Your pages
Buying guides, comparisons and FAQs that answer what shoppers ask, plus product and collection copy that says it plainly.
(horizon 01 / get found) AI visibility for consumer product brands, measured on 7 assistants.
(back to home)Be the brand AI recommends.
Shoppers ask ChatGPT, Gemini and the others what to buy, and the answer names a few brands. We measure how often you're one of them, every month, with error bars. Then we raise those odds with root-cause fixes we can prove.
(or) see every price. baselines from $3,900, ready in 10 business days.
(shopper asks ai)
(shopper) Best wooden play kitchen for a preschool classroom?
(ai)
For a classroom, durability and safety come first. Three names come up again and again:
- 01 a brand
- 02 a brand
- 03 a brand
- 04 you?
ai named 3 brands. is yours one of them?
Shoppers now ask AI what to buy before they ever search. The answer is a short list. That list is the new shelf and most brands have never checked whether they're on it.
-
+60%
better conversion from AI-referred retail visitors than other traffic
-
+53%
more revenue per visit from AI-referred retail visitors
-
86%
of shoppers who researched with AI checked the pick somewhere else before buying
Shopify now puts eligible catalogs in front of ChatGPT, Microsoft Copilot and, for US stores, Google's AI Mode and Gemini. Being in the catalog makes you eligible. What gets you recommended is your product data and what reviews, roundups and Reddit say about you.
(how we measure)
Odds, not ranks.
Ask an assistant twice and you get a different shortlist. In one study, ChatGPT and Google's AI returned the same brand list less than 1% of the time. So a single "position" is noise. We count how often you're named across many runs, and say how sure we are. (sparktoro and gumshoe, 2,961 responses, jan–feb 2026)
- (01) mine
Start from what shoppers actually ask.
Your reviews, support tickets, site search and returns become 8–12 clusters of real shopping questions. Two clusters become controls: measured, never worked on.
- (02) sample
Ask every assistant, many times.
Each prompt runs 10–20 times on each of 7 assistants, every month. Every raw answer is stored, so any number traces back to the answers behind it.
- (03) trace
Find the sources behind the answers.
We rank the pages, roundups, review sites, Reddit threads and videos each cluster leans on. Assistants repeat what their sources say.
- (04) move
Fix, then re-test against controls.
We ship the fixes that matter and re-sample monthly. A change only counts when it clears the error bars and beats the controls.
(illustrative) one brand, one question
“Best wooden play kitchen for a preschool classroom?”
40%
named in 8 of 20 runs. 95% range 22–61%.
(7 assistants, measured separately)
- ChatGPT openai
- Gemini google
- Perplexity perplexity
- Claude anthropic
- Google AI Mode google
- Microsoft Copilot microsoft
- Meta AI meta
(google ai overviews are included wherever your buyers' questions trigger them.)
Root causes, not symptoms.
Assistants recommend what their sources say about you. So we fix the sources: three places, every month, each change logged and re-tested.
Your structured data
Product markup, Merchant Center feeds, Agentic Storefront settings, and prices and stock that bots can read without a browser.
The sources AI relies on
The roundups, review sites and threads behind each answer: earned placements, profile fixes and corrections. Always disclosed.
(the clean aeo standard)
- (01) Real reviews only. No fake, bought or gated reviews. Incentivized reviews are disclosed. The FTC's 2024 rule on consumer reviews and testimonials bans fake and bought reviews outright.
- (02) Disclosed placements. No undisclosed paid placements in "best of" roundups. Affiliate and gifting relationships are disclosed wherever the placement appears.
- (03) The feed matches the page. Product feeds and markup always match what shoppers see: the same price, stock and claims. No hidden text or prompts aimed at AI shopping agents.
- (04) No fake shoppers. No sock-puppet Reddit, forum or community posts. A brand's team takes part openly, as itself.
(prices)
Every price, published.
Start with a baseline: where you stand on all 7 assistants, in 10 business days. Then pick a plan only if the numbers say it's worth it. 90 days, then month-to-month.
(selling on shopify)
(01) start here 100% upfront
Shopify Odds Baseline
$4,500 one-time
See which of your products AI assistants recommend, and why, before you commit to anything.
8–12 shopping clusters, up to 20 hero products, 7 assistants, one readout. 10 business days from receiving your data.
Credited in full against month one if a retainer starts within 30 days of the readout.
book a Shopify baseline- 01 Shopping Questions. 8–12 clusters (40–80 prompts) built from your reviews, support tickets, site search and returns
- 02 Shelf Sheet. How often each assistant recommends your brand and each hero product, with error bars, plus who takes the shelf instead
- 03 Source Map. The 10–20 sources behind each cluster, and the roundups you're missing from
- 04 Catalog & Feed Check. Agentic Storefronts, Merchant Center, product markup, and prices and stock that bots can read
- 05 Accuracy & Leakage Check. What AI gets wrong about your products, and how often it sends shoppers to Amazon or other retailers
- 06 AI Revenue Baseline. What AI already drives today, labelled as a floor
- 07 A 90-day plan of ready-to-ship tickets, a 60-minute readout and a one-slide summary
(02) then, only if the numbers say it's worth it. the baseline is credited to month one.
(most growing brands start here)
Shelf
$4,500 /month
For growing DTC brands. We find and spec the fixes, your developer ships them, and we prove the change every month.
3 clusters worked, up to 6 store-fix tickets and 3 content or offsite actions a month
what's included (9)
- Monthly re-sampling of every cluster on all 7 assistants, for your brand and hero products
- 3 clusters worked (the rest monitored), with 2 control clusters
- Up to 6 store-fix tickets a month, ready to ship, checked once your developer ships them
- 1 answer page a month: a buying guide, comparison or FAQ, drafted for your store
- 2 offsite actions a month: roundup pitches, review-site profile fixes and corrections
- An updated Shelf Sheet and Source Map every month, with a 45-minute readout
- A quarterly test readout: worked clusters against control clusters
- Revenue Link setup in month one: Shopify's AI channel, a GA4 AI channel and an AI answer in your post-purchase survey
- A shared Slack channel with replies within one business day
90-day first term, then month-to-month with 30 days' notice
start with a Shopify baseline(monthly)
Storefront
$9,500 /month
For Shopify Plus brands. We ship the fixes on your store and earn the placements that get you recommended.
6 or more clusters worked, up to 8 store fixes shipped by us and about 3 placements a month
what's included (9)
- Everything in Shelf, with 6 or more clusters worked
- We ship the fixes on your store: theme, metafields, structured data, feed and catalog settings, product and collection copy (up to 8 a month)
- Changes made on a duplicate theme and published only with your approval; we never touch checkout or payments
- 2 answer pages a month
- About 3 earned placements a month in roundups, reviews and with creators, each disclosed
- Retailer-leakage monitoring and fixes
- A peak-season readiness review before Black Friday, Cyber Monday and major drops
- A full attribution build joining AI-channel orders, survey answers and GA4
- A monthly executive readout
90-day first term, then month-to-month with 30 days' notice
start with a Shopify baseline(add-ons)
- Fix Sprint: we ship the baseline's store fixes once (up to 12 tickets in two weeks) $3,500 one-time
- Extra collection in the baseline $750 each
- Extra worked cluster $750/month
- Extra earned placement $600 each
(plain terms)
- Credited in full against month one if a retainer starts within 30 days of the readout.
- On Storefront we work on a duplicate theme and publish only with your approval. We never touch checkout, payments or app billing.
- Prepay a year and take 5% off. Prices exclude sales tax.
- No guarantees of placements or revenue. We commit to a process and to measured change against control clusters.
(not on shopify)
(01) start here 100% upfront
Odds Baseline
$3,900 one-time
Find out where you stand before you commit to anything.
8–12 buyer-question clusters, 7 assistants, one readout. 10 business days from receiving your data.
Credited in full against month one if a retainer starts within 30 days of the readout.
book a baseline- 01 Buyer Questions. 8–12 clusters (40–60 prompts) built from your own sales calls and support tickets
- 02 Odds Sheet. How often each assistant recommends you, per cluster, with error bars, plus who it recommends instead
- 03 Source Map. The 10–20 sources that most influence each cluster
- 04 Accuracy Check. What AI gets wrong about you, traced to the source of each error
- 05 Crawl Check. AI bot access, server rendering, bot activity in your logs, structured-data accuracy
- 06 A 90-day plan. Prioritized fixes, each with an owner and an expected effect
- 07 A 60-minute readout and a one-slide summary for your CEO or board
(02) then, only if the numbers say it's worth it. the baseline is credited to month one.
(monthly)
Growth
$6,500 /month
We work your three most valuable clusters and prove the change every month.
3 clusters worked (the rest monitored), 4 fixes a month
what's included (7)
- Monthly re-sampling of every cluster on all 7 assistants (10–20 runs per prompt)
- 4 fixes a month, typically 2 answer pages and 2 offsite actions (earned placements, profile updates, corrections)
- An updated Odds Sheet and Source Map every month
- A 45-minute monthly readout
- A quarterly test readout: worked clusters against control clusters
- Pipeline Link setup in month one: AI-referral channel in GA4, an AI option in your “how did you hear about us” field, AI bot checks in your server logs
- A shared Slack channel with replies within one business day
90-day first term, then month-to-month with 30 days' notice
start with a baseline(monthly)
Category
$11,500 /month
For Series B companies in contested categories.
6 or more clusters worked, 8 fixes a month
what's included (5)
- Everything in Growth
- A digital PR program of about 4 earned placements a month
- A full attribution build joining AI referrals, survey answers and CRM opportunities
- A monthly executive readout
- Competitive monitoring of up to 5 rivals
90-day first term, then month-to-month with 30 days' notice
start with a baseline(monthly)
Advisory
$2,500 /month
Your team does the work. We supply the method and the measurement.
Up to 5 clusters measured, your team executes
what's included (4)
- A monthly Odds Sheet for up to 5 clusters
- A Source Map refresh every month
- A 60-minute working session each month
- Async review of your team's planned fixes
90-day first term, then month-to-month
talk to us(add-ons)
- In-house team workshop (half day) $1,500 one-time
- Extra monitored cluster $750/month
- Extra earned placement $600 each
- Baseline refresh (non-retainer clients) $2,500 one-time
- ChatGPT Ads planning partner referral
(plain terms)
- Payment by ACH bank transfer through Stripe Invoicing, net 15. Cards are accepted, with processing fees passed through above $1,000.
- The baseline fee is credited in full against month one if a retainer starts within 30 days of the readout.
- Prepay a year and take 5% off. That's the only standing discount.
- Prices exclude sales tax.
- No pass-through costs at standard scope. Extra placements are billed only with your written approval.
- No guarantees of positions or outcomes. We commit to a process and to measured change against controls.
questions
(10)(01) Isn't Shopify's Agentic Storefronts enough?
It's a good start, and it's on by default for eligible stores. It makes your products available in ChatGPT and Microsoft Copilot and, for US stores using Google's sales channel, in Google AI Mode and Gemini. It doesn't decide which products get recommended. That comes down to your product data and what reviews, roundups and Reddit say about you. The baseline checks your setup, then measures what actually gets recommended.
(02) Will you touch our theme?
Only on Storefront, and only on a duplicate theme you approve before anything is published. We never touch checkout, payments or app billing. On Shelf, we write exact tickets and your developer or agency ships them.
(03) What about Amazon?
We measure how often assistants send shoppers to Amazon or other retailers instead of your store, and fix what's in our control: your product data, your site and the sources assistants cite. We don't manage Amazon listings. Amazon's own assistant, Rufus, can't be measured through official APIs, so we don't report on it.
(04) How do you measure revenue?
Three ways: Shopify's AI-channel orders, an AI channel in GA4, and an "AI assistant" answer in your post-purchase survey. Analytics undercounts AI, so we always report the result as a floor, not a total.
(05) We have a small catalog. Is this worth it?
The baseline covers up to 3 collections and 20 hero products, which is most of the catalog for many growing brands. It's a one-time $4,500, credited in full if you continue.
(06) What do you need from us?
Staff access to Shopify, GA4, Search Console and Google Merchant Center, plus exports of your reviews, three months of support tickets, site-search terms and returns reasons. We sign a DPA first and redact personal data before any analysis.
(07) Can you guarantee we'll show up in ChatGPT?
No one honestly can. Assistants rarely return the same products twice, so we commit to a process and to measured change against control clusters, reported with error bars.
(08) Do you work with stores that aren't on Shopify?
Not yet. The store and feed work is built around Shopify. If you're on another platform, tell us and we'll say honestly whether a baseline would help.
(09) Isn't this just SEO?
Mostly, and Google says so too. The differences are which questions matter (conversational and long-tail), which sources answers cite (review sites, Reddit, YouTube, publishers), and how you measure (odds across many runs, not ranks). We do all three.
(10) Can you post on Reddit for us?
Only as disclosed, real participation from your own team. Undisclosed seeding now carries FTC and platform-ban risk, and it breaks the Clean AEO Standard we work to.
(free scan)
See where you stand.
The free scan asks one AI model three questions your customers would likely ask, blind, and checks whether it names you. It also checks what AI needs from your site.
- 01Does AI recommend you, of 3
- 02AI assistants can read your site
- 03Product and structured data
- 04Reviews visible to AI