.
That last rule existed because we broke it.
The mail merge mistake
Early analysis flagged mail merge as a gap category. The evidence looked clean: a category leader at 1M+ installs with a 4.8/5.0 rating, and a set of reviews describing broken tracking, slow merges, unresponsive support.
The problem: those reviews weren't from the 4.8-rated product. They were pulled from smaller, lower-tier competitors and then cited as if they described a general category problem. The category leader is genuinely well-executed. There is no gap there.
This is the kind of error that's easy to make when you're pattern-matching across multiple sources quickly, and hard to catch unless you go back and verify which specific product each review is about. We caught it by re-reading the raw sources. Documenting it here because it's exactly the type of analytical mistake that doesn't usually get published.
E-signature: real gap, wrong floor
PandaDoc for Google Docs has ~640K installs and a 2.9/5.0 rating. The core complaint is structural: recipients get funneled into PandaDoc's paid ecosystem just to sign a document they were sent. That's a real wedge.
But e-signatures carry legal weight — ESIGN Act, eIDAS, audit trails, timestamping, identity verification. These aren't optional polish items you add in v2. The minimum viable bar is high, and the liability for doing it sloppily is real. We passed.
That left PDF conversion, which is the next post.