> For the complete documentation index, see [llms.txt](https://helps.ptengine.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://helps.ptengine.com/en/insight/glossary.md). # Insight Glossary ### Visitors (UV) The number of people who visited the website within a specified time period. Within that period, the same user is counted as 1 UV regardless of how many times they visit. Whether a user is the same person is determined by the user identifier (uid). If the uid is the same, they are considered the same user. The uid is stored in the browser's cookie, so switching browsers or devices will result in being counted as a different user. Additionally, in Safari, because the cookie retention period is 7 days, if a user revisits after more than 7 days, the cookie will be deleted and they will be counted as a new user. ### Sessions The number of visits to the website. Also known as session count. When a user visits the website, it counts as 1 session. If the user leaves the website and then returns, it counts as another session. A session ends (the user leaves) in the following situations: 1. The browser (tab) is closed. 2. The user navigates to a page within the project that is not being tracked. 3. The user has no activity for a certain period of time. Note: Definition of no activity for a certain period of time. Activity: Behaviors such as clicking, scrolling, typing, reloading a page, or opening a new page have not occurred. Time: The default session timeout is 30 minutes. You can also customize this duration. Within the above time period, if the user returns from a different traffic source (excluding direct traffic), it will be counted as a second session. If the traffic source is the same, a new session will not be counted. ### Pageviews (PV) The number of times a page is loaded. "Loaded" refers to the number of times Ptengine's JavaScript tracking code is loaded; each load counts as 1 PV. ### Pages per Visitor The average number of pages viewed per visitor, reflecting both website content quality and visitor quality. ### Clicks PC: The number of "mouse clicks".\ Smartphone: The number of "touches." All taps are counted regardless of whether a link exists, and consecutive clicks/touches are also counted. \*For smartphones and tablets, touches from swiping and pinch-to-zoom gestures are not counted as clicks. ### New Visits (New Session Count) The total number of sessions from users visiting the website for the first time since the profile was created.\ If there is no historical visit record within the project's website, the visitor is considered new. Even if they have visited before, if the user has deleted cookies or other data, they will be considered a new visitor. ### New Visit Percentage The proportion of new visitor sessions out of total sessions.\ New Visit Percentage = New Sessions / Total Sessions. ### Return Visits (Return Session Count) The number of sessions from users who have visited the website 2 or more times within the cookie retention period (\*).\ Return Sessions = Total Sessions - New Sessions. (\*) When a user interacts with Ptengine's tracking code, Ptengine provides a persistent cookie to the browser that does not expire. Whether a visitor is classified as "new" or "returning" depends on whether this cookie is retained. However, due to ITP, for Safari users, we distinguish between new and returning visitors based on the past 7 days. ### Return Visit Percentage The proportion of return visitor sessions out of total sessions.\ Return Visit Percentage = Return Sessions / Total Sessions. ### Average Load Time The average time it takes for a page to finish loading.\ This measures the time from the page request to the completion of page loading for each page.\ Average Load Time = Total page load time across all visits / Number of visits. ms = 1/1000 of a second; for example, 629ms = 0.629 seconds. ### Average Time on Site There are two types of average time metrics — Average Time on Site and Average Time on Page. They differ in both how they are calculated and what they are used for. #### Average Time on Site (Average Session Duration) Unit of measurement: a single complete visit (session).\ It represents the average total time a user spends from entering the website to leaving it.\ Average Time on Site = Total duration of all sessions / Total number of visits (sessions). How a session is counted: 1 session = a visitor opens the website → the session ends after 30 minutes of inactivity, closing the page, or navigating to an external site.\ Example: In one visit, a user views the homepage for 30s and a detail page for 90s. The session duration is 120s, which is added to the site-wide total. #### Average Time on Page (Average Page Duration) Unit of measurement: a single page view (PV).\ It represents the average time a user spends on one specific page.\ Average Time on Page = Total time spent across all PVs of that page / Total PV count of that page. #### Example Scenario: In one session, a user browses Homepage (40s) → Product page (80s) → Checkout page (60s), for a total session duration of 180s. 1.Average Time on Site: This session contributes 180s. The total across all sessions divided by the number of sessions reflects how long users spend browsing the entire site on average. 2.Average Time on Page for each page:\ ・Homepage: only the interval from opening the homepage to navigating to the product page is counted, i.e. 40s\ ・Product page: the interval from opening it to navigating to the checkout page, i.e. 80s\ ・Checkout page: as the last page of the session, its full 60s stay is captured via heartbeat tracking #### How to Use Each Metric in Analysis What Average Time on Site tells you: 1.Channel quality: for the same channel, a longer average session duration means more precisely targeted traffic\ 2.Overall site experience: whether the navigation and content flow work smoothly across the site\ 3.Conversion potential: longer session durations usually correlate with better purchase/lead conversion\ 4.Companion metrics: best viewed together with bounce rate and pages per visit What Average Time on Page tells you: 1.Landing page quality: short time on the homepage or an ad landing page = content does not match the traffic\ 2.Content page value: the longer users stay on articles or detail pages, the more carefully they are reading\ 3.Drop-off points: high exit rate + short time on page = the page has experience issues (slow loading, cluttered information)\ 4.Targeted optimization: you only need to optimize that single page (adjust copy, images, button placement) #### Common Misconceptions Misconception 1: Average Time on Page × Pages per Visit = Average Time on Site\ Not exactly equal: the last page of a session has no navigation interval, and Ptengine's heartbeat-based time supplementation creates a small difference. Use this equation only as an approximation. Misconception 2: Bounce sessions (viewing only 1 page before leaving) are not counted in time on page\ Ptengine fully captures the time spent on the bounced page, so page-level metrics are recorded normally. The visit is also fully counted in the site session duration. Which metric to use:\ ・To evaluate the overall appeal of your site: use Average Time on Site\ ・To identify which page fails to retain users: use Average Time on Page ### Dimension We commonly use "dimension" to describe characteristics such as size or color, but in data analysis, it typically refers to user attributes: traffic source, device, channel, geographic information, etc. ### Metric In web data analysis, metrics generally fall into two types: volume metrics and quality metrics. Volume metrics: Sessions, PV, UV\ Quality metrics: Average session duration, pages per visit, bounce rate, new visit percentage. ### Filter Toggle You can quickly view a specific attribute in the data center. For example, you can use the segment feature to view metrics such as PV and sessions for smartphone users only. ### Segment A feature that allows you to group specific sessions or users together. Using segments, you can split and analyze subsets of your data. For example, by applying a segment to an ad channel, you can determine which ads led to conversions. ### Bounces A bounce occurs when a user lands on a page and ends their visit without navigating to any other page. ### Bounce Rate Bounce Sessions / Total Sessions.\ Reflects both page quality and visitor quality. Note: Compared to Google Analytics, Ptengine's bounce rate tends to be higher because Ptengine considers a visit ended if the user has no activity on the site for a certain period of time. Different measurement tools have different session timeout limits. Ptengine:\ By default, PC: 5 minutes (for projects created on or after January 11, 2021, PC is 30 minutes), Mobile: 5 minutes. Google Analytics:\ 30 minutes for both PC and mobile.\ Therefore, especially for smartphones, Ptengine's criteria are stricter than Google Analytics, which tends to result in a higher bounce rate. ### Exits The number of times a page was the last page viewed in a session. This means the visit ended on that page. ### Exit Rate Exit Rate = Number of Exits / PV count for that page.\ The exit count is the number of times a page was the last page viewed in a session, meaning the visit ended on that page. ### Keywords The keywords a user searched for on a search engine when visiting the website. ### Traffic Acquisition A method of grouping traffic by rules. Ptengine's default channels include External Links, Search Engines, Social Media, Direct Traffic, and Ads. You can also create custom channels based on traffic sources. For information on custom channels, please refer to \[here]. ### Profile Refers to the website used for data measurement. By setting up a website as a profile, you can measure that website's data. ### Entry Page Also known as a "landing page." The first page a user visits. ### Entry Count The number of times a page was visited as the entry page.\ The entry page is also known as a "landing page" -- the first page a user visits. ### Entry Rate Entry Count / PV count for that page. ### Raw Page Refers to pages with unique URLs.\ For example, you can view the original URL and URLs with parameters separately:\ `http://aaa.com/about/`\ `http://aaa.com/about/?gclid=18jnie97dg`\ `http://aaa.com/about/?utm_source=google` ### Merged Page Used to merge data for the original URL and URLs with parameters.\ Example:\ Merged page:\ `http://aaa.com/about/`\ When viewing data with a merged page, the data includes all of the following URLs:\ `http://aaa.com/about/`\ `http://aaa.com/about/?gclid=18jnie97dg`\ `http://aaa.com/about/?utm_source=google` ### Heatmap The heatmap feature visually displays user behavior on a website. It shows where users look and click, as well as what content attracts their attention. ### Conversion (CV) Refers to detecting when a desired goal or action is achieved by a user on the website.\ In Ptengine, conversions are measured in the following 3 scenarios: 1. Visiting a specific page URL. By specifying the page a user will see after completing a specific action, you can measure outcomes.\ Examples:\ \*Member registration: A page containing "Submit"\ \*Product purchase: A "Thank\_You" page, etc. If a user reaches the specified page, 1 conversion is counted. 2. Visiting a specified page group. When a user visits a page in the specified page group, 1 conversion is recorded. 3. Triggering an event. When a user completes an event configured in Ptengine, 1 conversion is recorded. ### Conversion Count The number of sessions in which a conversion occurred. ### Conversion Rate Conversion Count / Total Sessions. ### Standard Conversion Standard conversions such as revenue, orders, and form submissions. ### Negative Conversion Negative conversions such as membership cancellations and returns. ### Conversion Value The assigned value per conversion multiplied by the number of conversions. For example: 1 conversion = $20, 20 conversions, so conversion value = 20 x 20 = $400. ### Conversion Funnel The conversion funnel lets you understand the steps users take to reach a conversion goal, how many users proceed to the next step at each stage, and how many drop off at each stage. ### Event Events refer to user interaction behaviors on a website that can be configured through Ptengine. Examples:\ How many "Add to Cart" button clicks occurred\ How many clicks occurred on the hero banner\ How many clicks occurred on key CTAs... and so on.