What data does an AI assistant need to prepare advisor meetings?
The same data the advisor consults today before seeing a client: CRM, portfolio holdings, recent emails and market data. Before choosing an assistant, test each of these sources on the firm's own systems, and ask that the brief show where each piece of information comes from: that way the advisor can check it in seconds and decide what to bring to the meeting. According to a Kitces Research study from 2019 covering more than 1,000 advisors, the advisors who own the client relationship spent 5.3 hours a week preparing for meetings, out of 43 hours of work: the figures are self-reported by respondents and not recent.
Which systems should the assistant read from?
The same systems the advisor uses, with the same permissions. Ask the vendor which systems the assistant can read in the proposed configuration and where the data is processed; then test each connection on the firm's systems rather than assuming it works. For market data, Streetbeat draws on more than 170 financial datasets: real-time and historical data, company data, news and macro indicators.
| Source | What to ask the vendor |
|---|---|
| CRM and portfolios | Does the assistant read the CRM and the portfolio systems? How current is the data? |
| Can it use correspondence with the client, within the limits of the firm's permissions? | |
| Source documents | Can the advisor open the document behind each piece of information in the brief? |
| Performance and scenarios | Can the advisor see the performance and the assumptions behind each scenario? |
| Advisor checks | What must the advisor check before the meeting and record afterward? |
How long should this data be retained?
For the periods that already apply to the firm's records, which vary by type of firm. The brief is built on data subject to retention, and meeting notes can become subject to it too: the firm's compliance team decides.
| Rule | Who it applies to | What it requires |
|---|---|---|
| FINRA Rule 4511 | FINRA member firms | At least 6 years for books and records with no specified period |
| SEC Rule 204-2 | SEC-registered investment advisers | At least 5 years for most required records, the first 2 in an office of the adviser |
How should a firm test the assistant before choosing it?
With cases prepared for the purpose, because errors are rare but not absent: at a bank using Streetbeat, 99.64% of interactions were error-free. For example, take a client record with a withdrawal request from last year and a new instruction from last week. Does the brief tell them apart? Then correct the record and check that the brief changes accordingly. Compared with today, measure three things: forgotten commitments, data needing correction and preparation time. Finally, see whether advisors keep using the assistant after the test ends.
What does the advisor decide, and what does the firm decide?
The advisor sets the agenda, confirms the client's current situation and chooses whether a scenario belongs in the conversation: the brief is the starting point for their work. The firm chooses the approved sources and the access permissions, and decides who corrects and who approves, without assuming that the assistant has checked every input. FINRA's Regulatory Notice 24-09, dated June 27, 2024, reminds firms that its rules apply whatever the technology, generative AI included.
What has Streetbeat measured in production?
Streetbeat's AI Suite for Advisors uses CRM, portfolio and market data to prepare suggestions that the advisor reviews. The figures below come from one European bank over three months: results in the United States may differ, and none of these numbers measures the quality of the briefs.
| Measure | Result | What it does not show |
|---|---|---|
| Adoption | More than 2,500 of 3,000 advisors (85%) active | Whether the briefs were accurate or useful |
| Error-free interactions | 99.64% (0.36% had errors, including user errors) | Whether the answers were correct or compliant: the advisor verifies each brief |
| Model-token cost | €0.01 to €0.05 per interaction | Implementation, infrastructure and oversight costs |
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.