Use Cases • 4 min read

Agentic commerce is live on your site. Can you see it?

Van Staub

Manager Solutions Architecture at Fullstory

Last updated: 10/15/2026

Your teams have spent lots of time and money over the years understanding who is coming to your site and designing their optimal path. In the past couple months, a whole new group of users have arrived on your site and most customers don’t know what it is doing. As our data science team recently pointed out, Muse gives itself away by how it moves. No name tag, no “Hi, I’m a bot,” just a pointer that teleports from click to click.

Spotting Muse is step one. The questions customers ask next are bigger. What is it buying? Where does it go? When does a person take over?

So we built a dashboard that answers them, and it’s live in Fullstory today.

Fullstory template selection screen with the Meta Muse AI agent activity dashboard highlighted.

Which camp are you in?

In the past couple of weeks, we’ve talked with teams across retail, travel, hospitality, and more about Muse. Their stances land in three camps.

Block it. One brand has tightened its CDN settings to keep Muse out. It still wants to know how often Muse shows up and what it tries to do.

Welcome it. One brand is encouraging agentic shopping for now. Employees who own the digital experience have already tested Muse on their own site.

Not sure yet. Plenty of teams told us the topic hasn’t yet bubbled to top priority internally, and are not sure how they want to approach it.

Still working out your approach to agents? Hear how Wingstop, Bilt, Superbet, and Happy Cabbage are doing it at #Bash_ on Nov. 10. Save your spot →

Meet the Meta Muse activity dashboard

Our new dashboard is a ready-made template built on a Meta Muse segment added to each metric. It uses multiple signals, so every customer sees real activity, even if you're not capturing IP addresses. If you are, check out the IP address metric to make your Meta Muse segment even more accurate.

Open it to see what Muse does on your site and where your experience needs work.

Meta Muse activity dashboard showing completed orders, total revenue, average order value, and average orders per buyer, alongside a Muse sessions activity chart and a list of recent AI agent sessions.

What you can see

Commerce. Completed orders, total revenue, average order value, and orders per buyer. This is where you see how much agentic commerce flows through your site.

Engagement. Clicks, rage clicks, dead clicks, error clicks. Each card is a single number, and you can click into any of them for the breakdown behind it.

Human takeovers. A fully agentic Muse session doesn’t highlight, copy, or paste. It pulls what it needs straight from the page. So when those actions show up, a person likely took the wheel. Maybe they weren’t ready to let an agent handle the checkout. Either way, you can see where pure agent sessions end and human sessions begin, plus the handoffs in between.

Errors. Muse may have a different error rate than humans have. That can reflect how agents interact with a page. It can also point to something on your site that isn’t working for them.

Navigation. Pages per visit, scroll depth, entry and exit pages, active versus inactive time. Humans spend more time being active. Muse spends a lot of time inactive. And the top pages surprised us. On one site, Muse landed on flight results more than the homepage. It figured out the URLs and skipped the front door. If you’ve built your experience around the homepage, that’s worth knowing.

Make it your own

The dashboard is tuned for Muse, but the segment behind it is yours to change.

Meta Muse dashboard showing navigation metrics, including pages per visit, single-page visits, time on page, scroll depth, and a line chart comparing visited-page activity across categories.

Duplicate the template and swap in a segment like “Suspicious Activity is AI Agent when the Device type is Desktop.” Every card updates, and you’re telling the story for a whole class of agents instead of one. When we did this, the dashboard surfaced a session running on Claude, and we could read it the same way: what it did, where it struggled, how it checked out.

That matters because the question for most teams isn’t “how do we stop agents?” It’s “how do we serve humans and agents well?” The answer starts with seeing both.

See Muse on your site

The Meta Muse activity dashboard is in Fullstory now. You’ll find it in our Template Gallery, or you can jump straight in for North American and European orgs.

Not yet a Fullstory customer? We’d love to show you a demo.

Google logoWant more like this in your Google results? Add Fullstory as a preferred source.
Van Staub
Manager Solutions Architecture, Fullstory

Additional Resources

Illustration of a monitor with charts and a pie graph connected to icons for questions, notes, and search, representing analytics data being queried through Fullstory MCP.
3 early MCP insights and learnings from our customers

3 early Fullstory MCP customers on querying data directly, catching issues before customers report them, and auto-generating bug tests.

Read the blog
Pizza-Hut-A/BTesting-Blogpost-Image
How Pizza Hut turns behavioral data into AI-powered experimentation

How Pizza Hut moved from manual A/B testing to AI-built product affinities—using Fullstory behavioral data to see the "why" behind every experiment.

Read the blog
Vio Traveling
How Vio.com creates smooth booking experiences for over 100M travelers

Learn how Vio.com makes data-driven UX decisions and sparks product ideas with Digital Experience Intelligence.

Read the blog