Sector & Industry Research for Equity Coverage

A name does not trade in isolation; it trades inside a sector thesis. Before you can defend a price target you have to know whether the whole pool is expanding or shrinking, where the profit pools sit, who is taking share, and what the channel is actually seeing. Picking up a new coverage assignment used to mean a week of reading sell-side primers, trade-body data, and Bloomberg articles before you could form a view. AI compresses that ramp to an afternoon — and lets you refresh the thesis every quarter instead of every coverage hand-off.

This lesson covers the sector-research workflows equity analysts use to build and maintain a coverage thesis: TAM/SAM/SOM sizing as an input to the bull case, competitive dynamics and share-shift analysis, sector primers, regulatory reviews, and — most importantly — synthesizing channel checks into a position view. The mechanics of building the underlying valuation come in the next module; this lesson is about the thesis the valuation has to support.

What You'll Learn

• How to size a market top-down and bottom-up as an input to your bull/bear case
• How to map competitive dynamics and detect share shifts across your coverage
• How to build a sector primer that frames your whole coverage universe
• How to summarize the regulatory and policy environment that moves the sector
• How to synthesize a channel check into a position view without losing the speed advantage

Sizing the Market as a Thesis Input

For an equity analyst, market sizing is rarely an academic exercise — it is the denominator behind your revenue forecast. If the bull case rests on a company growing into a $40B TAM, the size and growth of that pool is a load-bearing assumption a PM will challenge. The discipline: do it both top-down and bottom-up, then triangulate, and treat the divergence as a signal of how much of your thesis is genuinely underwritten.

Top-down with AI:

I need to size the total addressable market for [PRODUCT/SERVICE]
in [GEOGRAPHY] for [YEAR]. Build a top-down estimate.

Walk me through:
1. The largest umbrella market this product sits in. State its
   reported size and cite a source (Gartner, IDC, McKinsey, World Bank,
   or a major trade body).
2. The relevant sub-segment as a % of the umbrella market.
3. The serviceable share of that sub-segment (geography, regulation,
   product fit).
4. The realistically obtainable share given likely competition.

Show each step's logic and end with a TAM / SAM / SOM table in USD.
Flag any number you are uncertain about with [verify].

Bottom-up with AI:

Now estimate the same market bottom-up. Build it from:
1. Number of potential customers (B2B: number of qualifying companies;
   B2C: number of households or eligible individuals)
2. Penetration of the product category among those customers
3. Average revenue per customer

Walk through where each number comes from. Identify the assumption
that drives the most uncertainty in the final estimate.

The two answers will diverge. The interesting analysis is explaining why — and the gap usually maps directly onto the spread between your bull and bear cases.

Competitive Dynamics and Share Shifts

For an equity analyst the question is not just "who are the players" but "who is taking share, from whom, and is my covered name on the right side of it?" A static competitor map is table stakes; the alpha is in the direction of travel. Prompt:

I cover [TICKER] in the [SECTOR] sector. Map the competitive
landscape with an emphasis on share shifts, not just a static list:

1. The top 10-15 public and private players. For each: revenue
   scale band (under $50M / $50M-$500M / $500M-$5B / $5B+), ownership
   (public ticker / private / PE-backed), primary geography, and a
   one-sentence positioning.
2. The two axes of differentiation that most explain why share is
   moving (e.g. price vs premium, platform vs point solution).
3. Place each on a 2x2 grid using those axes.
4. Direction of travel: who has gained or lost revenue share over
   the last 3 years, with the disclosed reason (pricing, product
   cycle, distribution, M&A).
5. Where my covered name [TICKER] sits and whether it is a share
   gainer, defender, or donor.

If you do not know a specific number, write [verify]. Do not invent
competitors that do not exist.

You will spot a share donor or gainer you under-weighted in the model. You will also discover that AI sometimes invents companies — always verify by ticker or company website before any of this reaches a note. The share-shift framing is what turns a generic competitor list into a coverage thesis: it tells you whether your name is structurally winning or quietly bleeding share.

Sector Primers

A sector primer is the 5-page document a new associate reads on day one of a coverage assignment. AI produces a usable draft in 20 minutes:

Generate a sector primer on [SECTOR]. Use this structure:

1. SECTOR DEFINITION — what is included, what is adjacent but excluded
2. ECONOMIC SHAPE — total revenue pool, growth rate, profit pools
3. KEY VALUE CHAIN PLAYERS — who captures value at each step
4. BUSINESS MODELS — the 3-5 dominant business models in the sector
5. KEY METRICS — the operational and financial metrics analysts watch
6. SEASONALITY AND CYCLICALITY — how sector revenue and margins behave
7. REGULATORY ENVIRONMENT — the rules that materially shape competition
8. THREE STRUCTURAL TRENDS — that will shape the next 5 years
9. COMMON ANALYTICAL PITFALLS — mistakes new analysts make in this space

Each section: 80-120 words. Cite sources where possible.

A primer like this used to take a week. The AI version is your first draft, which you sharpen with sector specialists and live data.

Regulatory and Policy Reviews

Regulation is one area where AI shines because the rules are public and stable:

Summarize the regulatory environment for [INDUSTRY] in [GEOGRAPHY].
For each major regulation:

1. Name of the rule and adopting body (SEC, EU Commission, etc.)
2. What it requires in plain English (one paragraph)
3. Who it applies to (revenue thresholds, business types)
4. Penalties for non-compliance
5. Most recent material change (date, what changed)
6. Implications for company valuations or competitive dynamics

End with a forward look: which regulatory changes are in proposed
or consultation stage that could materially shift the industry in
the next 12-24 months.

This is the kind of work that used to require a call to outside counsel. AI gives you the analyst-grade version that lets you focus your legal calls on the genuinely uncertain parts.

Synthesizing Channel Checks Into a Position View

A channel check is a structured set of conversations with customers, suppliers, ex-employees, or industry consultants to test a thesis on a name you cover. For an equity analyst, the point of the check is not the transcript — it is the decision: does this confirm, contradict, or resize my position? AI does not replace the conversation, but it makes preparation faster and, crucially, turns raw call notes into a synthesized view across multiple checks. A word of caution before you start: a channel check is exactly the kind of activity where MNPI and Reg FD risk is highest, so the compliance lesson at the end of this course is required reading, not optional.

Pre-call preparation:

I am doing a channel check on [COMPANY] by interviewing a former
sales director (left the company 8 months ago). Help me prepare:

1. List the 15 best questions to ask, ordered by importance,
   covering: sales process changes, win/loss against key competitors,
   pricing pressure, sales rep productivity and tenure, demand outlook
2. Flag any questions that could be perceived as MNPI requests and
   suggest reframings
3. Suggest 3 questions that would help me detect if the interviewee
   is venting personal grievances vs reporting factual patterns

Post-call synthesis:

I just finished a 30-minute channel check call with a former
[ROLE] at [COMPANY]. My notes are below. Summarize:

1. The five most important things I learned
2. Which of my prior thesis points were confirmed
3. Which were contradicted
4. New questions I should investigate
5. How I should adjust my model assumptions if at all

[PASTE NOTES]

Do not invent details I did not write down.

Synthesizing across multiple checks is where the position view actually forms. After three or four calls on the same name, paste all the cleaned notes into a Claude Project and ask:

Below are my notes from four channel checks on [TICKER] (two
customers, one former sales lead, one distributor). Synthesize:

1. Where do the four sources AGREE? Treat agreement as higher-conviction.
2. Where do they CONTRADICT each other, and what would explain the
   contradiction (geography, segment, recency, incentive to spin)?
3. Net read on the three thesis pillars: [PILLAR 1], [PILLAR 2],
   [PILLAR 3] — confirmed, mixed, or refuted.
4. The single channel signal that should most change my position
   sizing, and in which direction.

Do not invent details. Quote which source said what.

Same pattern: AI structures and synthesizes; you supply the facts and own the position decision.

Verifying Without Losing Speed

The risk with AI-generated research is plausible nonsense. The mitigation is structured verification, not redoing the research by hand.

Three verification habits:

1. Spot-check the citations. AI cites real sources roughly 90% of the time but invents the rest. Click every citation before relying on it.
2. Anchor to one ground-truth number. If AI says the market is $X billion, verify that number against the original source. If it is wrong, the rest of the analysis is suspect.
3. Cross-check with Perplexity. Perplexity is built for cited research. Run the same question through both ChatGPT and Perplexity; if they agree, you can trust it. If they disagree, dig in.

Building a Weekly Sector Thesis Workflow

The end-state for an analyst maintaining a live sector thesis across coverage:

1. Monday: Use NotebookLM or a Claude Project loaded with last month's broker reports and your own notes to flag where the sector thesis has drifted from the data
2. Tuesday: Use ChatGPT or Claude to produce structured drafts (TAM sizing, share-shift map, regulatory summary)
3. Wednesday: Verify the drafts — citations, ground-truth numbers, cross-checks with Perplexity
4. Thursday: Run channel checks with AI-built prep, then synthesize the calls into a position read
5. Friday: Roll it all into the sector note that frames every name you cover for the PM

Total time: about 60% of what the same workflow took without AI. Once you have done this twice for a sector, the natural next step is to bake the coverage context into a persistent assistant — covered in the Sector-Coverage Assistants lesson later in this course.

Key Takeaways

• Size the market two ways — top-down and bottom-up — and treat the divergence as a read on how underwritten your bull case is
• Map competitive dynamics by direction of travel, not a static list: who is gaining share, who is donating it, and where your covered name sits
• Sector primers frame your whole coverage universe, but always layer in specialist conversations
• The output of a channel check is a position view, not a transcript — synthesize across multiple calls
• Spot-check citations, anchor to one ground truth, cross-check with Perplexity — and never let unverified channel-check detail near a published note