It’s the end of another week, and I want to leave you with something worth carrying into your Monday planning. Not a prediction. Not a trend piece. Actual data from financial institution websites — 316 of them — audited for AI visibility.
I’ve been writing about generative engine optimization (GEO) and answer engine optimization (AEO) for a while now. I know the topic can feel abstract. “AI search is growing.” “Buyer behavior is shifting.” It’s everywhere. But abstraction is a good reason to set something aside, and this is one topic you don’t want to set aside. So before you close the laptop today, let me make it concrete.
What 316 audits actually show
The benchmarking data compiled by Gleam Strategies comes from 316 real AI visibility evaluations of financial institution websites. These aren’t simulations or projections. They’re measurements.
The headline finding: 27% of financial institution websites score below 50 on AI/GEO discoverability, putting them in a range where AI answer engines — ChatGPT, Perplexity, Google AI Overviews — effectively can’t surface them in response to relevant queries. Not underperforming. Not partially visible. Invisible.
Only 15% of audited sites pass what we define as high-reliability standards: the threshold at which an AI system has enough structured, corroborated, extractable content to confidently recommend or cite your institution. The average GEO/AEO score across all 316 audited sites is 54 — just above the failing threshold, firmly in “needs work” territory. And 51% of sites score below 60 across all visibility categories combined, meaning most financial institution websites aren’t well-positioned on any of the core dimensions AI systems use to evaluate content.
These aren’t outliers. This is the norm.
The gap you might not expect
One finding surprises most people: the category with the highest failure rate isn’t GEO/AEO. It’s Accessibility.
Fifty-three percent of financial institution websites score below 50 on ADA/WCAG compliance. The average Accessibility score across all 316 sites is 47 — lower than any other category we track, lower than SEO (avg 50), lower than Reputation (avg 55), lower than GEO/AEO itself (avg 54). That’s a legal exposure story, but it’s also an AI visibility story that doesn’t get told enough.
Here’s why it matters for GEO and AEO: the same infrastructure that makes a website readable by assistive technologies — semantic HTML, properly labeled elements, structured organization — is what makes a website parseable by large language models. When your content requires human interpretation to navigate, an AI system will struggle to extract it, which reduces how often you show up in AI-generated answers.
The most commonly prescribed fix across all 316 audits is implementing structured data — recommended for 79 of the sites evaluated. Schema markup, the tags that tell AI and search systems exactly what your content contains, what your institution offers, who it serves, is the single highest-leverage action most financial institutions haven’t taken yet.
The window is still open
GEO and AEO are not yet crowded in financial services. Most institutions are not optimizing for this layer in any systematic way, which means the bar to stand out is still relatively low. A site that implements structured data, addresses its accessibility gaps, and builds a baseline of third-party validation has a real opportunity to show up in AI-generated recommendations before competitors get there.
But “still open” isn’t a permanent condition. The institutions that build this infrastructure now will have a structural advantage that compounds. The ones that wait will pay a higher cost to close a much larger gap.
If you want to understand exactly where your institution stands, Gleam Strategies built a platform for precisely this. It audits your site across GEO/AEO, SEO, Accessibility, Technical performance, and Reputation, benchmarks you against the 316 institutions in our dataset, and prioritizes the specific actions most likely to move your score. Learn more at gleamstrategies.com.
The bottom line
The shift to AI-mediated discovery isn’t a theoretical risk for financial institutions. It’s a current-state measurement problem with a known size and clear starting points. The data exists. The gaps are measurable. The actions are ranked.
That’s a good thing to know heading into the weekend.
Kevin Farley writes about AI visibility, GEO & AEO strategy, and digital discoverability for financial institutions. Read more at kevinfarley.org.