Red Reap

We build web data tools that tell you what they missed.

Most scrapers hand you rows and leave you guessing whether that was all of them. Ours report their own completeness: what the source actually holds, how much of it you received, and why collection stopped. If a run gets 71% of something, it says 71%.

---

Amazon returns at most 100 reviews behind any one filtered view. Ten pages of ten, then it stops. Every scraper hits that wall, which is why the field quietly ships a few hundred reviews from a product that has twelve thousand, with nothing to tell you that is what happened.

This one splits the review pool into hundreds of small filtered views, collects them all, deduplicates, and then reads Amazon's own written-review count for every star rating so it can tell you exactly what fraction you hold.

---

• Numbers in our listings are measured, not estimated. Every figure above came from a real run you can reproduce.
• We say what we could not get. A coverage report ships with every run, naming the star ratings that fell short and why each one stopped.
• Public pages only. No logins, no cookies, no accounts, ever.
• No personal data. We never collect the names or profile links of the people whose reviews we read, even when the source offers them.
• Duplicates are free. Reaching high coverage means deliberately buying the same review in several views. You pay for unique results, not for the work behind them.