Bot exclusion
Learn how we exclude bot traffic from your metrics.
As AI integrates into daily life, web crawlers and bots are increasingly active online. They serve legitimate purposes like SEO and content indexing, but their traffic can skew analytics, experiments, and growth metrics.
Most bots operate server-side, but advanced ones also execute client-side code. Since we support both server- and client-side rendering, bots can fetch and render your defined content, inflating monthly visitor counts and polluting tracking data.
Our bot exclusion features address this by restricting bot access to experiences and blocking bot-generated events, minimizing their impact on experiments and ensuring cleaner, more accurate metrics.
This feature is available only to accounts on the Growth and Scale plans. See the pricing page for more information.
How it works
We use the WhatIsMyBrowser user-agent list to identify non-human traffic. The list is actively maintained and catalogs user agents for known spiders and bots.
You can create specific experiences and experiments for bot traffic by using the following condition in your audience definitions:
device's category is "bot"If you do not want to serve specific content to bots (for example, SEO-focused content) and want to remove them from your monthly visitor count, you can turn on the exclusion flag in workspace settings.
When active, bot exclusion applies at two levels: content rendering and event tracking.
Content rendering
When a request comes from known bots, we serve the default content for the slot instead of experience content, or experiment variations.
Event tracking
For known server-side and client-side bots, we do not track any events, so their activity and sessions are excluded from monthly visitors, general analytics, and all experience and experiment metrics.
Besides, for unknown server-side bots that request content on the server but never run on the client, we also do not track any events.