TL;DR
- AI recommendation traffic is starting to show up as a distinct revenue line, not just a traffic bump. Two recent merchant data points on Agentic Page — a large new brand partnership and a two-month attribution result on an existing merchant — both point to the same underlying shift.
- The size of brand now testing this channel is changing. A fashion and accessories brand doing $1M+ in monthly store-wide GMV recently began working with DeepLumen on Agentic Page — a signal that AI-driven discovery is no longer a small-merchant experiment.
- On a separate merchant, two months of tracked data show what "AI as a channel" looks like in practice. $20K in revenue was directly attributed to Agentic Page in that window, alongside a reported 110% increase in AI-attributed orders after setup.
- Neither figure is a guarantee for every store. Both are specific to one merchant, one time window, and one metric definition — useful as evidence of what becomes measurable, not as a universal outcome.
What actually changed — an ad click, or something upstream of it?
For most of digital commerce's history, "channel performance" meant measuring clicks, impressions, and conversions from search or social ads. AI chat and AI-generated recommendations complicate that model because they sit upstream of a click: a shopper asks an AI system a question, gets a recommendation, and may visit or buy without ever seeing a traditional ad. That traffic used to be invisible — folded into "direct" or "other" in most analytics setups, with no way to tell whether an AI recommendation drove it. What's changing now isn't that AI influences shopping — it already did — it's that platforms are starting to make that influence measurable and attributable, the same way a paid channel is measurable.
Why does a $1M+ GMV brand partnership matter more than the dollar figure itself?
A fashion and accessories brand running $1M+ in monthly store-wide GMV (based on recent performance exceeding $3M in GMV over a two-month window) recently entered a collaboration with DeepLumen to grow through Agentic Page. This is disclosed as a new-merchant partnership, not a performance result — the brand's own GMV describes the scale of the store, not anything attributable to AI yet, since the collaboration is just beginning. Why it's still notable: it signals which size and category of merchant is now choosing to test AI-driven discovery deliberately, rather than discovering it by accident. Fashion and accessories is a category where product variation, styling context, and comparison shopping are exactly the kind of decisions AI chat tools are increasingly asked to help with.
What does a real attribution result look like, once the setup period is over?
On a separate fashion merchant that has been running Agentic Page for two months, DeepLumen's tracking shows $20K in revenue directly attributed to AI-driven visits and orders in that window — a single-merchant, two-month, AI-attributed-sales figure, not a platform-wide average. The same merchant's AI-attributed order count grew 110% after enabling Agentic Page, per DeepLumen's internally reported comparison. Both numbers describe the same underlying capability: once product and brand data is structured for AI systems to read, a merchant can see — with a dollar figure and an order count, not a vague "AI seems to be helping" impression — what AI-driven traffic is actually contributing.
Does this mean every merchant should expect similar numbers?
No — and that caveat matters as much as the figures themselves. Results depend on category, product competitiveness, price, inventory, brand awareness, and the quality of the underlying product information, the same variables that determine performance on any other channel. A $20K attributed-revenue figure and a 110% order-growth figure describe what was measured for one merchant over one specific window; they are evidence that attribution is possible and that AI-driven demand is real, not a forecast for what a different store in a different category will see. The more durable point is structural: before a merchant can know whether AI-driven traffic is helping, the traffic has to be visible and separated from "direct" and "organic" in the first place — that visibility is the prerequisite, not the result.
What's the common thread between a new $1M+ partnership and a two-month attribution result?
Both are evidence of the same shift restated at different points in a merchant's timeline. A large brand choosing to test the channel shows where demand for AI-readiness is heading. A smaller merchant's two-month attributed-revenue and order-growth data shows what becomes visible once that readiness is in place and running. Across DeepLumen's broader network of 850+ Shopify merchants on Agentic Page, the pattern holds directionally even though individual results vary: AI-driven visits and sales stop being an assumption and start being a line item a merchant can look at, the same way they'd look at performance from any paid or organic channel.
What should a merchant actually take from this, before assuming their store would see the same thing?
The useful takeaway isn't "expect $20K" or "expect 110% growth" — it's that those numbers only exist because the underlying product data was structured so an AI system could read it accurately and the resulting visits and sales could be separated out and measured. A merchant evaluating this for their own store should start by asking whether they can currently tell which of their visits and orders, if any, are AI-driven at all. For most stores today, the honest answer is no — not because AI isn't influencing their shoppers, but because nothing in their current setup is built to detect and attribute it. That gap, not the specific dollar figures above, is the actual starting point.
Where DeepLumen fits
Agentic Page structures a Shopify merchant's product and brand data so AI systems can read it correctly, keeps that data synced as the catalog changes, and gives merchants visibility into the AI-driven visits and sales that would otherwise be invisible in standard analytics — priced on actual attributed sales rather than upfront cost. It doesn't promise a specific dollar figure or growth rate for any given store; what it offers is the measurement layer that makes a number like "$20K attributed" or "110% order growth" possible to see in the first place, for merchants of any size, from newly onboarded $1M+ GMV brands to smaller stores just starting to test the channel.
Curious what AI-driven traffic looks like for your own store? Learn more about Agentic Page or book a demo to see what's currently measurable in your catalog.