The thesis page is the one page that will be read. It carries three blocks, and a page missing any of them is an investment-highlights slide wearing better clothes.
The claim is the rating, the target, the horizon and the upside, stated as numbers. Vindhya Speciality Chemicals is synthetic, and every figure in these examples is computed, none of it observed. It trades at ₹1,009.22 and the analyst's base case values it at ₹1,303.92 over twelve months, which is 29.2% of upside. That's the claim, and it takes one line.
How do you work out what the market believes?
The variant perception is the block that makes the page worth reading, and it's the one most drafts leave out. State what the market believes in the same units you use for your own view. You do that with the reverse DCF: hold the valuation framework fixed, and solve for the driver set that makes the model return the traded price.
One caution travels with the technique, and it belongs in the footnote of any page that uses it. Five free drivers against one equation admit many solutions, so the market case you recover is one internally consistent driver set out of many. Pin it down. Hold every driver you aren't arguing about at consensus or at the company's own recent history, and solve only for the ones your thesis disputes. Build a market case any other way and it's your own assumption wearing the market's name.
Advertisement
Worked example: Vindhya Speciality Chemicals
For Vindhya, the model returns the traded price of ₹1,009.22 on revenue growth of 9%, 8%, 7%, 6% and 5% across the five forecast years, an EBITDA margin held flat at 18.2%, capital expenditure at 5.0% of revenue, a discount rate of 11.9% and terminal growth of 4.5%. The analyst's base case differs in five places, and only five: growth of 11%, 10%, 9%, 8% and 7%, margin rising 18.4% to 19.2% by the third year and holding, capital expenditure falling to 4.5% by the fourth year, a discount rate of 11.7% and terminal growth of 4.8%.
Now you can state the variant perception in a sentence with numbers in it. The market is paying for a company whose margin never improves. The analyst thinks a mix shift already visible in the reported segments takes the EBITDA margin up a full point by the third year. That same shift lowers capital intensity, because the growing segments need less plant per rupee of revenue.
The falsifier is the observable that would end the argument: a specific measurement, with a threshold and a date. Risks and caveats go elsewhere in the deck. For Vindhya, if the pharma-intermediates share of revenue doesn't exceed 33% by the March quarter, or if the blended gross margin falls back below 30% for two consecutive quarters, the mix thesis is wrong and the rating should go.
How can you tell if a thesis page is weak?
Here's a test you can run on any thesis page in eight seconds. Cover the exhibit with your hand and read only the words. If a reader can't restate your variant perception in one sentence with a number in it, the page has described a company instead of making a claim about it.
You've probably seen the commonest wrong version: strong growth outlook, margin expansion potential, attractive valuation versus peers, best-in-class execution. Four bullets, no numbers, nothing anybody could disagree with, and no way for the reader to work out what the price already assumes. Barbara Minto's The Pyramid Principle names the underlying fault. A list of supports without a governing claim leaves the reader to construct the argument themselves.