2026-03-25 brighton data forum

This time, we are diving deep into a specific data challenge and the resulting investigation.
"Are They Lying to Our Bots? A Travel Pricing Investigation"
Andres Baravalle, Principal Data Engineer, On The Beach
When building intelligence on top of web-scraped data, there is an implicit assumption: that the target site served the same content to your bot as it would to a human. The normal approach is to ingest data, clean it, match it and build dashboards on top of it.
But what if the data was wrong before it even landed?
I have spent years on both sides of the problem (back and forth, if we're being honest).
As an academic, my research focused on mining and analysing web data - from open source communities through to dark web markets. As a cloud technology specialist, I helped businesses protect their assets from online threats, including bots. I've alternated between building walls and trying to push through them.
Anti-bot solutions are often used to block or slow scrapers, but that's a trivial use. If you can instil the fear that you are serving plausible-but-wrong data, your adversary will have to commit significant resources to monitoring data quality. The goal isn't always to stop collection. Sometimes it's just to make it expensive, or to make your adversaries less confident in what they have.
In my role at On The Beach, I lead Price Intelligence. We process millions of competitor prices daily.
When routine checks hinted at systematic discrepancies between bot-collected and human-observed prices, I had to ask myself a question: could the target be feeding us what they wanted us to see?
This talk covers what happened next.

The Brighton data forum is sponsored by Rittman Analytics and supported by Silicon Brighton.