How Kolsquare Doubled Its AI Citation Rate and Became Spain's Most-Cited Influencer Marketing Brand

For most SEO teams, that's a fire alarm. For Kolsquare, it wasn't and the reason is the whole point of this story.

The conversion events Nicolas tracked in GA4 did not fall.

Fewer clicks, same conversions. The traffic that remained was simply better: more qualified, closer to intent, more likely to turn into a real opportunity. Two things explain that resilience:

1. Kolsquare started on GEO early. While competitors were still treating AI search as a curiosity, Nicolas was already monitoring brand mentions and source citations across LLMs —and acting on what he found.
2. The strategy shifted from volume to visibility-where-it-converts. Instead of mourning lost clicks, Kolsquare doubled down on being recommended at the exact moment a buyer asks an AI model for help.

The old SEO question was "how much traffic can we bring in?"

The new one Nicolas asks is: "When someone asks ChatGPT how to find the right influencers, does our name come up and does the model speak well of us?"

Results at a Glance

On the Spanish market, where Kolsquare leaned hardest into the GEO playbook, the numbers over 7 months speak for themselves:

The same questions that barely surfaced Kolsquare in the beginning now return it as the reference brand, and the domain most frequently cited as a source.

Here are the results for Kolsquare at the beginning of october 7th: And the results on the 9th of June:

What Kolsquare Did with Qwairy

Track 1 - Reverse-engineering what LLMs cite

Nicolas's core workflow starts with a simple question: on our most strategic prompts, what sources are the models actually pulling from?

Using Qwairy, he breaks down high-value clusters: "best influencer-marketing platforms," "top creators in the UK / France," and looks at which pages and domains the LLMs cite (competitors like Modash, aggregator round-ups, even TikTok and Instagram listicles).

From there it becomes a content-gap analysis: Why is that page cited and not ours? Is it an intent mismatch? A length problem? Content we simply don't have yet?

This is the old SEO playbook applied to LLM SEO.

Those answers feed concrete briefs to the content team.

One article created off the back of this analysis began generating citations in ChatGPT. Exactly the outcome the whole approach is designed to produce: stop being absent from the answer, and become the source. Here is the example of the Citation Sources on Qwairy:

Track 2 — Content built for bots, written by experts

Two principles guide the content itself:

• Structure for machines. Structured data, clear bullet points, clean and recognizable markup formats that LLMs and crawlers can parse and reuse with confidence.

• Real expertise, not filler. Kolsquare invests in genuine experts to write, rather than junior content at volume. The logic is straightforwardly E-E-A-T: if a model searches the web and lands on a credible, expert-authored Kolsquare page, that's what justifies the citation.

It's a budget trade-off : expert content costs more than a junior writer but it pays off in authority.

Track 3 - Tracking the right AI models

When Nicolas started, the AI landscape that mattered was essentially Perplexity and ChatGPT. He's since added Gemini to his tracking, and is weighing Claude as the next one to include.

This is exactly where a tool earns its place: the AI answer ecosystem isn't transparent, it shifts constantly, and you can't optimize for what you can't see.

Qwairy gives Kolsquare a single, comparable view across models and the ability to spot when a new engine starts driving real visibility.

Track 4 - Brand perception and repositioning

This is where GEO stops being a traffic channel and becomes reputation management.

Using Qwairy's **Brand Perception **feature, Nicolas audited how Kolsquare is described across LLMs (starting with the UK market).

The output confirmed hypotheses the team had been debating internally: perceived pricing concerns, an image that read as "very European," and topics the brand simply wasn't talking about enough.

Seeing it reflected back from the models was the push to be seen differently. Concretely:

• A dedicated AI page crawlable by Google, surfaced discreetly in the footer that spells out who Kolsquare is, how to talk about them, the right terms to use, and crucially what they are not (not an agency, not an influencer-marketing marketplace).

• Segmented positioning pages rebuilt to address each audience in its own language: agencies, mid-market companies, and large enterprises.

Track 5 - A market-by-market Reddit strategy

Reddit is increasingly cited by AI models but you can't barge in as a brand account shouting that you're the best.

Kolsquare's approach (run with an agency, overseen by Nicolas) is to participate authentically across multiple accounts and specific threads, feeding genuinely useful discussion so it gets seen, relayed, and picked up and, here and there, surfacing the Kolsquare name as the premium reference.

The key lesson: this is deeply market-specific.

The Reddit play works in the UK today and is emerging in Spain, where the platform is heavily cited.

You can't run the same playbook in France as in Spain, and you can't brute-force it with raw translation; the expressions, the platforms, and the way people search all change by country and by segment. 💡 Further read : Our complete Reddit strategy for SEO and GEO.

The Bigger Picture: GEO Becomes a Cross-Functional Discipline

The most forward-looking shift at Kolsquare is organizational. If the goal is for AI models to recommend you positively, then SEO can no longer operate in its own corner.

Reputation signals like customer reviews directly influence what models say about a brand.

So Nicolas's GEO strategy now reaches into Sales and Customer Success: when customers are happy, the customer-relationship team is mobilized to capture reviews that feed back into how Kolsquare is perceived.

GEO becomes a central, coordinating function rather than a siloed channel.

What Made This Work

For other brands and consultants looking to replicate Kolsquare's results, the actionable takeaways:

1. Don't panic over click loss, check your conversions. If conversions hold while clicks fall, your remaining traffic is higher quality. Optimize for visibility where it converts, not raw volume.
2. Reverse-engineer the sources. Find what LLMs cite on your priority prompts, diagnose the gap versus competitors, and create or rewrite content to become the cited source.
3. Structure content for machines, and back it with real expertise. Structured data and clean markup make you parseable; expert authorship makes you citable.
4. Track multiple models and add new ones early. The landscape isn't transparent and it moves fast. Adding Gemini early (and watching Claude) is how you stay ahead.
5. Audit your brand perception, then fix the narrative. Use perception data to align Brand, SEO, and GEO. Publish a dedicated AI page that states who you are and who you're not.
6. Make GEO cross-functional. Coordinate with Sales and Customer Success to generate the reviews and reputation signals that shape how AI describes you.
7. Treat every market as its own project. No brute translation. Build market- and segment-specific prompt sets and tactics.

About Qwairy

Qwairy is the Generative Engine Optimization platform that helps brands measure and improve their visibility across AI answer engines ChatGPT, Claude, Perplexity, Gemini, Copilot, Grok, and more. From brand mentions and source citations to sentiment and competitive share of voice, Qwairy turns the opaque world of AI search into something you can monitor, act on, and win. Want to see how your brand shows up in AI search? Get a demo or start free.