We make stores readable.

Recommended by Agents is an agency that makes ecommerce stores legible to AI shopping agents such as ChatGPT, Gemini, Claude and Perplexity. Shopping is moving from search results to agent shortlists, and most stores are built for the first and invisible to the second. We exist to fix that.

Run the free check See what we do
The agent's shortlist Carry-on luggage

“What is the best carry-on under $300 with a long warranty?”

Recommended

  1. Brand APrice, delivery and returns read from the feed
  2. Brand BQuoted in two buying guides
  3. Brand CWarranty stated in product markup

Your storeNot on the list: price and warranty could not be read

Example shortlist, for illustration only.

Who you will work with.

A founder-led practice. The person who scopes your audit is on every call, and the developers who write your fixes are named to you before they start.

Harry Harrison, founder of Recommended by Agents

Founder

Harry Harrison

Harry founded Recommended by Agents in 2026 because the way people find products is changing faster than the stores are. Shoppers now ask ChatGPT, Gemini, Muse, Claude and Perplexity what to buy, and the agent answers with a handful of products. Most stores were built for a search results page, not for that.

Harry leads every engagement: the audit, the report and the calls. Fixes arrive as pull requests, written by Harry or by a developer we have worked with before, and you know who that is before they touch your code. Nothing is passed to a junior, and nothing is handed off between the person who scoped the work and the people who do it.

We keep the client list short on purpose, so every store gets senior attention from the first audit to the last pull request.

Harry on LinkedIn

On every engagement

  • Harry runs the audit and writes the report.
  • Fixes arrive as pull requests your developers review.
  • The Index and the weekly prompt battery measure you against your competitors.
  • Harry is on every call.
Founded
2026
Registered in
Wyoming, United States
Clients in
United Kingdom and United States
Services
Audit, Fix, Cite and Track

Three things we hold to.

They decide what we take on, how we do the work and what we publish about ourselves.

  1. The shortlist is the new shelf.

    When a shopper asks an agent what to buy, the agent builds a shortlist of a handful of products. Being on it is worth more than any ranking was. Being off it means the customer never learns you exist.

  2. Readable beats clever.

    Agents reward stores that state plainly what a product is, what it costs, when it arrives and how to send it back. Most of our work is making true things easy to read.

  3. Show your work.

    We publish our methodology, our prices and the prompts we test with. If we ask stores to be legible, we should be too.

    See our prices

Alongside your team, in your codebase.

A small, senior practice. The people who audit your store are the people who fix it, and they work with your developers or your agency, not around them.

Pull request yourstore/storefront

Add Product and Offer schema to the product template

Opened by Recommended by AgentsReviewer: your developers

  • templates/product+42−6
  • feeds/products+18−2
  • docs/product-schema.md+31
  • Markup and feed validated
  • Readiness score run again
  • Waiting for your team's review
Example pull request, for illustration only.
  1. Us

    We audit.

    We test your store the way the agents do. The people who run the audit stay on to make the fixes.

  2. Us

    We open pull requests.

    Changes arrive as pull requests or theme changes, on the platform you already use.

  3. Your developers or agency

    Your team reviews.

    Nothing ships without their review. They know the store, and they keep control of it.

  4. Your team

    You keep it.

    We leave documentation your team can maintain, so the work outlasts the engagement.

Where
UK and US We work with ecommerce brands in both markets.
Platforms
Hosted and headless Hosted platforms and headless builds alike.
Conflicts
One brand per category We do not take on two direct competitors in the same category at the same time.

See what we do

Our clients get no head start.

Every quarter we publish an Index of the brands agents recommend. Our clients are ranked the same way as everyone else.

Questions per category
40
Agents ChatGPT, Gemini, Muse, Claude, Perplexity
5
Runs per question, per agent
5
Markets UK and US, ranked separately
2

The rules we set for ourselves.

Read the methodology See the Index

  • Nobody can pay for a position.
  • Nobody sees the results before they are published.
  • Clients that appear in a category are marked as clients.
  • We publish the full list of questions with each edition.

One partner makes you findable.

Start with the free check, or book a thirty-minute readiness call.

Run the free check Book a call

Want the detail first? Read how we measure.