SeekLab
SEO & GEO Tool
Most agency case studies look the same. Metric goes up. Client is happy. Everyone looks good.
This isn't that kind of case study.
What follows is an honest breakdown of six months of GEO work on a Web3 wallet client — active across Asian markets, solid product, real user base. Citations grew significantly. Then they declined. Cited pages kept growing. Mentions barely moved. And every one of those patterns taught us something we didn't fully expect going in.
The Starting Point
When we started in November 2025, our client had near-zero presence in AI-generated answers. Ask ChatGPT about multi-chain wallets — they weren't mentioned. The product was solid. The AI visibility wasn't there.
Six months later, here's what the data actually shows:
MetricCurrent (Apr 2026)TrendAI Citations277-6.1% (declining from March peak)Mentions56-6.7% (declining)Cited Pages61+10.9% (still growing)Active AI platforms3ChatGPT, Google AI Mode, Gemini
That's a more complicated picture than most agencies would choose to publish. We're publishing it anyway because the complications are where the learning is.
What Worked: Citations Grew Significantly
From November 2025 to March 2026, AI citations grew from near-zero to approximately 300.
Three tactics drove most of that growth:
Comparison content "Client vs [competitor]" articles and on-site comparison tables were cited far more frequently than any other content type. If you want to appear in "best X for Y" answers, you need to exist in comparison contexts first. This was the single highest-leverage tactic in the campaign.
Reddit participation Consistent, expert-level contributions in relevant subreddits produced a clearer correlation with citation growth than almost any other off-site activity. Reddit's conversational format is heavily weighted by AI retrieval systems for recommendation queries.
The rule that worked: 90% genuinely useful contributions, 10% brand mention only when directly and honestly relevant.
Publishing velocity Maintaining a consistent weekly publishing cadence across Medium, dev.to, Hashnode, HackerNoon, and similar platforms created sustained signals that AI retrieval systems respond to over time. The first six weeks showed minimal change. The compound effect kicked in around week 8.
What Surprised Us: Citations Peak, Then Pull Back
By March 2026 the client had reached approximately 300 AI citations — the peak. Since then, citations declined to 277 (-6.1%). Mentions followed the same pattern (-6.7%).
AI citations are not sticky in the way traditional search rankings are.
A page that ranks #1 on Google tends to stay there unless something actively displaces it. AI citations work differently — they reflect the current state of the web ecosystem around a brand. When activity slows, citations follow.
The practical implication: AI visibility requires maintenance, not just an initial build.
What's Still Growing: Cited Pages
The one metric that kept moving in the right direction — still growing at +10.9% — is cited pages.
This matters more than it appears. Cited pages represent the structural foundation of AI visibility: the number of distinct pages AI systems consider credible enough to reference. When citations fluctuate, cited pages is the better indicator of underlying visibility health.
This reframed how SeekLab thinks about GEO success metrics:
Citations = leading indicator. Responds quickly to activity, pulls back quickly when activity slows.
Cited pages = lagging indicator. Grows slowly but represents more durable AI visibility infrastructure.
What Barely Moved: Mentions
Mentions stayed relatively flat across the entire six months, ending at 56 with a slight decline.
Our current hypothesis: mentions in AI-generated content are more heavily influenced by brand recognition and training data than by structured content and off-site signals. For a brand operating primarily in Asian markets, breaking into the mention layer of Western AI systems requires either more time or a different set of tactics — possibly more focus on high-DA English-language coverage.
This is an open question we're still working through.
The Honest Summary
Six months of structured GEO work produced real results — the client went from invisible to actively cited across ChatGPT, Google AI Mode, and Gemini.
But the data also revealed something the industry doesn't talk about enough: AI visibility is more like a living ecosystem than a ranking. It requires ongoing care, not just initial optimization.
Three things we'd tell any brand starting a GEO campaign now:
Start with comparison content. Fastest path to appearing in recommendation queries.
Build cited pages, not just citations. The page count is the foundation everything else grows from.
Plan for maintenance from day one. If your strategy doesn't include ongoing publishing and community activity, your citation peak will be temporary.
Want to know where your brand currently stands in AI-generated answers? Open ChatGPT, Perplexity, and Gemini. Search your brand name and core category queries. Record what comes back.
After auditing websites across SaaS, FMCG, and Web3 clients, one pattern holds: most teams either skip technical SEO checks entirely or run audits they don't know how to read. SeekLab's free audit tool was built to close both gaps — it runs in under five minutes and produces a prioritised report any team member can act on.
Here's exactly how to use it, what you'll see, and what to do next.
What the SeekLab Audit Tool Does
The SeekLab Audit Tool scans your website for technical SEO issues and surfaces them in a single structured report. It crawls up to 100 pages per scan, assigns an overall SEO score out of 100, and groups every issue into three priority levels: Critical, Warning, and Notice.
In SeekLab's own test scan, the tool completed a 100-page crawl and returned a score of 77/100 — with one Critical issue, a cluster of Warnings, and several Notice-level items spread across template and page-level problems.
How to Run Your Audit
Go to seeklab.io/audit
Enter your domain in the "Enter your domain to audit" field
Click to start the scan
Wait approximately 5 minutes for the crawl to complete
When the report status shows Success, your results are ready
That's the full workflow. No setup, no account required to start.
How to Read the Report
The report organises every issue into three tiers, so you always know what to fix first.
Critical These are issues that directly harm crawlability, indexing, or user experience. Fix these before anything else. In the SeekLab test scan, this tier flagged one duplicate meta description.
Warning These are issues that reduce ranking potential but won't break the site. Common examples from the SeekLab scan included:
Missing Alt Text: 10 instances
Title Too Long: 2 instances
Missing Meta Descriptions: 3 pages
Missing H2 Tags: 14 pages
Meta Description Too Long: 8 pages
Notice These are lower-priority patterns worth tracking over time. The SeekLab scan returned notices for duplicate H1/title tags (40 instances), nofollow external links (90), and a small number of redirect and anchor text issues.
One practical rule from SeekLab's consulting work: when Warning or Notice issues appear repeatedly across many pages, they're almost always template-level problems. Fix the template once rather than editing each page individually.
What to Do After the Report Is Ready
Three actions are available directly from the report:
Copy Link — saves the report URL so you can share it or bookmark it
Export — downloads the full report for your records or team review
Contact Us — connects you with SeekLab's team if you need help turning findings into an action plan
SeekLab provides paid consulting for teams that need structured support moving from audit output to resolved issues.
When to Run an Audit
Run the SeekLab audit:
After any significant website update or migration
Before launching a new content push
When traffic drops unexpectedly
Quarterly as a routine health check
Exporting the report each time gives your team a clear before/after record across audits.
The One Thing Most Teams Miss
Running the audit is easy. The common failure point is letting the report sit unread. The most valuable use of the tool is to focus the Critical tier first, then batch the Warning items by type — because grouped fixes at the template level take a fraction of the time that page-by-page edits do.
A website that gets audited and acted on outperforms one with a perfect-looking homepage every time.
Want to run your first scan? Visit seeklab.io/audit — it's free, no signup required, and results are ready in about five minutes.
For teams needing deeper analysis or help prioritising fixes, seeklab.io offers full-scope technical SEO consulting across website structure, content, and AI search visibility.
In March 2026, researchers simulated a complete fruit fly brain — around 130,000 neurons, 50 million synapses — placed it in a virtual body, and watched it walk, groom, and feed without any explicit task training.
No reinforcement learning. No behavior trees. No supervised training on labeled datasets.
Just structure, feedback, and an environment to react to.
The SEO lesson hit immediately: most content strategies fail for exactly the same reason most AI systems fail — not because of bad inputs, but because of bad architecture.
Here's what a fruit fly can teach you about building a website that actually performs.
Structure beats scale — always
The mainstream assumption in both AI and content strategy is that more is better. More compute, more data, more parameters. More pages, more categories, more blog posts, more translated copies.
The fruit fly simulation challenges that directly. A system with 130,000 neurons, organized correctly, produced coherent behavior. A website with 5,000 pages, organized poorly, produces confusion.
In SeekLab's audit work, we see this consistently: sites that struggle aren't usually thin on content. They're thin on structure. Dozens of blog posts that overlap, location pages with barely differentiated copy, internal links that come mostly from menus and footers rather than contextual pathways.
That's not scale. That's noise.
Think in circuits, not pages
The fruit fly brain works in loops. A leg movement changes position. Position changes sensory input. Sensory input alters neural activity. That closed-loop feedback is what produces behavior.
A well-structured website works the same way:
Discovery circuit: educational article → solution page → proof material → contact form
Validation circuit: industry page → technical detail → FAQ → inquiry point
Recovery circuit: related links, breadcrumbs, and hub pages that help users change course without bouncing
If your site doesn't have these circuits, it doesn't matter how many pages you have. Users — and search engines — will bounce.
What a structurally intelligent site looks like
Brute-force site patterns:
Many near-duplicate pages
Weak contextual internal linking
Broad publishing with little depth
Confusing path to conversion
Translation everywhere, maintenance nowhere
Structurally intelligent site patterns:
Fewer but clearer topic clusters
Purposeful links between related intent stages
Modular content built around real tasks
Clear user journey from discovery to inquiry
Selective localization with strong architecture
The second pattern is what SeekLab.io builds toward in every audit. Not more pages — better information architecture, stronger semantic relationships, and fewer dead ends.
The 5-circuit check for your own site
Use these five questions as a diagnostic:
Does each important topic cluster support a real task?
Are internal links behaving like circuits — or just template links?
Are you localizing strategically rather than mechanically?
Are technical issues cutting key behavioral paths?
Are you publishing because the strategy is clear, or because the calendar says so?
If any of these reveals a gap, that's where the work starts. Not another content sprint — a structural fix.
SeekLab's technical SEO roadmap for early-stage growth and high-quality blog content optimization guide are directly relevant here.
A fruit fly with 130,000 neurons outperforms a website with 5,000 pages when the wiring is better. Build for coherence first. Then grow.
References
A lot of startup content is still written for one outcome: ranking.
That still matters, obviously, but it feels like a lot of teams are missing a second layer now.
A page can rank on Google and still never show up inside an AI-generated answer.
For a bigger brand, that may not feel urgent yet. For a smaller startup, I think it matters more.
Smaller teams usually do not have much margin for wasted visibility. If your best pages are technically ranking but never getting picked up, summarized, or cited in AI search flows, you may be less visible than your SEO dashboard suggests.
This is one of the visibility shifts we’ve been paying closer attention to at SeekLab lately. What seems to matter more now is not just whether a page can rank, but whether it is easy to use:
clear answer near the top
simple structure
consistent terminology
real FAQs
less filler
content that is easy to extract from
I’m starting to think more founders should treat this as a content design problem, not just a keyword problem.
We wrote a deeper breakdown here if useful:https://seeklab.io/blog/from-seo-to-geo-adapting-content-for-ai-search/
Comment