Intrigue DigitalHow Does ChatGPT Choose Which Finance Broker to Recommend?
The short answer
An AI assistant names the broker it can most confidently identify as a real, licensed business, whose details are consistent everywhere it looks, and whose content clearly answers the question being asked. Confidence is the whole game: a model would rather name a broker it is sure about than one it half-recognises.
You influence that confidence by controlling three inputs: how clearly you are defined, how well you answer, and how consistently the rest of the web backs you up.
Identity and consistency
First, the model tries to work out exactly who you are. A brokerage with structured data stating its name, licensing, service areas and specialties, and with matching details across its Google Business Profile, MFAA or FBAA listing and aggregator profile, is easy to identify. A brokerage with a different phone number or business name in three places is not, and the model hedges.
This is why consistency across the directories AI reads matters so much. Every contradiction is a reason for the model to leave you out.
Relevance and trust
Next, the model matches the question to your content. If someone asks about refinancing an investment loan and your site has a clear, answer-first page on exactly that, you are a strong candidate. If your site only has a generic home page, you are not.
Finally, it weighs trust: genuine reviews, mentions in finance or local press, and a credible, named broker behind the business. These off-site signals are what tip a model from listing options to naming yours. The full build is in AEO for finance brokers.
What the research says
Independent studies on search and AI behaviour, last reviewed October 2026. We re-check these figures at least every six months.
| Measure | Before | After |
|---|---|---|
| Organic CTR on queries with an AI Overview | 1.76% | 0.61% |
| Paid CTR on queries with an AI Overview | 19.7% | 6.34% |
| Extra organic clicks for brands cited in the Overview | +35% | |
| Extra paid clicks for brands cited in the Overview | +91% |
Source: Seer Interactive, AI Overviews impact on Google CTR, September 2025 update. 3,119 informational queries across 42 organisations, June 2024 to September 2025. Seer notes citation may reflect, not cause, stronger brands.
| Platform or forecast | Figure | Date |
|---|---|---|
| ChatGPT weekly active users | 800 million | October 2025 |
| Google AI Overviews monthly users | 2 billion | July 2025 |
| Gartner forecast for traditional search volume | 25% lower by 2026 | February 2024 |
Source: TechCrunch, ChatGPT hits 800 million weekly active users (October 2025); Search Engine Journal, AI Overviews reach 2 billion monthly users (Q2 2025 earnings); Gartner, Search engine volume forecast (February 2024). Gartner's figure is a forecast, not a measurement.
Sources and further reading
Frequently asked questions
Does ChatGPT use Google to pick a broker?
AI assistants draw on a mix of their training, live web results and trusted sources. In practice that means your Google presence, directory listings and reviews all feed what the model knows about you.
Can I pay to be recommended by ChatGPT?
Not for the organic recommendation. You earn that by being the clearest, most trusted and most relevant broker for the question. Paid AI ad formats are separate from being named in a normal answer.
Why does AI recommend my competitor and not me?
Usually because the competitor is easier to identify and trust: clearer structured data, consistent details across directories, better reviews, or content that answers the question more directly. Each of those is fixable.
Want to be the brand AI name-drops?
Intrigue Digital does SEO, AEO and GEO for agencies and businesses. One business per industry, per postcode, so your competitor cannot take your spot.
Book a free strategy call →Some links in this article are affiliate links to tools we use and trust. If you sign up through them we may earn a commission, at no extra cost to you. We only recommend tools we actually use.