What the job involves
A typical sell-side analyst covers 10 to 25 companies in one industry, often with one or two associates. The weekly work falls into a few parts:
- Models. A spreadsheet for each company with revenue, margins, earnings per share and cash flow forecasts. These feed the consensus.
- Reports. An initiation report when coverage starts, then notes after each earnings release, guidance change, deal or industry event.
- Ratings and price targets. A label such as Buy, Hold or Sell and a 12-month price target, both defined by the firm's own rating system.
- Management access. Conferences, non-deal roadshows and calls that put clients in front of company executives.
- Client service. Calls and meetings with portfolio managers and buy-side analysts, who often vote on which analysts to reward with commissions.
Sell-side, buy-side and independent research
| Type | Employer | Who reads it | How it is paid for |
|---|---|---|---|
| Sell-side | Broker-dealer or investment bank | Firm's institutional and some retail clients | Commissions, research payments, wider client relationship |
| Buy-side | Mutual fund, pension, hedge fund | Internal portfolio managers only | Fund's management fees |
| Independent | Research firm with no banking or trading business | Paying subscribers | Subscriptions |
Independent research grew after the 2003 Global Research Analyst Settlement between regulators and major Wall Street firms, which required separation between research and investment banking and funding for independent research for a period.
The rules that apply
Two sets of rules matter most in the US.
FINRA Rule 2241 governs research analyst conflicts. It restricts investment banking staff from directing research, limits analysts' role in pitching banking business, sets quiet periods around offerings the firm underwrites and requires disclosures such as whether the firm has a banking relationship with the company.
The SEC's Regulation Analyst Certification, known as Regulation AC, requires each research report to include a certification that the views are the analyst's own, and a statement on whether the analyst was paid for a specific recommendation.
How to read a sell-side report
The most useful parts of a report are rarely on the first page.
| Section | What it tells you | How much weight to give it |
|---|---|---|
| Rating and price target | The analyst's headline view | Low on its own |
| Estimate changes | How the forecast moved and why | High; revisions move stocks |
| Model | Assumptions on volume, price, margin and cash | High; test your own against it |
| Industry data | Channel checks, surveys, pricing data | Often high |
| Management Q&A | What the company said in meetings | Medium; check against filings |
| Disclosures | Conflicts and rating distribution | Always read |
Changes in estimates often matter more than changes in ratings, a pattern covered in earnings estimate revisions. Rating changes still move stocks for a day or two, as analyst upgrades and downgrades explains.
A year in the life of coverage
The sell-side calendar follows the reporting cycle, and knowing it helps you predict when research will appear.
- Before earnings. Analysts publish previews that set out their estimates and the questions they will ask. Changes to estimates in the week before a report often show which way the analyst leans.
- Earnings night and the next morning. A short first-take note goes out within hours, followed by a fuller note with revised estimates after the call.
- Conferences. Each bank runs industry conferences where companies present and meet investors. Notes from these meetings can carry new detail on demand or pricing.
- Investor days. When a company holds an investor day, analysts often reset their long-term models around the new targets.
- Initiations. When a new analyst starts coverage, the initiation report is usually the most thorough piece the firm will publish on the company, with a full model and industry background.
Coverage can also end. A dropped initiation or a note saying coverage is suspended often follows an analyst leaving the firm or the firm advising on a transaction, not a view on the stock. Check the reason before reading anything into it.
Where the incentives lead
Sell-side research is useful and also shaped by the business around it.
- Coverage follows commissions. Large, liquid stocks attract many analysts. Small companies, spin-offs and post-bankruptcy equities often have none, one reason small caps misprice events more often.
- Ratings skew positive. Analysts need management access, and companies are more open with friendly analysts. Sell ratings are uncommon.
- Herding. An estimate far from consensus carries career risk if it is wrong, so forecasts tend to cluster.
- Short horizons. Reports react to the latest quarter, while many value and event-driven theses take longer to play out.
Where this goes wrong
Sell-side work is a useful input, but it fails in predictable ways when readers lean on it too heavily. The pattern is usually the same: a reader treats a secondary, incentive-shaped source as a conclusion rather than a starting point.
- Using the price target as a forecast. Targets are often set by applying a multiple to an estimate, and they move with the share price.
- Treating consensus as fact. Consensus is an average of models built on similar inputs. It can be wrong together.
- Skipping primary sources. A report is a secondary source. Check the numbers against the filings, the approach set out in primary vs secondary sources in equity research.
- Missing what the analyst cannot say. Quiet periods and conflicts mean an analyst may be silent around an offering or a deal their firm advises on.
The value for your own work is usually in finding where your view differs from the sell-side and why, the idea behind variant perception. Reading the Q&A in an earnings call, which our guide on how to analyze an earnings call transcript covers, also shows which questions the covering analysts think matter.
Individual investors who do not get sell-side reports can find much of the same data in the tools listed in the best stock research tools for individual investors.
The bottom line
A sell-side analyst publishes research for a bank or broker's clients, and the firm earns from the trading and relationships that research supports. FINRA Rule 2241 and Regulation AC limit the conflicts and require disclosure, but the incentives still favor large companies, positive ratings and consensus views. Read the model, the estimate changes and the disclosures before the rating. For small companies and special situations, expect little coverage and plan to do the primary work yourself.