Does AI Recommend Your Brand? Key Learning from the Ecommerce Expo Event September 2026

AI shopping agents are changing the consumer experience.

Increasingly, customers aren’t just searching Google and clicking through a list of websites. They’re asking ChatGPT, Gemini, Perplexity and other AI tools questions such as:

“Which laptop brand has the best value for money?”
“Find me a winter jacket that meets these criteria.”
“What are the downsides of this purchase?”
“Will I regret buying this?”

The difference is important. Consumers aren’t asking AI to find websites. They’re asking it to make a decision.

So how do you know whether AI is recommending your brand?

GEO isn’t just another version of SEO

At Ecommerce Expo 2026, Rob Stacey from Lomo.ai made a useful distinction:

“GEO is not SEO. GEO starts with a brand. GEO is strategic, not tactical.”

Getting found in AI answers isn’t simply about keywords. It’s about building a trusted, consistent brand that AI systems can understand.

That means connecting brand, content and media. Your website, product pages, pricing information, guides, FAQs and how-to content all contribute to the picture an AI system builds of your business.

So the question isn’t simply: “Does AI know our brand?”

It’s: “What does AI think our brand is known for?”

Four metrics to measure AI recommendation

Rob Stacey’s Expo session highlighted four useful GEO KPIs.

1. Brand mentions

How often does your brand appear in relevant AI answers? Create a set of questions your customers might realistically ask, such as “What are the best [category] brands?” or “What are the best alternatives to Brand X?”. Run them across the AI platforms that matter to your audience and track whether your brand appears.

2. Sentiment

Being mentioned isn’t necessarily a good thing. AI might describe your brand as reliable and good value, or as expensive with mixed reviews. Track the associations that repeatedly appear around your brand: quality, value, innovation, sustainability, reliability and price.

3. Citations

Look at where AI gets its information about you. When sources are provided, record whether they come from your website, retailers, review sites, industry publications, comparison websites, forums or editorial coverage. If AI recommends your brand but relies mainly on third-party sources, that’s useful to know. If your website contains plenty of information but isn’t being cited, that’s an opportunity.

4. Position

Finally, track where you appear in the answer. Are you the first recommendation, the third, or buried among alternatives? Position is useful, but don’t obsess over it. As the Lomo.ai session noted, position matters less than brand fit. Being recommended for the right question is more valuable than appearing first for an irrelevant one.

Build an AI recommendation benchmark

The easiest way to start is to create a fixed set of prompts covering the customer journey:

Discovery: “What are the best [category] brands?”

Comparison: “Brand A vs Brand B — which is better?”

Value: “Which [category] offers the best value for money?”

Objections: “What are the downsides of Brand A?”

Recommendation: “If you had £X to spend, which would you buy?”

Run the same prompts regularly across the AI platforms that matter to your customers.

You’re looking for movement in four areas:

Mentions. Sentiment. Citations. Position.

This gives you something much more useful than simply asking whether ChatGPT “knows” your brand.

And where does direct mail fit?

There’s an interesting connection between GEO and another major theme from Ecommerce Expo: trust.

The session on Monzo and Mums in Marketing highlighted that successful brands don’t simply enter communities and start broadcasting. They listen, research and contribute something useful. Brands should behave as “guests in someone else’s home”. The same principle applies to the customer journey. At Paperplanes, we think physical touchpoints can play an important role in building that relationship — particularly when they’re triggered by customer behaviour.

In their 2025 report, Recharge stated that just 60% of subscribers make it to order three, and 50% to order four. Piyush Jain of Loop Subscriptions recommended educating customers early, celebrating milestones and treating cancellation as a conversation. These are precisely the moments where direct mail can make a difference. A welcome piece can educate a new customer before their second order. A milestone card can recognise a customer reaching order three. A triggered piece of mail can give a lapsed customer a physical reason to reconsider.

The point isn’t to replace digital marketing.

It’s to use digital data to decide when something physical should arrive through the letterbox.

The bigger picture

AI recommendation is still evolving, but one principle is already clear:trust matters.

AI needs consistent information to understand what your brand is, what it stands for and when it is relevant. Customers need confidence to choose you. So start measuring your AI visibility. Track mentions, sentiment, citations and position.

But don’t stop there. Ask what those measurements tell you about your brand — and how you can build trust throughout the wider customer journey. Because if AI is becoming part of how customers decide what to buy, the brands that win won’t necessarily be the ones that shout the loudest. They’ll be the ones that are trusted, understood and consistently useful — wherever the customer encounters them.

Sometimes that’s on a screen.

Sometimes, it arrives through the letterbox.

Want to put your customer data in the letterbox?

At Paperplanes, we help ecommerce brands turn customer data and behaviour into relevant, timely direct mail — from welcome and onboarding through to milestone, retention and win-back campaigns.

Visit the Paperplanes website to discover what you could trigger next.