There’s a new question every financial advice firm should be asking, and it’s not “How’s our Google ranking?” It’s this: when someone asks ChatGPT, Perplexity, or Google’s AI Overview to recommend a financial adviser, does your firm get mentioned?
For most of the firms we checked, the answer is no. Not because they’re not good advisers, but because almost nobody in the profession has caught up with how AI search actually works.
What is a GEO score?
GEO stands for Generative Engine Optimisation — the AI-era successor to SEO, but with completely different rules.
Traditional SEO is about ranking. Google crawls your site, weighs up backlinks and keywords, and places you somewhere on a results page. Even at position 8, you’re still visible. Someone might scroll down and click.
AI search doesn’t work like that. When someone asks ChatGPT or Perplexity a question, the AI doesn’t return ten blue links. It picks a handful of sources, reads them, and writes an answer — citing the businesses it trusts enough to mention by name. There’s no “position 8”. You’re either cited, or you don’t exist for that answer at all.
That’s the entire game now: citation, not ranking.
A GEO score measures how likely a website is to get picked when an AI tool is forming an answer. We built a free scanner that checks six things AI tools actually care about:
- Content extractability — can an AI easily pull a clear answer from your page, or is the useful information buried?
- Entity & authority signals — does the AI know who you are, where you’re based, and who’s behind the business?
- Technical & bot access — can AI crawlers even read your site properly?
- Social proof — reviews, testimonials, and case studies with real outcomes
- Query-answer structure — FAQs, clear headings, and content laid out the way people actually ask questions
- E-E-A-T signals — the trust markers Google and AI engines both look for: team pages, credentials, contact details
Each dimension is scored, added up out of 100, and banded from Critical through to Excellent.
How did UK IFA websites actually score?
We ran the scanner across 54 IFA firm websites, sourced from local directory listings across 10 UK cities. The results were not great, but also not catastrophic. They were something more interesting: uniformly mediocre.
Average score: 53.1 out of 100
| Band | Score range | Firms |
|---|---|---|
| Excellent | 80–100 | 0 |
| Good | 60–79 | 13 |
| Fair | 40–59 | 36 |
| Poor | 20–39 | 5 |
| Critical | 0–19 | 0 |
Two thirds of every firm we scanned landed in the same narrow “Fair” band. Nobody scored Excellent. Nobody scored Critical either. The picture isn’t “some firms are invisible and some are fine.” It’s that almost the entire sector is sitting in the same mediocre middle — which means the few firms that fix this properly will stand out by a wide margin.
The top performer was Executive Wealth Services, scoring 79 — just shy of “Excellent” and the only firm to get genuinely close. The lowest was Penneck Wealth at 20, driven almost entirely by having no social proof, no FAQ-style content, and thin extractable content on the page itself.
The real story is in the dimension breakdown
The overall average hides the more useful finding: firms are good at one thing and bad at almost everything else.
| Dimension | Average score | % of maximum |
|---|---|---|
| Content Extractability | 20.4 / 25 | 81% |
| Technical & Bot Access | 11.2 / 20 | 56% |
| E-E-A-T Signals | 2.8 / 5 | 56% |
| Entity & Authority | 9.8 / 20 | 49% |
| Social Proof | 6.0 / 20 | 30% |
| Query-Answer Structure | 2.7 / 10 | 27% |
The standout figure is Content Extractability at 81%. Most IFA sites are reasonably well written — clear language, sensible paragraph lengths, a decent summary near the top. That part of the job, the actual writing, is mostly done well.
Everything past that falls off a cliff.
Query-Answer Structure averaged just 27%.This is the dimension that checks for FAQ sections, FAQ schema markup, and headings written the way people actually phrase questions (“How much does a financial adviser cost?” rather than “Our Fees”). One in five firms scored almost nothing here at all. AI tools are disproportionately drawn to content structured as direct answers to direct questions — and almost nobody is doing it.
Social Proof averaged 30%.Reviews, testimonials with named clients, case studies with measurable outcomes. This is consistently the single biggest gap across the entire sector, and it matters more for AI citation than most firms realise. An AI deciding whether to recommend you is, in effect, asking “can I trust this claim?” Vague claims without evidence don’t get cited. Specific outcomes do.
Entity & Authority sat at 49%.This is the dimension checking whether AI tools can clearly establish who you are: company name, location, named team members with credentials, a linked LinkedIn page. A surprising number of firms are, in AI terms, faceless. They are technically present online but without the structured signals that confirm they’re a real, credible, located business.
Why this matters more for financial advice than most sectors
Financial advice runs almost entirely on trust. Nobody chooses an adviser the way they choose a coffee shop. They want evidence: who are these people, what’s their track record, can I trust what they’re claiming?
That happens to be exactly what AI citation algorithms are also checking for. The dimensions scoring worst across our sample — social proof, named authority, structured trust signals — are not cosmetic SEO details. They’re the same evidence a prospective client would want to see, just formatted in a way an AI can actually read and act on.
The firms scoring well aren’t doing anything exotic. They’re doing the basics that good client communication already requires — clear bios, real testimonials, specific outcomes — and just making sure that content exists in a form AI tools can extract.
What this means in practice
If your firm’s website doesn’t have:
- A visible FAQ section answering the questions clients actually ask
- Testimonials with a real name, role, and specific result attached
- A team page with named advisers and their qualifications
- Basic structured data (schema markup) telling search engines what kind of business you are
There is a reasonable chance an AI tool, asked to recommend a financial adviser in your area, simply won’t mention you. Not because a competitor outspent you on ads, but because there was nothing on your page an AI could confidently cite.
The good news buried in this data: the bar is low right now.With two thirds of the sector clustered in “Fair”, the firms that close even one or two of these gaps will separate themselves from almost everyone else in the profession.
Check your own score
We built the same scanner used for this research as a free tool. It takes about 30 seconds, no sign-up required, and gives you the same six-dimension breakdown shown above for your own site.
If you’re curious how your firm compares to the 54 in this study, that’s the way to find out.
Methodology note:100 firm listings were sourced from local IFA directories across 10 UK cities (10 per city), reducing to 87 unique firm profiles after deduplication. Of these, 62 had a resolvable website link on their directory listing (25 had no website, or linked only to social media, the FCA register, or network pages). 54 of those 62 sites were successfully scanned; 8 failed to fetch due to dead domains, SSL errors, or timeouts on the firm’s own infrastructure. Scores are heuristic indicators of AI citation likelihood, not guarantees of citation.