Bot filtering

Savri automatically filters out bots and spam to give you reliable statistics. Filtering is on by default for all accounts and requires no configuration.

Why filtering?

Without filtering, your statistics can be distorted by:

  • Search engine spiders (Googlebot, Bingbot, etc.)
  • Referrer spam (fake visits from spam domains)
  • Automated tools (SEO tools, monitors, etc.)
  • Malicious bots (scrapers, vulnerability scanners)

What is filtered?

1. User-Agent-based filtering

We identify and filter out known bots based on their User-Agent:

Googlebot
Bingbot
YandexBot
Baiduspider
DuckDuckBot
Slurp (Yahoo)
facebookexternalhit
Twitterbot
LinkedInBot
WhatsApp
Discordbot
TelegramBot
Semrush, Ahrefs, Moz
Lighthouse, PageSpeed
GPTBot, ClaudeBot
PerplexityBot

AI bots like GPTBot, ClaudeBot and PerplexityBot are also filtered out of your visitor statistics. Want to see which AI bots read your site instead? Use AI crawler tracking.

2. Headless browsers and automated tools

The tracking script sends nothing when the browser flags navigator.webdriver. This catches Selenium, Puppeteer, Playwright and similar automated browsers before they even reach our servers.

3. Datacenter IP filtering

Visits from known cloud and hosting provider IP ranges (AWS, Google Cloud, Azure, Alibaba, Tencent, Hetzner, OVH and more) are filtered out server-side. Real visitors don't browse from cloud servers, traffic from there is scrapers and bots even when the user agent looks like a regular browser.

4. Referrer spam filtering

We block visits from known spam domains and patterns:

Exact domains

~100 known spam domains are blocked

semalt.comdarodar.combuttons-for-website.com

Patterns

Domains containing spam-related terms

*apk**casino**porn**seo-**free-traffic*

Suspect TLDs

Short domains with spam-associated top-level domains

.xyz.top.tk.gq.ml.click

Updated regularly

The spam list is based on Matomo's community list and supplemented with domains reported by our users.

5. Browser normalization

In-app browsers (Facebook, Instagram, TikTok etc.) are grouped as "Mobile App" to provide a more accurate picture of browser distribution.

What is NOT filtered?

We do not filter out:

  • Real visitors - all legitimate browsers are counted
  • RSS readers - if they generate pageviews
  • Your own traffic - see below for how to exclude it

Exclude your own traffic

If you want to exclude your own traffic, there are two ways:

1. localStorage flag

Open your browser console on your site and run:

localStorage.setItem('va_ignore', 'true')

Your browser will now be ignored by tracking (only applies to that browser).

2. Manual exclusion (coming soon)

The ability to exclude traffic based on IP, country, page, or hostname is coming in a future update.

Report spam

See suspicious traffic from a domain that isn't being filtered? Contact us and we'll add it to the blocklist.

Information to include

  • The domain sending spam
  • Your site (so we can verify)
  • Approximate time/date

Future improvements

We are continuously working to improve filtering:

  • Traffic pattern analysis - identify suspicious behaviors
  • Manual exclusion - exclude IP, country, pages