Sophia at Wellspent <[email protected]>
two AI assistants, mostly different advisors
Hi there,
Thanks for signing up. For the next seven days I'll send one short email a day. Each has one thing we found in our data and one change you can make that day.
Today's finding is the reason for the course. On October 31 we asked ChatGPT and Perplexity to recommend a fee-only advisor in 100 US and Canadian cities, three different ways each, and read all 600 answers.
- Every answer named specific firms, 7.9 on average.
- For the same city, the two assistants agreed on 18% of the firms they named. In three cities they had none in common.
- Asking for a "fee-only financial advisor" and for "the best fee-only financial planners for retirement planning" in the same city, from the same assistant, gave lists that overlapped by 28%.
So there is no single list to get onto. A client describes their situation, every situation is a slightly different question, and the answer is rebuilt each time from whatever the assistant can read about the firms nearby.
Today's change: write down the question your best client would have typed before they found you. Not "financial advisor near me", but their real situation, something like "a flat-fee planner in Raleigh who works with teachers close to retiring". Keep it somewhere handy. We'll use it on day 7.
The full study, with the method and its limits: How AI assistants recommend financial advisors
Tomorrow: what the assistants read before they answer.
Sophia