How an AI agent can convert fund research into a comparison brief with differences, risks, assumptions, and next due diligence steps
This workflow gives an AI agent a defined job, a bounded set of records, and a result a person can review. The agent reads the relevant Morningstar context, applies the rules in the prompt, and keeps the source behind every recommendation. It returns a proposed handoff rather than taking consequential actions on its own.
Can an AI agent convert fund research into a comparison brief with differences, risks, assumptions, and next due diligence steps?
Yes. Start with the scope, date range, decision rules, and fields that identify the right records. The agent can collect the evidence, compare states or sources, mark conflicts and missing data, and organize the result around the outcome above. A reviewer then checks the matches and judgment calls before approving messages, record updates, bookings, purchases, publishing, or other write actions. The guide below shows the records, boundaries, prompt, and handoff needed for this specific workflow.
What this agent helps you do
A Morningstar fund comparison agent helps analysts organize differences between candidate funds or investments. It summarizes available context, highlights risk questions, and prepares due diligence next steps.
When to use this workflow
Use it before portfolio reviews, fund screens, client research preparation, or investment committee discussions. Keep final decisions with qualified humans.
How Morningstar gives the agent context
Connect Morningstar and define the fund list, peer group, horizon, and comparison criteria. Ask the agent to distinguish observations from recommendations and to call out missing data.
Example starter prompt
Compare these funds using Morningstar context. Prepare a table of key differences, risk questions, assumptions, follow-up research, and caveats. Do not provide personalized investment advice.Suggested workflow steps
The agent gathers available fund context, compares dimensions relevant to the request, identifies risk areas, and prepares follow-up questions.
Keep Morningstar data dates, share classes, currencies, benchmark or category context, and fee basis aligned before comparing values. A difference caused by mismatched share classes belongs in data cleanup, not the investment conclusion.
Expected handoff
The handoff should include a comparison table, summary, caveats, and next due diligence steps. It can feed an analyst memo or review agenda.
Questions this workflow answers
What would let me compare several funds fairly when their categories, share classes, fees, and reporting periods differ?
Begin with the investor’s research question, fund identifiers, share classes, currency, as-of date, and measures to compare. Morningstar supplies available fund, category, holdings, performance, fee, and risk context. The agent should verify identity and normalize periods and units before presenting a ranking.
The comparison can include objective, category, benchmark, share-class terms, fees, performance periods, volatility or drawdown measures, holdings concentration, sector or geographic exposure, and stated strategy. Missing or stale values remain visible. A lower fee or higher historical return does not establish suitability, and figures from different periods should not share one column without qualification.
Ask the agent to explain meaningful differences and data limitations rather than collapse them into one score. A concentrated fund may be intentional. Category averages can be useful context without becoming a recommendation. The agent should cite the record and period behind each figure and separate reported data from any calculation it performs.
The handoff includes a comparable table, tradeoffs, identity and date checks, caveats, and due-diligence questions. An analyst verifies the sources and applies client, mandate, tax, and regulatory context outside the workflow. The agent does not recommend a purchase or predict returns. It makes the evidence comparable enough for a qualified person to continue the review.
Share-class identity can change the conclusion before performance analysis begins. Two tickers may represent the same underlying portfolio with different fees, currency hedging, distribution policy, or investor eligibility. The agent should show those attributes beside the fund name and exclude a class that does not match the requested investor or market. It should also align trailing periods to the same end date and mark funds without a complete history instead of treating missing returns as zero. For holdings, compare dates and levels of disclosure; a quarter-old top-ten list is not equivalent to a current full portfolio. These checks keep a neat table from implying comparability the source records do not support.