Analytics
Know who’s reading. Including the machines.
Where every visit came from, ChatGPT and Perplexity included, how far each reader got, and which AI crawlers fetched your posts. First-party, and no cookies.
Your blog, at a glance.
Visits and views against the period before, how long people read, what they clicked, your Google clicks, and who is on the blog right now.
Where readers come from.
Every arrival is sorted into a channel and named by its source, and AI assistants are a channel of their own.
AI assistants, by name. ChatGPT, Perplexity, Gemini, Claude, Copilot and more, so you can see which of them send people your way.
Search, social and email. Google and Bing, LinkedIn and Reddit, newsletters and links from other sites, each counted where it belongs.
Your own campaigns. Tag the links you share with UTM parameters and every campaign gets a line of its own.
Read by machines, too.
Before an assistant can quote a post, something has to fetch it. Bloggable records every fetch by a crawler it recognises.
Who came. GPTBot, ClaudeBot, PerplexityBot, Googlebot and the rest of the 28 it knows, each with when it was last seen.
Why they came. Training a model, indexing for an answer engine, or fetching the page live because someone asked about it.
What they read. Posts, their Markdown copies, llms.txt and the sitemap, and which of your posts they fetch most.
Whether they stayed.
A view says someone arrived. These say whether the post held them.
Engaged time. Counted only while the page is on screen and the reader is active. A tab left open in the background adds nothing.
Read-through. How many readers got three quarters of the way down a post, and how many reached the end.
CTA clicks. Clicks on the calls to action, buttons and sign-up links in your posts, and where each one led.
Ask it your own questions.
Choose the stretch, choose what to compare it with, and take the numbers with you.
Compare any stretch.
Against the period before, or the same weeks last year with the weekdays lined up.
Leave yourself out.
Exclude your own network in one click, so checking your own post never counts as a reader.
Ask Joe.
“Why are visits up?” He answers from the same figures you are looking at.
Take it with you.
Export the posts table to CSV for a report, a spreadsheet or a client.
Explore the rest of the platform
What people ask about the numbers
No. Analytics sets no cookies and stores no IP addresses. A visit is counted from a one-way hash that includes the date, so it changes every day and a reader cannot be followed from one day to the next.
Partly because they are undercounted, and the panel says so. Assistant apps often open a link without saying where it came from, and those visits land under Direct. The figure you see is a floor, not the whole number.
From the user-agent each crawler sends, matched against the names the AI companies and search engines publish for their crawlers. Anyone can claim to be GPTBot, so it is a strong signal rather than proof, and the panel says that too.
The last seven days of visits and views, the posts you published, who is reading right now, and how fast your pages load. Advanced analytics adds longer ranges and year-on-year comparison, where readers came from, how they read, CTA clicks, AI crawlers, Search Console figures and CSV export. The plans page shows which plans include it.