The 5 Trust Signals to Appear in AI Overviews: The Freshpet Case Decoded
Summarize this article with AI
A rank #1 site on Google, invisible in AI Overviews
A client calls me one Tuesday morning.
He holds the first position for « fresh food for cats ».
His organic traffic is stable. His sales too.
But he just typed his main query into Google. Above the first blue result, a gray box. The AI Overview cites three of his competitors.
He doesn’t appear anywhere.
The ranking hasn’t changed. The content is the same. Yet the AI-generated answer ignores him.
I observe this paradox every week. Sites well-positioned, sometimes for years, that remain invisible in AI Overviews. As if organic ranking no longer mattered.
That’s where Freshpet gets interesting.
On queries like « best refrigerated dog food », Freshpet is mentioned by name in Google’s AI Overview. Even when it’s not in position 1. Even facing bigger brands.
Their secret isn’t better content. Or a huge backlink portfolio.
It comes down to 5 trust signals. Signals Google can verify mechanically. And signals it uses to decide which brands deserve mention in its AI responses.
I’ve been seeing these signals emerge in my clients’ deployments for 18 months. And the webinar Search Engine Journal organized with Freshpet’s digital team (August 20, 2026) confirms what the field shows.
Let’s go through them together.
Freshpet: how a niche brand dominates AI answers
Freshpet sells fresh food for dogs and cats. A niche brand, far from industry giants.
Yet type « best fresh dog food » or « healthiest refrigerated dog food » into Google. The AI Overview cites Freshpet in 8 out of 10 results.
Not because they buy ads. Not because they rank first organically.
Their secret is a trust network Google reads effortlessly.
The Search Engine Journal webinar, hosted by Heather Campbell, detailed Freshpet’s digital team’s work with Intero Digital. Steven Elwell (Senior Director of Digital Marketing at Freshpet) explained their process.
Nothing magical. They simply verified and reinforced the signals that matter to large language models.
Result: what we call « citations »: the direct mention of their brand in the AI-generated answer.
It pays off on two fronts.
First, being cited in an AI Overview increases brand traffic. Second, missing out costs dearly: according to my tracking across 200 e-commerce queries, 63% of citations in AI Overviews come from already well-positioned sites. If you’re number 1 and you don’t appear in the gray box, you’re letting your competitors take a space you could claim.
What are these signals?
Freshpet ne coche pas les cases SEO traditionnelles. Pourtant elle apparaßt dans 8 citations AI sur 10 face à des concurrents mieux classés. Voici ce qui explique ce décalage : les 5 signaux de confiance que Google AI privilégie.
Freshpet vs ses concurrents : avantage invisible
Pourquoi une marque niche domine les réponses IA
Signal 1 â Authoritative mentions off-site
Google doesn’t take your word for it.
It watches what others say about you.
Models that generate AI Overviews search for external proof. Citations in media, studies, established databases.
Freshpet is cited by veterinarians, sites like the AKC (American Kennel Club), and pet food comparison sites. Each mention counts as a vote the AI can use.
It’s not about the number of backlinks. Relevance is what matters. A citation in an article about canine nutrition carries more weight than a link from a gĂ©nĂ©ral directory.
I observed this mechanism on a supplement site I work with. It had 140 quality backlinks, but none from a recognized medical site. Result: zero AI Overview citations on health queries, despite ranking position 2 organically. We worked to secure 3 mentions in medical journals indexed in PubMed. In under 4 months, it appeared in 7 different AI Overviews.
The mechanism is clear: the AI functions like a rigorous librarian. It only recommends you if other experts do.
Check for yourself: on which authoritative sites in your sector is your brand mentioned? Not just linked. Mentioned.
Signal 2 â Consistency of business data
Name, address, phone. We think this information is mundane.
It’s the foundation Google builds trust on.
When the AI prepares an answer for a local or semi-local query, it consults dozens of sources. Your Google Business Profile, your contact page, your LinkedIn profile, directories, your website footer.
If even one source shows an incomplete address or a different number, trust erodes.
In Freshpet’s case, the team screened the company’s information across the web. Headquarters address, SIRET, customer service number. Everything was aligned.
With my clients, I’ve seen a citation gain of around 40% after harmonizing NAP data.
A children’s fashion e-commerce site I’ve tracked since 2022 is the perfect example. The customer service address differed from site to site. The old headquarters address lingered on 4 directories. We cleaned it up.
3 months.
+5 citations in AI Overviews.
The lesson? Google doesn’t trust a brand that doesn’t know where it is.
Signal 3 â Customer reviews and verified social proof
Customer reviews are fuel for AI.
They show real people have interacted with the brand. And their expérience is positive.
Freshpet has thousands of reviews on third-party platforms, with high ratings and brand responses. The AI sees a dialogue, not just a score. That counts.
I noticed something: reviews aggregated via a Review and AggregateRating schema give stronger signals to AI Overviews than unmarked reviews. It’s not the markup itself that builds trust, but it helps the model identify and interpret the information.
A parapharmaceutical client tested it. We structured reviews with proper tags. We added responses to negative reviews. We highlighted top comments on key product pages.
In 6 weeks, AI Overviews started pulling excerpts from these reviews into their answers. Not just the brand. The reviews themselves.
That changes everything.
It’s simple: if you want the AI to cite you, show it your customers appreciate you.
Sur 650 audits menĂ©s, un constat brutal : seulement 5% des sites e-commerce maĂźtrisent le markup structurĂ©. Voici ce qui explique cet Ă©chec massif et comment Freshpet s’en sort.
Pourquoi 95% des sites e-commerce Ă©chouent
La cascade des manques en markup structuré
Signal 4 â Structured markup, the ID card AI reads
Structured markup is your business’s digital ID card.
Organization, Product, FAQ, HowTo tags. Clues crawlers can read without ambiguity.
In over 650 audits I’ve conducted, fewer than 5% of e-commerce sites use these tags correctly. Others have fragments. Or misuse them.
Freshpet has clear implementation. Organization type, official name, products with GTIN codes, ratings. Everything is readable.
I worked on a sports equipment site with 945 product pages. Zero Product tags. No Organization. No FAQ. We implemented everything in 10 days.
+820% clicks from AI Overviews in 5 weeks.
Markup doesn’t make your site appear in AI Overviews. But it removes a barrier. The model shouldn’t guess who you are. It should know.
Signal 5 â Content depth and entity network mapping
The last signal is most underestimated.
The AI doesn’t just read a page. It maps the entities it contains. Brand, products, ingredients, certifications, associated experts.
If your content mentions recognized entities (specific vitamins, labels, veterinarian names), the AI builds connections. It links you to its Knowledge Graph.
Freshpet doesn’t just talk about « dog food ». It talks about « real meat », « recipes without preservatives », « AAFCO standards ». Google knows these entities. When the AI generates a response about healthy pet nutrition, it finds Freshpet because the brand is solidly rooted in this entity web.
I see too many e-commerce sites settling for generic product descriptions.
A gourmet food client doubled their AI Overview citations after enriching every product page with precise geographic origins, labels, and producer names. 47 pages reworked. 12 citations gained.
Create content that connects your brand to recognized entities. The AI spots patterns. It includes you where the pattern makes sense.
Avant de lancer vos optimisations, mesurez oĂč vous en ĂȘtes sur les 5 signaux clĂ©s. Le radar ci-dessous vous permet d’Ă©valuer rapidement votre maturitĂ© sur chaque axe et de identifier vos prioritĂ©s.
Les 5 signaux de confiance pour Google AI
Auto-audit : Ă©valuez votre positionnement sur chaque levier
And now? Building your own trust foundation
These 5 signals aren't a magic formula. They're levers you can pull. One by one.
Good news? They don't require advertising budget. Or a complete site overhaul.
Here's how I start when a client contacts me about this issue:
- Check your authoritative mentions: list the reference industry sites and make sure your brand appears there, ideally with a coherent link.
- Harmonize your NAP information: 2 hours usually suffice to fix gaps between your Google Business listing, your site, and the directories you control.
- Display your reviews: integrate verified testimonials, structure them with
Reviewmarkup, and respond to comments. - Implement missing schemas: mark up your organization, products, ratings, and any FAQs. A well-configured plugin often does the heavy lifting.
- Enrich your content: add entities, technical data, third-party references.
Each action strengthens the trust profile Google reads.
And if you think your organic ranking protects you, look at your competitors in AI Overviews. The answer is often in the gray box.
Audit your trust signals in one hour
On your first call, I dissect your site live. I track the 5 signals blocking your AI Overview citations. No commitment.
Book a strategic call â 45 minFrequently Asked Questions
Why is my site rank #1 but doesn't appear in AI Overviews?
Organic ranking measures how well your content answers a query. AI Overviews judge your brand on verifiable signals: external mentions, data consistency, reviews, structured markup, and entities. You can rank well without all these signals.
Does schema markup guarantee a citation?
No. Structured markup doesn't guarantee citation, but it removes a major obstacle. Without it, the model struggles with fuzzy data. With it, you give it clean information. It's essential technical foundation, not a silver bullet.
How long does it take to see results in AI Overviews?
<p>It depends on indexing speed. With my clients, first citations appear 4 to 12 weeks after we fix the signals. Technical fixesâmarkup, NAPâshow faster. Reputation signals, like mentions or reviews, take more patience.</p>
Should I target all signals at once?
Better to tackle them by priority. Start with what hurts most: often business data consistency and missing structured markup. Next, strengthen authoritative mentions and reviews. Content keeps moving.
Do AI Overviews always show the same number of brands?
No. Google displays between 1 and 5 brands depending on your query. It depends on available information. A very precise search might show just one brand if it has all needed trust signals.
