Constructing an Evidence-Based Stock Thesis: Quality, Valuation, Catalysts, and Risks

Hello. In the previous lesson, you learned to compare earnings growth, ROE, P/E, price-to-sales, and free-cash-flow yield across a company’s history and a sensible peer group. Those metrics are inputs, not a conclusion.

This lesson turns that research into a compact stock thesis: a written, evidence-backed view of the business and its shares. The aim is not to produce a confident-sounding “buy” call. It is to state what you think matters, why the current valuation may or may not reflect it, what could change the market’s view, and what facts would prove you wrong.

By the end, you should be able to write a roughly 200–300 word thesis for a US or UK stock as a paper-research exercise.

---

A thesis is a claim about a stock, not a company description

A company can be excellent while its shares are unattractive at the current price. Equally, a troubled company can have an interesting stock if expectations are excessively pessimistic. Your thesis therefore needs to connect two separate judgments:

1. Business quality: Is this a durable business with credible economics, financial resilience, and a defensible competitive position?
2. Stock valuation: What expectations appear embedded in the share price, and are your expectations meaningfully different?

A useful thesis is:

• specific — it identifies the few drivers that matter;
• evidence-based — each factual claim can be traced to a filing, results release, or dated market-data snapshot;
• conditional — it says what needs to happen for the thesis to work;
• falsifiable — it names outcomes that weaken or invalidate the view;
• brief — it filters rather than repeats every fact discovered in research.

A long company history or a list of ratios is not a thesis. For example, “the company has growing revenue, a 17% ROE, and a P/E of 15” describes data, but does not explain why the shares could be mispriced.

The better question is: What does the market seem to expect, what is my alternative view, and what evidence would distinguish the two?

---

Start with business quality: explain why the economics might persist

The first part of a thesis should answer four practical questions:

• What does the company sell, to whom, and how does it make money?
• What primarily drives revenue, margins, and free cash flow?
• Why might customers continue to choose it over alternatives?
• Does its balance sheet support the strategy, or constrain it?

Your prior work on revenue, margins, operating cash flow, free cash flow, debt, and equity belongs here—but selectively. Include a figure only when it supports the argument. If a company’s appeal is recurring software revenue and improving margins, revenue mix, retention, operating margin, and cash conversion may matter. If it is a retailer, like-for-like sales, store economics, inventory, and lease obligations could be more important.

Competitive advantage is often described as an economic moat: a feature that makes it difficult for competitors to erode profits over time.

Use those categories as a checklist, not as labels to award automatically:

Possible moat	Mechanism	Evidence worth seeking
Network effects	The product becomes more valuable as more users, buyers, sellers, or participants join.	User engagement, transaction liquidity, retention, market share, or a two-sided marketplace’s growth.
Switching costs	Customers incur financial, operational, training, or data-migration costs when changing provider.	Renewal rates, contract duration, embedded workflows, customer concentration, and churn trends.
Economies of scale	Larger scale lowers unit costs or improves distribution and purchasing power.	Stable or improving margins, cost per unit, logistics density, or a cost advantage versus peers.
Intangible assets	Brand, licenses, patents, regulatory permissions, proprietary data, or trusted reputation protect returns.	Pricing power, premium positioning, renewal behaviour, legal protection, or regulated access.

A moat is not proven by saying “the company has a strong brand.” The evidence must show an economic consequence, such as pricing power, resilient margins, repeat purchases, or lower churn.

A simple test is to finish the business-quality section with a causal statement:

Higher customer retention supports recurring revenue; recurring revenue supports more predictable cash generation; predictable cash generation can justify a higher valuation than a similarly sized but cyclical business.

That is an argument. It can later be tested against results.

---

Valuation: identify the expectation gap, not merely a “cheap” multiple

Your valuation paragraph should state what the market is currently paying and whether that appears reasonable in context.

From the prior lesson, you already have a basic toolkit:

• trailing P/E;
• price-to-sales;
• free-cash-flow yield;
• EPS growth;
• ROE;
• comparison with history and carefully chosen peers.

A good valuation statement contains three pieces:

1. A dated valuation fact
   For example: “At the closing price on 15 May 2025, shares trade at 18 times trailing diluted EPS and a 5.5% trailing FCF yield.”

2. A relevant comparison
   Compare with the company’s own historical valuation range and peers with similar business models, growth, margins, and financial risk.

3. An explanation for the difference
   A lower P/E may reflect weaker growth, cyclicality, debt, poor cash conversion, or a real competitive threat. A premium multiple may reflect stronger margins, recurring revenue, a superior balance sheet, or simply optimistic assumptions.

The most important distinction is between a low multiple and a mispriced stock. A stock trading at 10 times earnings can be a bargain, or it can be cheap because earnings are cyclical and about to fall. Your thesis must explain which possibility you consider more likely.

For a beginner’s first thesis, do not force a precise target price. You can write a valuation range or a conditional scenario instead. If you do use a simple P/E-based scenario, make the assumptions visible:

$$\text{Implied share price}=\text{Estimated diluted EPS}\times \text{Assumed P/E multiple}$$

The estimate is only as useful as its assumptions. State what would justify the chosen earnings estimate and multiple; do not treat either as a fact.

---

Catalysts: what could cause a reappraisal?

A catalyst is an event or development that could reveal new information and change investors’ expectations. It is not simply “the stock goes up.”

Possible catalysts include:

• an earnings release showing acceleration in revenue, margins, or free cash flow;
• a product launch that is large enough to affect future revenue;
• a debt reduction or refinancing that changes financial risk;
• a regulatory decision;
• an acquisition, disposal, or share-buyback announcement;
• evidence that a temporary headwind is easing;
• an investor-day update that clarifies credible long-term targets.

Be precise about the mechanism. “Next earnings” is not automatically a catalyst; every company reports earnings. A more useful statement is:

The next two quarterly updates could be catalysts if customer retention remains above management’s stated threshold and operating margin continues to expand, because that would challenge the market view that growth is deteriorating.

A catalyst can also fail to occur. That possibility belongs in the risk section.

---

Risks: write the strongest counterargument

A risk section should not be a generic warning that “markets can be volatile.” It should identify the most material ways the thesis could be wrong.

For each risk, include:

• the threat — what might happen;
• the mechanism — how it would damage revenue, margins, cash flow, or valuation;
• the monitoring evidence — which future metric, disclosure, or event you will watch;
• the consequence — whether it weakens the thesis or invalidates it.

For example:

Weak risk statement	Testable risk statement
“Competition is a risk.”	“A major competitor could reduce prices, lowering renewal rates and gross margin. Reassess if retention declines for two consecutive reporting periods or management cuts margin guidance.”
“Debt is a concern.”	“Refinancing risk is material because debt matures before expected cash generation can reduce leverage. Reassess if refinancing occurs at substantially higher interest cost or net debt rises.”
“The sector is cyclical.”	“Demand may decline if industrial production weakens, reducing volumes and operating leverage. Treat unusually high current margins as non-normal until tested against a downturn.”

This is similar to debugging a system design: do not merely note that a component may fail. Specify the failure mode, the observable signal, and its effect on the whole system.

A robust thesis is allowed to conclude “watchlist only” or “insufficient evidence.” Deciding not to form a conviction is often better research than manufacturing a recommendation.

---

Cite evidence so another person can verify it

Use citations for factual claims, especially numbers, management targets, segment facts, market share claims, and valuation data. Your citation should answer: Where exactly did this come from?

A compact source format works well:

• [1] Company name, Annual Report 2024, “Business Model,” p. 18.
• [2] Company name, FY 2024 Results Release, cash-flow table, 12 March 2025.
• [3] TradingView price snapshot, primary listing, closing price on 15 May 2025.
• [4] Company name, Q1 2025 Earnings Call, management guidance section.

Keep facts and interpretations separate:

• Fact: “Operating margin increased from 18% to 21% in FY2024 [2].”
• Interpretation: “The increase may indicate operating leverage, but it should be checked against any one-off cost reductions.”

For a UK company, retain the unit discipline from the previous lesson: share prices may be quoted in pence while financial statements use pounds. Also record the fiscal year, reporting currency, and exact price date used for your valuation.

---

A complete mini-thesis: fictional example

The example below uses an invented company, documents, numbers, and citations. It demonstrates structure only; it is not based on a real security.

Harbor Software plc — paper-research candidate, 12-month horizon. Harbor provides workflow software to mid-sized logistics firms. Its business quality appears above average because 85% of FY2024 revenue was subscription-based, customer retention was 93%, and operating margin increased from 22% to 24% as revenue grew 12% [1][2]. These figures suggest switching costs from integrated customer workflows, although this claim depends on retention remaining durable rather than simply reflecting a strong industry cycle.

At a market capitalisation of £1.35 billion on 15 May 2025, the shares trade at 22 times trailing diluted EPS, 2.7 times sales, and a 6.7% FCF yield [3]. The P/E is below the 23–30 times range of similar software peers, despite comparable revenue growth and stronger cash conversion [4]. My working view is that the discount reflects concern about slowing customer additions, but may be too large if existing-customer expansion remains stable.

The next two results releases could be catalysts if renewal rates stay above 90% and management confirms margin expansion, which would support earnings estimates and a valuation closer to peers. Key risks are price competition, a fall in logistics-sector spending, and weaker retention following a competitor’s product release. The thesis would weaken materially if retention falls below 90% for two consecutive quarters or free-cash-flow margin declines without a credible investment explanation.

Notice the order:

1. What the business does and why it may be durable.
2. What the market currently pays.
3. What may be misunderstood or underappreciated.
4. What evidence could change the market’s view.
5. What could prove the view wrong.

The thesis does not claim certainty. It identifies the small set of facts worth following.

---

Your 20-minute first-draft workflow

Choose one liquid US or UK company with accessible annual and interim reports. Avoid starting with a company whose business you cannot explain in two sentences.

Collect only these items before drafting:

1. The latest annual report and latest results release.
2. A dated share price or market-capitalisation snapshot from TradingView.
3. Revenue growth, operating margin, diluted EPS, operating cash flow, free cash flow, debt or net cash, and share count.
4. Two genuinely comparable peers.
5. One recent development or upcoming event that could change expectations.

Then write within this structure, with a strict 250-word limit:

Company, listing, research date, and intended horizon

Working conclusion:
[Watchlist / paper-research candidate / no thesis yet], because ...

Business quality:
The company earns money by ...
The evidence for, or against, a durable advantage is ...
Relevant financial evidence: ... [1][2]

Valuation:
At [date], the shares trade at ...
Compared with [history / peers], this appears ...
My working expectation differs because ...

Potential catalysts:
The next evidence to watch is ...
It matters because ...

Material risks and invalidation:
The strongest counterargument is ...
I will reassess or reject the thesis if ...

After writing, perform one hostile review. Remove any sentence that is merely promotional, vague, or unsupported. Replace “strong,” “cheap,” “best-in-class,” and “huge opportunity” with a measurable claim, a cited fact, or a clearly labelled judgment.

---

Key takeaways

• A stock thesis is a testable claim about business quality, valuation, catalysts, and risks—not a collection of ratios or a company biography.
• Separate the quality of the business from the attractiveness of the shares at the current price.
• Evidence for a moat should show an economic consequence: retention, pricing power, margins, cost advantage, or durable market position.
• A valuation multiple is meaningful only relative to growth, cash generation, financial risk, history, and appropriate peers.
• A catalyst must explain how new information could alter expectations.
• Risks should name a mechanism, a monitoring signal, and an invalidation condition.
• Cite every material factual claim to a dated, traceable source.

This completes the fundamental-analysis module. Next, the course moves to price charts: reading OHLCV data and understanding how chart timeframe changes the information visible to a trader.