Skip to content

Statistics

Bot traffic statistics for local-business websites

By Sergey Sulimko, Founder5 min read

Imperva's 2025 report reported automated traffic at 51% of its global 2024 web-traffic profile: 37% bad bots and 14% good bots, with human traffic at 49%.12 That is a dated, global vendor-network measurement, not a percentage for local-business websites or a count of potential customers.

Cloudflare measures likely bot and human HTTP requests in its Radar population, while Google's documentation names crawler and fetcher classes without publishing their share of requests.345 These sources answer different questions, so their figures should not be combined into one local-business benchmark.

How much bot traffic do published measurements report?

What was measured Finding Source, population, period, and denominator
Imperva's global bot profile 51% automated traffic, split into 37% bad bots and 14% good bots; human traffic was 49% Imperva analysis from its global network, covering calendar year 2024; the denominator is the report's total global web or internet traffic profile.12
Imperva blocked bad-bot requests 13 trillion blocked bad-bot requests Imperva global network data from 2024 across thousands of domains and industries; this is a blocked-request volume, not the denominator for the 51%, 37%, 14%, and 49% shares.1
Cloudflare Radar's bot view A bot-versus-human percentage of HTTP requests over the selected period; the page showed a default range of last 7 days, but its readable text exposed no current numeric share Cloudflare Radar's selected HTTP-request population; the page does not provide a local-business denominator or a static current percentage in the readable view.3
Cloudflare Radar bot classes Likely automated requests use bot scores from 1 to 29; likely human requests use scores from 30 to 99 Cloudflare classification applied to the selected Radar HTTP-request set; the score bands classify requests and do not themselves report a traffic percentage.4
Named Google crawlers and fetchers Google publishes crawler names, purposes, and verification guidance, but no request-share percentage Google's crawler and fetcher documentation; no local-business sample or traffic denominator is stated.56

The Imperva percentages are the only numeric traffic shares in these sources. Imperva's report says the shares describe its global 2024 profile, and its 13 trillion figure is a separate count of blocked bad-bot requests.1 The 13 trillion requests must not be added to the percentage rows.

What does Cloudflare count as bot traffic?

Cloudflare's Radar documentation counts complete HTTP and HTTPS requests from end users reaching websites served by Cloudflare's CDN. It can return request data separated into likely human and likely automated series through the botClass parameter.7

Cloudflare's bot classes are score bands, not verified identities. Radar classifies scores from 1 through 29 as likely automated and scores from 30 through 99 as likely human.4 A request-level class therefore does not equal a person, a visit, a lead, or a customer.

What do Google's crawler documents show?

Google's documentation identifies crawler and fetcher types, but it does not publish the percentage of requests that named Google clients make to local-business websites.5 Google describes crawlers as programs that automatically discover and scan websites. Fetchers typically make a single request on a user's behalf, so a fetcher request is not automatically an ordinary search-crawler visit.5

Google's common-crawler reference names classes such as Googlebot, Googlebot-Image, Googlebot-Video, GoogleOther, and Google-Extended. The reference warns that its list is not exhaustive and that a user-agent string can be spoofed.6 Server-log identification therefore needs Google's documented verification method, not just a text match on the user-agent field.

This distinction matters when a business reviews how its site is accessed. Where does ChatGPT get local-business data? addresses information sources for assistant answers, while does schema move AI citations? addresses a separate citation question. Neither question can be answered by treating a crawler request as a customer visit. For the wider local-visibility context, see the Google Maps optimization guide.

What do these bot-traffic statistics not tell you?

The published measurements do not give a current bot-traffic rate for local-business websites. Cloudflare's request population is broader than local businesses, Imperva's shares describe a global vendor-network profile from 2024, and Google's documents describe request classes rather than a measured share.135

They also do not measure unique visitors, sessions, page views, leads, bookings, conversions, or traffic from possible customers. A bot request can appear in request or analytics totals without representing a person or a commercial opportunity.

Use the denominator beside the number when you cite it: Cloudflare for classified HTTP requests, Imperva for a dated global traffic profile and blocked-request volume, and Google for named crawler and fetcher definitions. None is a universal benchmark for a local-business site.

Sources

  1. Imperva, 2025 Bad Bot Report PDF, covering 2024 global network traffic and blocked bad-bot requests: open the report (read 2026-10-11)

  2. Imperva report release post dated April 15, 2025 and describing the 2024 global profile: Imperva report release (read 2026-10-11)

  3. Cloudflare Radar, “Bot Traffic Worldwide,” bot-versus-human HTTP-request view with a last-seven-days default at read time: Cloudflare Radar bot traffic (read 2026-10-11)

  4. Cloudflare, “Bot classes,” score ranges for likely automated and likely human requests: Cloudflare bot classes (last updated April 20, 2026; read 2026-10-11)

  5. Google Search Central, “Overview of Google crawlers and fetchers (user agents),” crawler and fetcher definitions: Google crawlers and fetchers (read 2026-10-11)

  6. Google Search Central, “Google's common crawlers,” crawler identities, purposes, and verification guidance: Google's common crawlers (read 2026-10-11)

  7. Cloudflare, “HTTP requests,” definition of complete HTTP(S) requests reaching Cloudflare-CDN-served websites and botClass examples: Cloudflare HTTP requests (last updated April 20, 2026; read 2026-10-11)

See where your profile stands

The audit shows how this applies to your own profile: it reads a verified Google Business Profile and scores it the same way every month.

Visibility Score

0 to 100

20 scored factors across profile completeness, performance, competitive position. Free, no signup, on any verified Google Business Profile.

Run the free Visibility Score audit

No signup, no card. You get the score and the factors behind it.