Final round at a hedge fund. You walk the portfolio manager through a clean DCF built on unlevered free cash flows, WACC and a terminal value. Your sensitivity table shows $32 to $45 per share. The stock trades at $40. The portfolio manager asks one question: “So, would you buy it?” Your range does not answer it.
That silence is the gap between a valuation and an investment thesis. A valuation estimates what a company is worth under stated assumptions. An investment thesis explains why the market price is wrong, by how much, what will close the gap, and what would prove you wrong.
Bankers are trained to produce the first. Buy-side investors are paid for the second. This guide shows the difference on one illustrative model, built both ways, and gives you four steps that turn a valuation into an investment view.
Key Takeaways
- A valuation answers “What is it worth?” An investment thesis answers “Why is the price wrong, and what closes the gap?”
- Banks value companies for clients. A fairness opinion judges a price for a defined group of holders; it is not a recommendation to buy.
- An investor compares the price with the expectations it implies (reverse DCF), not only with an intrinsic value range.
- A thesis needs a variant view, probability-weighted scenarios, a catalyst and conditions under which it fails.
- A valuation range that contains the share price says nothing about whether to buy.
What Is the Difference Between a Valuation and an Investment Thesis?
A valuation is a quantitative estimate of what a company is worth under stated assumptions. An investment thesis is a falsifiable argument that the market price is wrong. It states by how much, why, what will close the gap, and what would prove it wrong. The valuation is one input to the thesis, not a substitute for it.
The two use different reference points. A valuation compares a company with its own assumptions. A thesis compares the share price with the expectations that price already contains. NYU Stern professor Aswath Damodaran draws the same line: value comes from fundamentals, while price is set by supply and demand, including mood and momentum. An investor bets that the gap between the two will close.
| Valuation | Investment thesis | |
|---|---|---|
| Core question | What is the company worth? | Why is the price wrong, and what closes the gap? |
| Output | A value or a range | Buy, pass or short, with size and time frame |
| Reference point | Your own assumptions | Expectations implied by the current price |
| Who bears the outcome | The client who uses the number | The investor who commits capital |
| How it fails | An assumption is wrong | The thesis is wrong, or the gap never closes |
The habit starts earlier than most candidates think. Our advice for IB modeling tests applies here too. Read the CIM or case materials as if you had to defend an investment recommendation, not as a source of numbers to plug in. A banker who reads that way is already halfway to a thesis.
Why Does a Banker’s Valuation Stop Short of a Recommendation?
An investment bank values a company for a client, inside a mandate. Its work answers whether a price is defensible for that client, not whether anyone should own the stock. A fairness opinion only judges a price for a defined group of holders, and FINRA Rule 5150 requires member firms to disclose contingent fees.
Advocacy, not conviction
On a sell-side mandate, the bank supports the highest price it can defend. On a buy-side mandate, it supports the lowest. In practice, both sides often run the same methods on similar data and lean on different ends of the range. That is advocacy, and it is the job.
We draw a similar line between a pitch deck and a model memo. A pitch deck selects favorable data and packages it for an audience that has not committed yet. A valuation inside a pitchbook works the same way. It is built to support a client’s position, not to put capital at risk.
What a fairness opinion says, and what it does not
A fairness opinion concludes that a price is “fair, from a financial point of view” to a defined group of holders. It does not say whether the stock is a good investment. FINRA Rule 5150 applies when a member firm’s opinion is provided to public shareholders. The firm must disclose whether it acted as financial advisor and whether its fee is contingent on the deal closing. It must also disclose whether a fairness committee approved the opinion.
A Deal From the Public Record: Leaf Group, 2021
In April 2021, Graham Holdings agreed to buy Leaf Group for $8.50 per share in cash. According to the merger proxy filed with the SEC, the price was a premium of about 21% to the April 1, 2021 close. Canaccord Genuity opined on April 3, 2021 that the price was fair, from a financial point of view, to Leaf Group’s stockholders. A fairness committee approved the opinion. The proxy states that the opinion is not a recommendation on how any stockholder should vote. On April 21, 2021, Leaf Group closed at $9.11, above the deal price. The opinion judged the price for existing stockholders. It did not judge whether buying the stock at $9.11 was attractive.
One Model, Two Answers: A Worked Example
The same DCF can give a banker’s answer and an investor’s answer. The banker’s answer is a value range. The investor’s answer starts from the share price and solves for the growth it implies. It then compares that growth with a researched view, weights scenarios and states an expected value and a downside.
The illustrative company
All figures are illustrative and rounded to one decimal place. The model uses five forecast years and end-of-year discounting.
| Assumption | Value |
|---|---|
| Base-year revenue | $2,000.0 million |
| EBIT margin | 15% |
| Tax rate | 25% |
| Net reinvestment (capex less D&A, plus working capital) | 30% of each year’s revenue increase |
| WACC | 9.0% |
| Terminal growth (Gordon growth) | 3.0% |
| Net debt | $500.0 million |
| Shares outstanding | 100.0 million |
| Share price | $40.00 |
| Management case revenue growth | 8% per year |
Unlevered free cash flow (UFCF) equals NOPAT, which is EBIT × (1 − tax rate), minus net reinvestment. The terminal value uses year 6 UFCF divided by (WACC − g).
The banker’s output
On the management case, the DCF looks like this ($ million):
| Year 1 | Year 2 | Year 3 | Year 4 | Year 5 | |
|---|---|---|---|---|---|
| Revenue | 2,160.0 | 2,332.8 | 2,519.4 | 2,721.0 | 2,938.7 |
| EBIT | 324.0 | 349.9 | 377.9 | 408.1 | 440.8 |
| NOPAT | 243.0 | 262.4 | 283.4 | 306.1 | 330.6 |
| Net reinvestment | 48.0 | 51.8 | 56.0 | 60.5 | 65.3 |
| UFCF | 195.0 | 210.6 | 227.4 | 245.6 | 265.3 |
| PV of UFCF | 178.9 | 177.3 | 175.6 | 174.0 | 172.4 |
The present value of the five cash flows is $878.2 million. Year 6 UFCF of $314.1 million gives a terminal value of $5,234.5 million and a present value of $3,402.1 million. Enterprise value is $4,280.3 million, equity value is $3,780.3 million, and the value per share is $37.80. The terminal value carries 79.5% of enterprise value.
The banker then flexes the two assumptions nobody can prove:
| Value per share | g = 2.5% | g = 3.0% | g = 3.5% |
|---|---|---|---|
| WACC 8.5% | $39.02 | $41.88 | $45.31 |
| WACC 9.0% | $35.46 | $37.80 | $40.57 |
| WACC 9.5% | $32.41 | $34.36 | $36.62 |
The range is $32.41 to $45.31. The share price of $40 sits inside it. For a banker, that is a complete answer: the price is defensible. For an investor, it is no answer at all.
Step 1: Read what the price implies (reverse DCF)
A reverse DCF starts from the share price and solves for the expectations it contains. Michael Mauboussin and Alfred Rappaport built their book Expectations Investing on this idea, as the CFA Institute’s review of the 2021 edition describes.
In Excel, set the value-per-share cell to 40 by changing the revenue growth input. The tool is Goal Seek, on the Data tab in the Forecast group, under What-If Analysis > Goal Seek. Every other assumption stays fixed.
At $40 per share, the market implies revenue growth of 9.4% a year for five years. That is faster than management’s own 8% plan. The question is no longer “What is it worth?” It is “Will this company grow faster than its own management expects?”
Step 2: State your variant view on the drivers that matter
A variant view is where your forecast differs from what the price implies, with a reason. The benchmark is the price-implied path, not only the sell-side consensus. Keep the view to the few inputs that move value. Our rule for modeling tests is the same: identify the three to five assumptions that drive about 80% of the output. In this model, revenue growth and EBIT margin dominate.
Suppose your research shows the largest product line is maturing. Your base case is 5% growth at a 15% margin, well below the 9.4% the price implies. That gap is the thesis.
Step 3: Weight scenarios and state the downside
A single number hides risk. Build three coherent cases, assign probabilities, and compute the expected value. A simple data table on the key inputs shows that you treat the model as a decision tool, not just a calculation engine. Present the base case first, the downside second and the upside third.
| Scenario | Revenue growth | EBIT margin | Value per share | vs. $40 | Probability |
|---|---|---|---|---|---|
| Base | 5% | 15% | $33.34 | −16.6% | 45% |
| Bear | 2% | 13% | $24.41 | −39.0% | 25% |
| Bull | 9% | 16% | $42.72 | +6.8% | 30% |
| Expected value | $33.92 | −15.2% |
The expected value of $33.92 sits 15.2% below the share price. The bear case loses 39.0%, while the bull case gains only 6.8%. That risk/reward is asymmetric in the wrong direction. An investor who wants at least 15% upside to the expected value would need to pay about $29.50 or less.
For each scenario input, write down where it came from and the range of defensible alternatives. Then note what happens to the output if the input moves by 10%. That turns three numbers into an argument an interviewer can test.
Step 4: Name the catalyst and the kill criteria
A catalyst is the event that forces the market to update its expectations. Here it could be the next two quarterly reports, where growth well below 9% would challenge the implied path. A kill criterion is the evidence that proves you wrong: for example, two consecutive quarters of year-over-year revenue growth above 9%.
The investor’s answer at $40 is no. The same model that told the banker “fair” tells the investor something else: the price already assumes more growth than the business is likely to deliver.
What Does an Investment Thesis Need Beyond the Model?
An investment thesis needs five elements a valuation model does not supply. The first is a variant view against what the price implies. The others are a reason for the mispricing, a catalyst with a time frame, an asymmetric risk/reward, and written kill criteria.
- Variant view: where your forecast differs from the price-implied expectations, in numbers.
- Reason for the mispricing: why the market holds a different view (a maturing product line, a misread margin, an overlooked asset).
- Catalyst: the event that closes the gap, and roughly when.
- Asymmetric risk/reward: upside and downside, probability-weighted, with the downside stated first.
- Kill criteria: the evidence that would make you exit.
The catalyst matters because value and price can stay apart. Damodaran warns that you can be right and go bankrupt being right, when the pricing process keeps the price away from value for long periods. A thesis without a catalyst is a valuation with an opinion attached.
A thesis can also fail in two separate ways. Thesis risk means the business does not deliver what you assumed. Valuation risk means the business delivers, but the market applies a lower multiple to the same earnings. Name both. The first shows up in quarterly filings; the second rarely leaves evidence in advance.
Write down what your analysis cannot answer. Our guide to model memos puts it directly: stating what a model cannot do is the clearest signal of analytical discipline. The same holds for a thesis.
Thesis template, applied to the example
At $40, the market implies 9.4% annual revenue growth, faster than management’s 8% plan. The base case is about 5%, because the largest product line is maturing. Probability-weighted value is $33.92, with 39% downside in the bear case and 7% upside in the bull case. The next two quarterly reports are the catalyst. Two consecutive quarters of year-over-year growth above 9% would prove the thesis wrong.
Where Do Banking Habits Break on the Buy Side?
Banking trains completeness, client advocacy and defensible ranges. Buy-side interviewers test the opposite: a point of view, the few drivers that matter, and risk stated before upside. The habits below are where a technically strong banker loses a buy-side case, even with a flawless model.
| Banking habit | What the buy side expects instead |
|---|---|
| Adopt the management case | Build your own case and explain every assumption |
| Present a value range | Give a probability-weighted view and a decision |
| Lead with the upside | Base case first, then downside, then upside |
| Show outputs | Show the assumptions that produce them |
| Model everything | Focus on the three to five drivers that move value |
| Compare value with your own model | Compare price with what the price implies |
The first habit is the hardest to break. In the case-plus-model format, common at VP level and above or in boutique processes, the model is only the starting point of the discussion. Every assumption needs to be yours, and you need to explain it without the file in front of you. A buy-side case tends to apply that standard at every level.
From Our Practice
Our guide to model memos names the most common mistake: presenting an output without its drivers. A 22% IRR means little without the growth rate, exit multiple and leverage behind it. A second habit: most analysts show the upside scenario first because it looks better. In a buy-side case, both habits signal a seller, not an investor. State the assumptions, then lead with base and downside.
If your valuation work is solid but cases stall at the step from range to recommendation, consider our Expert Valuation Package. It covers scenario and sensitivity work in three two-hour sessions.
How Do Valuation and Thesis Fit in a Stock Pitch vs. a PE Case?
In a stock pitch, the thesis is a claim that the public market misprices expectations, and valuation shows upside against downside. In a private equity case, the thesis is a value-creation plan the buyer controls. Valuation becomes returns math: entry price, leverage, exit multiple, IRR and MOIC.
| Hedge fund stock pitch | PE case study | |
|---|---|---|
| What closes the gap | A catalyst the market will notice | Actions the owner takes |
| Valuation’s role | Price vs. implied expectations | Entry and exit price, debt capacity |
| Key outputs | Expected value, downside, catalyst date | IRR, MOIC, downside case returns |
| Main risk | Being right too late | Overpaying or missing the plan |
Both formats end in a written recommendation. Keep the documents apart: an investment memo makes the case for a transaction, while a model memo documents the analytical logic behind the model. Our guide on structuring a model memo that survives IC review covers the second. For timed cases, see how to build and defend a model in a 72-hour case.
When Are Banker-Grade Valuation Skills Enough?
Valuation skills without a thesis are enough when the job is to build and defend a model for someone else. Examples are an IB modeling test, a fairness or valuation analysis, or a client pitch. The thesis becomes the test once you are asked to commit capital or tell someone else to.
In an IB modeling test, model quality decides. Most candidates who fail a bank modeling test do not fail on the DCF. They fail because the file looks like it was built by someone who never worked in a transaction environment. Building a test-ready model is a separate skill from forming a view.
Self-study covers the concepts in this article: reverse DCF, scenario weighting and pitch structure. What is harder to learn alone is defending a view while someone pushes back on every assumption. That takes practice in front of a skeptical listener.
FAQ
Is a target price an investment thesis?
No. A target price is a valuation output. The thesis is the reasoning behind it: why the market is wrong, what closes the gap, and what would prove the view wrong.
Can a stock be undervalued and still be a bad investment?
Yes. A stock can trade below your estimate of value while the business deteriorates or no catalyst closes the gap. A cheap price without a reason for change is a screen result, not a thesis.
What is the difference between an investment memo and a model memo?
An investment memo argues for making an investment. A model memo documents the assumptions, structure, outputs and limits of the model behind it. Its reader is a reviewer who checks it line by line.
Do you need a full three-statement model for a stock pitch?
Usually not. A driver-level build of the few variables that move value, compared with what the price implies, carries most pitches. A full model helps when the thesis depends on balance sheet or cash conversion effects.
Do investment bankers form investment views?
Bankers form views on price, deal feasibility and buyer appetite for their clients. Their deliverables are advisory. A fairness opinion judges whether a price is fair to a defined group of holders. The Leaf Group proxy, for example, states that its opinion is not a voting recommendation.
Conclusion
A valuation tells you what a company could be worth. Only a thesis tells you why the price should move and what would prove you wrong.
Start with your last valuation:
- Run a reverse DCF at today’s share price.
- Write down where your view differs from what the price implies, and why.
- Weight three scenarios and state the downside first.
- Name one catalyst and one kill criterion.
Free 30-Minute Buy-Side Case Consultation
Building a valuation is a skill you already have. Defending a view under an interviewer’s questions is a different one. Talk directly with Samuel Schiano about where you stand.
- An assessment of how your current pitch or case holds up as an investment view
- 2–3 concrete next steps before your interview or case deadline
- An honest recommendation: which package fits, or whether self-study is enough
Book Your Free Consultation → Calendly
About Financial-Modeling.com Financial Modeling LLC (Financial-Modeling.com) provides Excel-based financial modeling training in New York City and live online. Active finance professionals teach 1-on-1 or in small groups of up to five participants, and every model is built from a blank Excel sheet. Programs cover three-statement, DCF, LBO, M&A, and credit models for students, career switchers, and analysts, as well as for corporate FP&A teams, boutique investment banks, and private equity firms.
This article is for educational purposes only and does not constitute investment, legal, or tax advice. The company and all figures in the worked example are illustrative.
Last updated: October 7, 2026
Sources
- Leaf Group Ltd., Preliminary Proxy Statement (Schedule 14A), April 23, 2021, SEC EDGAR: https://www.sec.gov/Archives/edgar/data/1365038/000110465921054452/tm212618-2_prem14a.htm
- FINRA Rule 5150, Fairness Opinions: https://www.finra.org/rules-guidance/rulebooks/finra-rules/5150
- Aswath Damodaran, “January 2019 Data Update 9: The Pricing Game,” Musings on Markets, February 22, 2019: https://aswathdamodaran.blogspot.com/2019/02/january-2019-data-update-9-pricing-game.html
- Kevin Harris, “Professor Aswath Damodaran on Valuation,” Forbes, July 17, 2018: https://www.forbes.com/sites/kevinharris/2018/07/17/professor-aswath-damodaran-on-valuation/
- Mark K. Bhasin, CFA, “Book Review: Expectations Investing” (Mauboussin and Rappaport, revised 2021 edition), CFA Institute Enterprising Investor, January 14, 2022: https://rpc.cfainstitute.org/blogs/enterprising-investor/2022/book-review-expectations-investing
- Microsoft Support, “Use Goal Seek to find the result you want by adjusting an input value”: https://support.microsoft.com/en-us/office/use-goal-seek-to-find-the-result-you-want-by-adjusting-an-input-value-320cb99e-f4a4-417f-b1c3-4f369d6e66c7
- Featured image: AI-generated illustration created for this article. It does not show real people or a real company model.