How GEO helped SteelSeries grow AI search conversions 3.2x in 6 months

SteelSeries grew AI search conversions 3.2x in 6 months and became the most retrieved brand for "gaming headset," "gaming keyboards" and "gaming earbuds" across ChatGPT, Gemini and Perplexity. It got there by fixing the sources AI already cited, cleaning up outdated product mentions, and rewriting product pages so models could quote them.

The challenge

SteelSeries makes award-winning gaming headsets, keyboards and mice. Its audience is tech-savvy gamers, and those gamers had started asking LLMs which gear to buy instead of scrolling a page of links.

That shift hurt twice. Organic traffic declined as zero-click answers replaced visits. And when gamers asked high-intent questions like "best gaming headsets" or "best gaming keyboards," SteelSeries was not consistently named. Its flagship headset lines were often missing, and competitor products surfaced instead.

The team set three goals:

1. Win prime placement for queries like "best gaming headset" and "best gaming keyboards."
2. Raise brand preference with AI-driven gamers by making unique product features visible to the models.
3. Track leads from AI answers through to product pages, so the work could be tied to conversions.

What they changed

The program ran in six steps.

1. Found the sources that mattered. The team mapped which sources AI models cited in gaming discussions. A small group of review sites and Reddit threads accounted for a large share of headset-related citations. That short list became the outreach target.

2. Replaced stale mentions. Outreach to those sources secured current product mentions, replacing outdated information the models had been citing.

3. Corrected perception and sentiment. Monitoring surfaced negative and outdated references to older headset models. The team updated content, fixed structured data and refreshed reviews so low-scoring mentions gave way to accurate, positive ones.

4. Fixed the technical layer. Product pages got improved structured data, schema and robots.txt rules. The team also added an llms.txt file so AI agents could parse and surface current product details.

5. Rewrote product pages for extraction. Key pages gained Q&A sections, bullet-point benefits and citation-ready phrasing such as "best wireless gaming headset for PC and Xbox." Copy was aligned to the exact queries gamers were asking.

6. Connected citations to revenue. Traffic and conversions were attributed to specific product citations, so the team could see which answers drove sales.

The results

SteelSeries became the #1 most retrieved brand in AI answers for gaming headsets, keyboards and earbuds, with consistent placement across AI search.

• 3.2x more AI search conversions in 6 months.
• Perplexity visibility score rose from 44 to 77, a 75% improvement.
• ChatGPT visibility score rose from 51 to 73, a 43% improvement.
• Gemini visibility score rose from 62 to 85, a 37% improvement.

AI search traffic also outperformed traditional search on conversions, which fed significant revenue growth.

What you can copy

Start with the citation map, not your own blog. SteelSeries learned that a handful of review sites and Reddit threads drove most headset citations. Find the few sources that shape answers in your category and focus there first.

Treat outdated mentions as a visibility problem. Models kept citing old product information and old models. Getting current details onto the pages AI already trusts replaced that stale content in the answer.

Write product pages that can be quoted. Q&A blocks, bulleted benefits and plain phrases that match real queries give a model a clean sentence to lift. Vague brand copy gives it nothing.

Clear the technical path. Structured data, sensible robots.txt rules and an llms.txt file make it easy for AI crawlers to read the current version of your catalog.

Measure conversions, not just mentions. Visibility scores show progress. Attribution from AI answers to product pages is what proved the program paid off.