How should AI prepare an advisor for a client meeting?
A useful pre-meeting brief gives the advisor the client profile and goals, holdings and performance, recent conversations and commitments, market events that touch the holdings, and suggested topics, all as a draft the advisor reviews and edits before the meeting. Preparation is a real share of the week: a Kitces Research study of more than 1,000 advisors, published in March 2019, found that lead advisors spend 5.3 hours a week preparing for client meetings, out of 43 hours of work.
A pilot should compare prepared and unprepared meetings on time, edits and advisor use, so the firm learns where the brief saves effort and where it still needs correcting.
What does a useful pre-meeting brief contain?
A useful brief holds five elements, and each one needs a known source and a check the advisor can run in minutes. Without a source, the advisor cannot tell a fact from a guess. Market and company data are the usual source for the fourth element; Streetbeat, for example, offers 170+ financial datasets.
| Brief element | Where it comes from | What the advisor checks |
|---|---|---|
| Client profile and goals | CRM profile, risk questionnaire, suitability records | Goals and constraints are current, nothing is missing |
| Holdings and performance | Portfolio and custody systems | Positions and returns match the system of record |
| Recent conversations and commitments | CRM notes, emails, call records | Each commitment is real and still open |
| Market events affecting the holdings | Market and company data feeds | The event touches this client's positions, with a date |
| Suggested topics | Combination of the four elements above | The topics fit the client's goals and the advisor's judgment |
Why should the brief be a draft for the advisor to review?
The advisor should review it because even an accurate assistant is sometimes wrong, and the brief sets the agenda for a client conversation. In a live deployment at a European bank, 99.64% of interactions ran without errors, and the 0.36% with errors include user mistakes. Those figures cover interactions in general, not meeting briefs. They show that errors are rare and not zero, so an advisor who reads the brief first can catch a wrong holding, a stale commitment or a misread event before the client sees it.
Three habits keep the review short:
- Every figure in the brief names its source system.
- Every market event carries a date.
- Suggested topics are labeled as suggestions, never as advice.
How do you measure a pilot of AI meeting preparation?
Compare meetings prepared with the brief against meetings prepared by hand, over the same period and with similar advisors. The standard Streetbeat plan starts with a 3-month live pilot, followed by about 8 to 13 weeks to production, roughly 5 to 6 months in total, so a measurement window of one quarter is realistic. The criteria below are proposals for a firm to adopt, not results from any deployment.
| Criterion | How to measure | What it shows |
|---|---|---|
| Preparation time per meeting | Advisor time logs or a short survey, before and after | Whether the brief replaces manual gathering |
| Share of briefs edited | Edits recorded before the meeting | How much the advisor corrects |
| Errors found by the advisor | Count of errors per batch of briefs, by type | Where the brief is wrong, and how often |
| Advisor usage | Share of advisors active each month | Whether advisors keep using it |
| Cost per interaction | Model token cost, tracked per interaction | Whether the volume stays affordable |
For advisor usage, the live deployment gives a reference point: 85% of 3,000 advisors, more than 2,500, were active over three months. For cost, the model token cost in that deployment was €0.01 to €0.05 per interaction. Both figures come from one institution and do not predict results elsewhere.
What does a worked example look like?
Streetbeat's AI Suite for Advisors is one example. On its public page, Streetbeat describes meeting preparation as "A complete client dossier in minutes: wealth profile, performance and what-if scenarios, in one view." The same page describes live context and compliance checks during calls, and an assistant that draws on the firm's CRM, portfolios and market data to surface prioritized action suggestions for advisor review. Its market and company data covers 170+ financial datasets.
In the table above, that maps to the client profile, holdings and performance, and market events. For conversations and commitments, a firm should ask any vendor how that element is sourced and checked.
Who owns the brief once the advisor opens it?
Everything that reaches the client. The brief proposes; the advisor decides which topics to raise, which suggestions to drop, and what to say. The institution stays responsible for what the assistant can read, how the brief is reviewed and who signs off. Even with 99.64% of interactions without errors in the European deployment, the remaining 0.36% is a reason to keep review in the workflow. A pilot should record how often advisors change or reject a brief, not only how often they open it.
About Streetbeat
Streetbeat helps organizations use AI to grow revenue, not costs. Streetbeat OS, the AI operating system for companies, is the foundation of Streetbeat for Finance, Streetbeat for Impact and Brain50 by Streetbeat.
Streetbeat for Finance
Streetbeat for Finance brings AI to banks, wealth and asset managers, insurers and real estate investment firms. It includes the AI Suite for Advisors, the Portfolio Engine, the AI Suite for Retail and the AI Suite for Operations, deployed on-premise or in the client's cloud and configured around the institution's workflows, data and controls. See Streetbeat for Finance.