Investment research: read the filings, then call the people who would know

Investment research starts with a thesis you could be wrong about: test it with filings and industry data, then customer and expert calls that stay clear of MNPI.

Instant Expert EditorialPublished Updated 6 min read

Investment research is deciding whether a security or company is worth owning at its price, and it should start with a thesis you could be wrong about. Test that thesis against filings and industry data, then against customers, suppliers, competitors and former employees, and revise it.

Primary research shows what the filings leave out. It is also where the legal risk sits, since a conversation can stray into material nonpublic information (MNPI) that you are not allowed to trade on. This is general information, not legal advice; if you work at a fund or firm, its compliance policies come first.

1. Write a thesis that could be proven wrong

In two or three sentences, write why the investment could work and what would prove you wrong: "Restaurant operators are switching point-of-sale systems faster than analysts expect, and Company X is winning most of those switches." Each part of that sentence is a question research can answer. Without one, reading more filings rarely changes a decision. (It does make the reading feel productive.)

2. Read what is already public

For a U.S. public company, start with the annual report on Form 10-K and the quarterly reports on Form 10-Q. Investor.gov explains what each part contains (Investor.gov):

  • Item 1, Business: products, services, markets, competition and regulation.
  • Item 1A, Risk Factors: the most significant risks, which companies generally list in order of importance.
  • Item 7, MD&A: management's own explanation of results, trends and uncertainties.
  • Item 8, Financial Statements: the audited statements, notes and auditor's report. Investor.gov suggests looking closely at any qualified opinion or disclosed material weakness.

All 10-Ks and 10-Qs are free on the SEC's EDGAR website. Investor.gov notes that the SEC sets disclosure requirements but does not vouch for a filing's accuracy; the company's CEO and CFO certify it. Then read competitors' filings, earnings call transcripts, trade publications and industry data. Secondary market research lists more public sources, and industry research has a one-week plan for learning an unfamiliar sector.

3. Talk to the people around the company

Filings tell you what the company says; the people around it tell you what they see. The CFA Institute's guidance on material nonpublic information calls reaching customers, contractors, suppliers and competitors normal analyst work for building the most complete picture (CFA Institute Standard II(A)).

  • Customers who chose, stayed with or left the product, and why.
  • Competitors' salespeople who win and lose deals against it.
  • Suppliers and channel partners who see order patterns across the industry.
  • Former employees who can explain how the business works without sharing their former employer's confidential information.

In the restaurant example, five or six operators who recently changed point-of-sale systems would test the first half of the thesis directly.

4. Know exactly what counts as MNPI

Investor.gov defines illegal insider trading as buying or selling a security, in breach of a fiduciary duty or other relationship of trust and confidence, on the basis of material nonpublic information. Violations can include tipping others and trading by the person tipped (Investor.gov).

The CFA standard defines the two parts:

  • Material means disclosure would probably affect the price, or reasonable investors would want to know it before deciding. Its examples include earnings, mergers and acquisitions, regulatory approvals, the gain or loss of a major contract, management changes and product trial results. Reliability matters: factual information from an insider about a new contract is likely material, while a competitor's speculation about the same contract is less likely to be.
  • Nonpublic means not yet available to the market in general. Information shared with a room of analysts, or in a members-only online group, is not necessarily public.

The standard also describes the mosaic theory: an analyst may combine public information with pieces of nonmaterial nonpublic information, such as opinions from customers and retailers, and act on the conclusion, even if that conclusion would have been material had the company disclosed it directly. It recommends saving and documenting your research so you can show how you got there.

5. Run expert calls with the boundary stated up front

The CFA standard says investment professionals may pay outside experts for insight, but they remain responsible for not requesting or acting on confidential information. Expert networks often have both sides sign MNPI agreements; even so, if an expert shares MNPI, you may not act on it until it becomes public.

Two CFA Institute examples show the line. An analyst who asks cancer treatment experts about the latest therapies, without asking for or receiving information such as preliminary trial results, is using expert calls appropriately. An analyst who hears from the scientist leading a company's product tests that results are disappointing, and tells colleagues at the analyst's fund, which then trades, has violated the standard and cannot rely on the agreements the expert signed. So:

  1. Choose experts whose knowledge is about the industry, not the confidential plans of the company you are researching. Be especially careful with current employees of the company and its close partners.
  2. Open every call with the boundary. Tell the expert you do not want confidential information about any current or former employer, or about unannounced results, deals or approvals.
  3. Ask about patterns, not secrets. "How do restaurants decide when to replace their POS system?" is safe ground. "What is Company X's churn this quarter?" is not.
  4. Stop the call if MNPI comes up, write down what happened, and tell your compliance team if you have one. Do not trade on it.
  5. Keep notes of every call as part of your research record.

What to do when an expert cannot share confidential information covers how to keep a conversation useful within those limits.

6. Go back to the thesis after every round

Mark each part confirmed, contradicted or still open. A thesis that survived conversations with people who would know is worth more than one supported only by a model. When experts disagree, what to do when experts disagree explains how to weigh their views.

Your next step

Write your thesis in two sentences and list the three facts that would change your mind. For each one public filings can't answer, list the roles of people who would see it in their daily work.

Instant Expert can find people who match a description such as "owners of multi-location restaurants who changed POS systems in the last year." You review who it finds, it sends your invitations and you pay for each call that gets booked. Screening for conflicts and following your firm's compliance rules stay your job. The directory pages for sales professionals in restaurants and finance professionals in restaurants are one place to start. For managed compliance workflows, compare expert network companies.