diff --git a/contents/customers/creatify.md b/contents/customers/creatify.md index 74dc3c521421..4241e99c3def 100644 --- a/contents/customers/creatify.md +++ b/contents/customers/creatify.md @@ -40,7 +40,7 @@ When the PostHog web analytics beta launched in November 2023, Creatify was amon /> -The team watched with anticipation over the following months as the beta grew and new features released, often sharing feedback to ensure the dashboard included the metrics they cared about most. When this wasn’t possible they simply created their own product analytics dashboards to break down the data in new ways. +The team watched with anticipation over the following months as the beta grew and new features released, often sharing feedback to ensure the [dashboard](/docs/web-analytics/dashboard) included the metrics they cared about most. When this wasn’t possible they simply created their own product analytics dashboards to break down the data in new ways. “I don’t think I could ever go back to GA4 now that we’ve adopted PostHog,” said Xin. “Web analytics gives us all the metrics we really care about, and we’ve set up dashboards for breaking down specific segments when we need to. It’s great that we have all the metrics we care about so readily available.” diff --git a/contents/customers/significa.md b/contents/customers/significa.md index 89d524d822d9..84f678910401 100644 --- a/contents/customers/significa.md +++ b/contents/customers/significa.md @@ -27,7 +27,7 @@ When it came to web analytics, Significa used to rely on Plausible. But, as the ## Web analytics was just the beginning -Web analytics in PostHog is similar to Plausible – it’s centered around a single, clear dashboard that summarizes relevant traffic data – Significa wanted the ability to go deeper. +Web analytics in PostHog is similar to Plausible – it’s centered around [a single, clear dashboard](/docs/web-analytics/dashboard) that summarizes relevant traffic data – Significa wanted the ability to go deeper. “We were using Plausible but it doesn’t really have any cool features to analyze data more deeply,” said Tomás. “PostHog, on the other hand, enables us to analyze our traffic in more ways and to set up funnels when we need to.” diff --git a/contents/docs/data/channel-type.mdx b/contents/docs/data/channel-type.mdx new file mode 100644 index 000000000000..fe14143e7821 --- /dev/null +++ b/contents/docs/data/channel-type.mdx @@ -0,0 +1,93 @@ +--- +title: Channel type +sidebar: Docs +showTitle: true +availability: + free: full + selfServe: full + enterprise: full +--- + +Channel type is the type of acquisition channel the user came from, for example: + +| Channel Type | Description of where the user came from | +|------------------|---------------------------------------------------------------------------------| +| Direct | User typed in the URL directly or used a saved link. | +| Paid Search | An ad from a search engine, e.g. Google, Bing, or Baidu. | +| Paid Social | An ad from a social media platform, e.g. Facebook, LinkedIn, or Twitter | +| Paid Video | An ad from a video platform, e.g. YouTube or Twitch. | +| Paid Shopping | An ad from a shopping platform, e.g. Amazon or eBay. | +| Paid Other | An ad from an unknown platform. | +| Cross-Network | A cross-network ad | +| Organic Search | A non-ad search result from a search engine, e.g. Google, Bing, or Baidu. | +| Organic Social | A non-ad link from a social media platform, e.g. Facebook, LinkedIn, or Twitter | +| Organic Video | A non-ad link from a video platform, e.g. YouTube or TikTok. | +| Organic Shopping | A non-ad link from a shopping platform, e.g. Amazon or eBay. | +| Affiliate | An affiliate link. | +| Referral | A referral link. | +| Email | A link from an email. | +| Display | A display ad, e.g. an ad on Google Display Network. | +| SMS | A link from an SMS. | +| Audio | An audio ad, e.g. a podcast ad. | +| Push | A push notification. | +| Other | A link from an unknown source. | + +It is calculated using the referring domain, [UTM tags](/docs/data/utm-segmentation), and advertising IDs that are present in the URL when the user first lands on your website,. + +### Channel type calculation + +Traffic is considered "Paid" if any of the following are true: + +* `utm_medium` is one of `cpc`, `cpm`, `cpv`, `cpa`, `ppc` or `retargeting` +* `utm_medium` starts with `paid` +* Either `gclid` or `gad_source` is present + +We use a list of known values for `referring_domain`, `utm_source`, and `utm_medium`, which you can view in the [source code](https://github.com/PostHog/posthog/blob/master/posthog/models/channel_type/channel_definitions.json). + +These rules are applied in order, and the first one that matches is used to determine the channel type. + +| Channel Type | Logic | +|------------------|---------------------------------------------------------------------------------------------------------------------------------------------------| +| Cross Network | `utm_campaign` equals `cross-network` | +| Paid Search | Traffic is paid, and `utm_source` matches our list of search sources (e.g. `google`, `bing`, `duckduckgo`) | +| Paid Social | Traffic is paid, and `utm_source` matches our list of social sources (e.g. `facebook`, `linkedin`, `twitter`) | +| Paid Video | Traffic is paid, and `utm_source` matches our list of video sources (e.g. `youtube`, `twitch`, `disneyplus`) | +| Paid Shopping | Traffic is paid, and `utm_source` matches our list of shopping sources (e.g. `amazon`, `etsy`, `ebay`) | +| Paid Search | Traffic is paid, and `referring_domain` matches our list of search sources (e.g. `google.com`, `bing.com`, `duckduckgo.com`) | +| Paid Social | Traffic is paid, and `referring_domain` matches our list of social sources (e.g. `facebook.com`, `linkedin.com`, `twitter.com`) | +| Paid Video | Traffic is paid, and `referring_domain` matches our list of video sources (e.g. `youtube.com`, `twitch.tv`, `disneyplus.com`) | +| Paid Shopping | Traffic is paid, and `referring_domain` matches our list of shopping sources (e.g. `amazon.com`, `etsy.com`, `ebay.co.uk`) | +| Paid Shopping | Traffic is paid, and `utm_campaign` matches the regular expression `^(.*(([^a-df-z]|^)shop|shopping).*)$`| +| Paid Social | Traffic is paid, and `utm_medium` matches our list of social mediums (e.g. `sm`, `social-media`, `social-network`) | +| Paid Video | Traffic is paid, and `utm_medium` matches our list of video mediums (e.g. `video`) | +| Display | Traffic is paid, and `utm_medium` matches our list of display mediums (e.g. `display`, `interstitial`, `banner`) | +| Paid Search | Traffic is paid, and `gad_source` equals `1` | +| Paid Video | Traffic is paid, and `utm_campaign` matches the regular expression `^(.*video.*)$` | +| Paid Other | Traffic is paid | +| Direct | Referring domain is the special value `$direct`, and `utm_medium` is not set, and `utm_source` is either empty or equal to `direct` or `(direct)` | +| Organic Search | `utm_source` matches our list of search sources (e.g. `google`, `bing`, `duckduckgo`) | +| Organic Social | `utm_source` matches our list of social sources (e.g. `facebook`, `linkedin`, `twitter`) | +| Organic Video | `utm_source` matches our list of video sources (e.g. `youtube`, `twitch`, `disneyplus`) | +| Organic Shopping | `utm_source` matches our list of shopping sources (e.g. `amazon`, `etsy`, `ebay`) | +| Email | `utm_source` matches our list of email sources (e.g. `email`, `e_mail`, `e-mail`) | +| SMS | `utm_source` matches our list of SMS sources (e.g. `sms`) | +| Push | `utm_source` matches our list of SMS sources (e.g. `firebase`) | +| Organic Search | `referring_domain` matches our list of search sources (e.g. `google.com`, `bing.com`, `duckduckgo.com`) | +| Organic Social | `referring_domain` matches our list of social sources (e.g. `facebook.com`, `linkedin.com`, `twitter.com`) | +| Organic Video | `referring_domain` matches our list of video sources (e.g. `youtube.com`, `twitch.tv`, `disneyplus.com`) | +| Organic Shopping | `referring_domain` matches our list of shopping sources (e.g. `amazon.com`, `etsy.com`, `ebay.co.uk`) | +| Organic Shopping | `utm_campaign` matches the regular expression `^(.*(([^a-df-z]|^)shop|shopping).*)$`| +| Organic Social | `utm_medium` matches our list of social mediums (e.g. `sm`, `social-media`, `social-network`) | +| Organic Video | `utm_medium` matches our list of video mediums (e.g. `video`) | +| Affiliate | `utm_medium` matches our list of affiliate mediums (e.g. `affiliate`) | +| Referral | `utm_medium` matches our list of referral mediums (e.g. `referral`, `link`, `app`) | +| Email | `utm_medium` matches our list of email mediums (e.g. `email`, `e_mail`, `e-mail`) | +| Display | `utm_medium` matches our list of display mediums (e.g. `display`, `cpm`, `interstitial`, `banner`) | +| Audio | `utm_medium` matches our list of referral mediums (e.g. `audio`) | +| Push | `utm_medium` matches our list of referral mediums (e.g. `push`, `notification`, `mobile`) | +| Organic Video | `utm_campaign` matches the regular expression `^(.*video.*)$` | +| Push | `utm_medium` ends with `push` | +| Other | No previous rule matched | + +If you would to be able to customise this logic, please upvote [this issue](https://github.com/PostHog/posthog/issues/21195). Our definitions were based on the [definitions that GA4 uses](https://support.google.com/analytics/answer/9756891?hl=en), but added our own logic to handle some edge cases they don't. + diff --git a/contents/docs/glossary.mdx b/contents/docs/glossary.mdx index 8cc7fbf738a9..d3392c30df92 100644 --- a/contents/docs/glossary.mdx +++ b/contents/docs/glossary.mdx @@ -48,6 +48,9 @@ Business to customer. It's when you make money by selling things or services to #### BAA Business Associate Agreement, or also known as a Business Associate Contract. It's a legal document signed by contractors who may have to handle [personal health information](/manual/glossary#PHI) in order to agree to and satisfy the requirements of [HIPAA](/manual/glossary#HIPAA). [Find out more about PostHog and HIPAA](/docs/privacy/hipaa-compliance). +#### Bounce rate +A bounce is a session where the user only had one pageview, no autocaptures, and spent less than 30 seconds on the page. Your bounce rate is the percentage of sessions that resulted in a bounce. Used on the [web analytics dashboard](/docs/web-analytics/dashboard). + ## C #### CAC diff --git a/contents/docs/web-analytics/dashboard.mdx b/contents/docs/web-analytics/dashboard.mdx new file mode 100644 index 000000000000..42ebcb37f76c --- /dev/null +++ b/contents/docs/web-analytics/dashboard.mdx @@ -0,0 +1,128 @@ +--- +title: Web analytics dashboard +sidebar: Docs +showTitle: true +--- + +The web analytics dashboard provides an overview of your website's traffic. + +For the selected time range, it starts with the number of visitors, views, sessions, along with trends for each, as well as average session duration and bounce rate. Each of these is compared with the previous time range, showing how much they increased or decreased. + + + +## Bounce rate + +A bounce is a session where the user only had one pageview, no autocaptures, and spent less than 30 seconds on the page. Your bounce rate is the percentage of sessions that resulted in a bounce. + +PostHog uses autocaptured events to calculate this. To make sure this value is accurate, make sure you enable [autocapture](/docs/product-analytics/autocapture) and are capturing both `$pageleave` and `$autocapture` events. + +## Paths + +Top paths and top entry paths drill down into specific pages on your site to show their views, visitors, bounce rate, and scroll depth. You can click on any of the paths to filter the dashboard for that path. + + + +### Scroll depth + +Both average scroll and deep scroll rate are calculated using how far a user has scrolled down the page and how much content has scrolled into view. + +- **Average scroll depth** is the average scroll percentage across pageviews. +- **Deep scroll rate** is the percentage of users who scroll far enough down a page to view 80% of the content. + +## Referrers, channels, UTMs + +To get an idea of where users are visiting your site from, you can see top referrers, channels, and UTM values. + + + +### Channel types + +Based on UTMs, referring domains, and more, PostHog automatically classifies traffic into specific acquisition channel types such as: + +| Channel Type | Description of where the user came from | +|------------------|---------------------------------------------------------------------------------| +| Direct | Typed in the URL directly or used a saved link. | +| Paid Search | An ad from a search engine, e.g. Google, Bing, or Baidu. | +| Paid Social | An ad from a social media platform, e.g. Facebook, LinkedIn, or Twitter | +| Paid Video | An ad from a video platform, e.g. YouTube or Twitch. | +| Paid Shopping | An ad from a shopping platform, e.g. Amazon or eBay. | +| Paid Other | An ad from an unknown platform. | +| Cross-Network | A cross-network ad | +| Organic Search | A non-ad search result from a search engine, e.g. Google, Bing, or Baidu. | +| Organic Social | A non-ad link from a social media platform, e.g. Facebook, LinkedIn, or Twitter | +| Organic Video | A non-ad link from a video platform, e.g. YouTube or TikTok. | +| Organic Shopping | A non-ad link from a shopping platform, e.g. Amazon or eBay. | +| Affiliate | An affiliate link. | +| Referral | A referral link. | +| Email | A link from an email. | +| Display | A display ad, e.g. an ad on Google Display Network. | +| SMS | A link from an SMS. | +| Audio | An audio ad, e.g. a podcast ad. | +| Push | A push notification. | +| Other | A link from an unknown source. | + + + +> **Read more:** [How channel type is calculated](/docs/data/channel-type) + +### UTMs + +UTMs include source, medium, campaign, content, and term. Each are set as URL parameters and autocaptured by PostHog. For example, the following URL has the source of `twitter`, medium of `social`, and `campaign` of `twitter-campaign`: + +`https://posthog.com/?utm_source=twitter&utm_medium=social&utm_campaign=twitter-campaign` + +Setting UTMs correctly is crucial for accurately classifying your traffic, not only for UTMs, but for channel types as well. + +> **Read more:** [How to capture, customize, and filter UTM parameters](/docs/data/utm-segmentation) + +## World map + +The world map shows where your users are located, but you can also select it to show top countries, regions (like California, England, or Ontario), or cities. + + + +## Retention + +Retention creates a cohort of unique users who performed any event for the first time in the last week. It then tracks the percentage of users who return to perform any event in the following weeks. + + + +> **Read more:** [Creating and understanding retention](/docs/product-analytics/retention) + +## Filtering your dashboard + +Like other dashboards in PostHog, the web analytics dashboard is filterable. This means you can filter for data with certain event or person property values. Options include browser, path name, device type, country, and UTMs. Just click the "Add filter" button next to the date range at the top of the dashboard. + +This enables you to dive into specific stats for regions, parts of the site, and specific marketing campaigns. + +For more complex queries, you can still use the product analytics tab as usual. diff --git a/contents/docs/web-analytics/index.mdx b/contents/docs/web-analytics/index.mdx index 8c6931df3efc..19625618fa8f 100644 --- a/contents/docs/web-analytics/index.mdx +++ b/contents/docs/web-analytics/index.mdx @@ -49,12 +49,4 @@ In PostHog, click your profile icon in the top right and then click "Feature pre ### 3. Go to the web analytics dashboard -Now, you can click the newly added [web analytics tab](https://app.posthog.com/web) to go to the dashboard. - -## Filtering your dashboard - -Like other dashboards in PostHog, the web analytics dashboard is filterable. This means you can filter for data with certain event or person property values. Options include browser, path name, device type, country, and UTMs. Just click the "Add filter" button next to the date range at the top of the dashboard. - -This enables you to dive into specific stats for regions, parts of the site, and specific marketing campaigns. - -For more complex queries, you can still use the product analytics tab as usual. +Now, you can click the newly added [web analytics tab](https://app.posthog.com/web) to go to the [dashboard](/docs/web-analytics/dashboard). diff --git a/src/navs/index.js b/src/navs/index.js index f7ff670e261d..ba69e4fd8987 100644 --- a/src/navs/index.js +++ b/src/navs/index.js @@ -1333,6 +1333,10 @@ export const docsMenu = { name: 'Annotations', url: '/docs/data/annotations', }, + { + name: 'Channel type', + url: '/docs/data/channel-type', + }, { name: 'Cohorts', url: '/docs/data/cohorts', @@ -1971,6 +1975,12 @@ export const docsMenu = { icon: 'IconHome', color: 'seagreen', }, + { + name: 'Dashboard', + url: '/docs/web-analytics/dashboard', + icon: 'IconDashboard', + color: 'orange', + }, { name: 'FAQ', url: '/docs/web-analytics/faq',