Why SEO Agencies Can't Do GEO

Every week, I see another SEO agency announcing a new "AI Search Optimisation" or "GEO" offering. They've updated their decks, added a few new slides about ChatGPT and Perplexity, and are positioning their existing content production and link-building services under a new label.

Some of them are doing it in good faith — they genuinely believe GEO is a natural extension of SEO, and that the skills translate. A few are doing it cynically, knowing the label will sell even if the substance doesn't follow. Either way, the CMOs on the receiving end of these pitches are getting a fundamentally misleading picture of what GEO actually is and what it requires.

Let me be direct: GEO is not a new SEO skill. It's a different discipline. The skill sets overlap at the edges, but the core competencies, the strategy logic, the measurement frameworks, and the tactics that actually work are fundamentally different. Here's why — and what the gap means for your brand if you're running GEO through an SEO team.

What SEO Is Actually Good At

SEO at its core is about ranking pages in search engine results. The discipline has evolved considerably over the past two decades, but its fundamental logic remains: optimise pages for crawlability, relevance, and authority signals (primarily backlinks) to rank higher for target keywords. The measurable outcomes are positions, impressions, and organic traffic.

Good SEO agencies are strong at:

• Technical site audits and crawlability optimisation
• Keyword research and content gap analysis
• On-page optimisation (headers, meta, structured data)
• Link acquisition through outreach and content
• Tracking rankings and organic traffic metrics

These skills are genuinely valuable for organic search. The problem is that almost none of them translate meaningfully to AI search optimisation.

What GEO Actually Requires

GEO's goal is for your brand to be recommended by AI models when users ask questions about your category. The mechanism is fundamentally different from keyword ranking. AI models synthesise answers from patterns in their training data — they're not retrieving ranked pages. Getting recommended requires:

• Entity architecture — ensuring AI models have accurate, well-structured knowledge of your brand
• Citation engineering — earning mentions in the authoritative sources that AI models weight heavily (Tier-1 press, analyst reports, academic papers)
• Citation-worthy content — original research and frameworks that other sources reference, creating the citation chain AI models follow
• Multi-platform measurement — tracking citation frequency and velocity across ChatGPT, Perplexity, Claude, and Gemini
• Authority narrative building — strategic positioning of your brand's intellectual territory in a category

The Skills Gap in Detail

The comparison makes the gap clear. The tactics that move SEO metrics (backlink volume, keyword density, content cadence) are either irrelevant or actively counterproductive for GEO. An SEO agency optimising your content for keyword density is spending time on activities that don't affect citation frequency. An SEO agency prioritising link volume over source quality is building the wrong kind of authority.

The "GEO-Enabled SEO" Trap

The most common failure mode is what I call the GEO-enabled SEO trap. The agency rebrands their existing content and link-building services as GEO-optimised. They add some new jargon to the reports. They point to AI overview appearances in Google as GEO results. They continue doing what they've always done — and because some of it loosely correlates with generic visibility, the client doesn't immediately notice it isn't working specifically for AI recommendations.

The tell is in the metrics. Ask your agency directly:

1. What is our current citation frequency across ChatGPT, Perplexity, Claude, and Gemini for our core category queries?
2. How has that frequency changed over the last 90 days?
3. Which specific tactics in our programme are driving citation improvement?
4. What is our Entity Authority Score, and what are the gaps?

If they can't answer questions 1 and 2 specifically and with real data, they're not running a GEO programme. They're running an SEO programme with a GEO label.

What to Look for Instead

The disciplines that actually underpin effective GEO are closer to strategic PR and brand authority building than to SEO. The skill sets that transfer are:

• Analyst relations — Building relationships with Gartner, Forrester, and category-specific analysts. Getting into Wave and Magic Quadrant reports. These are among the highest-weight citation sources for AI models in B2B categories.
• Strategic PR — Earning coverage in Tier-1 publications, not just any publication. Understanding which outlets carry weight in AI training data and how to get cited there.
• Content strategy for citation — Designing content with the goal of being referenced, not just read. Original research, proprietary frameworks, definitional content.
• Entity management — Wikipedia, Wikidata, Google Knowledge Graph, and industry database management. Ensuring consistent, accurate, rich entity representation across all sources AI models draw from.
• AI visibility measurement — Custom tracking of citation frequency across platforms, authority velocity measurement, competitive citation analysis.

The Honest Assessment for CMOs

If your current SEO agency is pitching you GEO services, the right response is not to reflexively reject it — some SEO agencies are genuinely investing in building these skills. The right response is to ask the questions above, demand measurement, and evaluate based on outcomes.

But if you're currently running what you're calling a GEO programme and you've never received a citation frequency report with actual AI query data, you're almost certainly not running a real GEO programme. You're running an SEO programme that's been renamed.

The opportunity in AI search is real, and the first-mover advantage is significant. Spending that opportunity budget on mislabelled SEO is one of the more expensive mistakes a marketing team can make right now.

Know what you're buying. Demand measurement. The discipline exists — but so does the rebrand.