Differences from Other Analytics Tools
Pintelly's visitor, session and page view counts will not match Google Analytics, Yandex Metrica or another analytics tool exactly. Each tool decides with its own rules who counts as a "visitor", which traffic is a bot, and where a day starts and ends. This page covers the most common causes and how to compare the numbers correctly.
Before comparing
- Use the same date range and time zone. Pintelly reports draw day boundaries in the time zone selected in the dashboard (Europe/Istanbul by default). Google Analytics uses the property's reporting time zone; Yandex Metrica uses the counter's time zone.
- Compare completed days. Google Analytics 4 processes intraday data with a delay in its standard reports; today's number can keep changing until the evening or the next day. Pintelly events reach reports within minutes. Seeing a several-fold gap during the day is therefore expected.
- Compare the same metric. Google Analytics "Active users" and "New users", Yandex Metrica "Visitors" and Pintelly "Unique visitors" are different definitions (see below).
Bots and automated traffic
The largest gaps usually come from bot traffic. Pintelly passes every request through stages ordered from cheapest to hardest; once a stage calls it a bot, the later stages do not run:
- Claimed crawler. If the User-Agent says it is a crawler such as Googlebot, Bingbot, GPTBot, ClaudeBot, Applebot or YandexBot, the request is checked against the IP ranges that company publishes, or by reverse DNS where it publishes no list or its list lags. A request that cannot be verified is counted as a bot labelled "(fake)".
- Known bot User-Agents. Community-maintained crawler and bot lists.
- Forged User-Agent. Version combinations no real browser sends.
- Data-center IP. Requests that look like a real browser but come from cloud and hosting providers such as Tencent, Alibaba, Huawei, AWS or Google Cloud.
Traffic counted as a bot is excluded from visitors, sessions and page views, and listed separately under Analytics → Crawlers & Bots.
Google Analytics and Yandex Metrica filter bots mainly by User-Agent and their own known-bot lists. Automated traffic that behaves like a real browser and comes from cloud servers can be counted as users there. Pintelly may therefore show fewer visitors than other tools on some days; the Crawlers & Bots tab shows what was filtered.
A real person browsing through a VPN or a cloud-based proxy comes from a data-center address and can be counted as a bot. On most sites this is a small share of visitors.
Consent settings
Pintelly does not track a visitor who refuses analytics in your site's cookie tool (see Consent Management). These visitors are counted anonymously in unique visitors, but no session or behavioral data is collected.
A tool that is not connected to the same cookie tool keeps measuring the visitor who refused, for example a Google Analytics tag without Consent Mode. If it is connected and the visitor refused, Google Analytics may not show that visitor at all, or only as a modeled estimate. These two cases push the gap in opposite directions.
Visitor identity
- Google Analytics recognises a visitor by the ID in the
_gacookie, Yandex Metrica by the_ym_uidcookie. - Pintelly recognises a consented visitor by its own ID. A visitor without consent is counted by an irreversible digest of IP and User-Agent that changes every day (UTC), so such a visitor is counted again on each day of a multi-day range.
- People who clear cookies, use private windows or switch devices may count as separate visitors in any tool; tools merge them to different degrees.
Metric definitions
| Pintelly | Closest equivalent | Difference |
|---|---|---|
| Unique visitors | GA "Active users", Yandex "Visitors" | Pintelly counts real (non-bot) identities that sent any event in the range. GA defines an active user through an engaged session or a first visit. |
| Sessions | GA "Sessions", Yandex "Visits" | In Pintelly a new session starts after 30 minutes without activity. Sessions are kept for consented visitors only. |
| Page views | GA "Views", Yandex "Pageviews" | How route changes in single-page apps are detected differs between tools. |
Blockers and load timing
Ad and tracker blockers block each tool's script with different lists. If a visitor leaves before a script loads, the tool loaded later on the page misses that visit. Both effects usually account for a few percent.
How bot detection uses the IP address
- Every request: The IP address is compared in memory with published data-center and crawler IP ranges and released right after. For visitors without consent it is not written to any record for this purpose.
- Only when suspicious: If a request claims to be a known crawler but is outside the ranges that company publishes, a reverse DNS lookup is made for the IP. The result is kept in memory briefly so the same address is not looked up again.
- Purpose and legal basis: This processing is necessary for accurate statistics and to protect the service against abuse: automated traffic that looks like a real browser cannot be told apart by its User-Agent. It is done in the moment to make that distinction, not to store the data. The legal basis is legitimate interest (KVKK Art. 5(2)(f), GDPR Art. 6(1)(f)).
- IP list on the Crawlers & Bots screen: It is built only from requests whose IP was already recorded under consented measurement. It is there so you can block that traffic at your firewall or CDN and spare your site the load.