From IM to Indicative Offer: financial modeling from zero

This is the first post in a short series on private-equity modeling for people with a quantitative background and no finance vocabulary — which is exactly where I started. No prior accounting knowledge is assumed; every term is defined the first time it earns its keep. By the end you should be able to read a company's headline financials, fill in a one-page deal sheet, and understand what a first offer actually commits to.

The P&L is a pipeline

A company's income statement (the P&L, profit and loss) is best read as a pipeline of subtractions. Each stage removes one category of cost, and each intermediate result has a name and a job:

  Revenue                      what customers paid
− Cost of goods sold           what the product itself cost
= Gross profit
− Operating expenses           salaries, rent, marketing, R&D
= EBITDA                       earnings before interest, taxes,
                               depreciation and amortization
− Depreciation & amortization  yesterday's capex, spread over time
= EBIT                         operating profit
− Interest                     the capital structure shows up
= EBT, − taxes = Net income

Two of these stages matter far more than the rest in deal work.

EBITDA is the profit of the operations, before anyone asks how the company is financed (interest), where it is taxed, or how old its machines are (D&A). That independence is the point: two identical businesses with different owners, debt levels and depreciation schedules print the same EBITDA. It is the closest thing the P&L has to an invariant of the business itself — which is why buyers quote prices as multiples of it.

EBIT re-subtracts depreciation and amortization. D&A is not a cash cost this year — the cash left when the machine was bought — but it is a real economic cost: machines wear out and must be replaced. So EBIT answers a different question: what does this business earn after paying for the wear on its asset base?

The gap between the two is the tell. For a software company, EBITDA ≈ EBIT and nobody argues. For a machine shop with heavy capex, the gap is wide, and pricing it on EBITDA as if that gap did not exist is how people overpay. The one-line summary a mathematician will appreciate: EBITDA is a cash-flow proxy that is only valid when capex is small; EBIT is the amortized truth.

And the standing warning, because every practitioner repeats it for a reason: EBITDA is not cash flow. Cash is EBITDA minus capex, minus the working capital the growth consumes, minus taxes. A business can grow its EBITDA and bleed cash the whole way.

Enterprise value vs equity value

The second piece of vocabulary is the difference between the value of the business and the value of its shares:

Enterprise value (EV)  =  value of the operating business,
                          independent of how it is financed
Equity value           =  EV − net debt
net debt               =  debt − cash

The bridge is just a change of coordinates. The business generates the cash flows; debt and equity are claims on them, in that order. When someone says "the company traded at 8× EBITDA", they mean EV = 8 × EBITDA — and if the company carries debt, the sellers' shares are worth correspondingly less: the buyer takes the obligations over, so they come off the price.

This is the single most common beginner confusion, so once more as a formula with the toy numbers we use below: EV = 8 × 18 = €144m, net debt = €35m, so the equity changes hands at €109m — not €144m.

From IM to IO: where the numbers come from

A sale process has a rhythm. The sell-side bank sends a teaser (one anonymous page). If it is interesting, you sign an NDA and receive the IM — the information memorandum, a 60–150 page book: the company's history, market, management, customers, and crucially its historical and projected financials. On the basis of the IM alone — no data room yet, no management meetings — you submit an indicative offer (IO): a non-binding letter saying what you would pay and under which assumptions.

The work between those two documents is, at its core, filling in one page.

The deal sheet

Every firm formats it differently, but the first page always carries the same fields, and each one comes from a known place in the IM:

Then the arithmetic that turns it into a price. Try it — the sheet below is live, pre-filled with a toy IM (call it Alpenfeinkost GmbH: a specialty food producer, €120m revenue, €18m reported EBITDA, €6m D&A, €35m net debt). Change any input and everything below the line recomputes:

Reading the computed rows: 15% EBITDA margin is a decent business; the EBIT of €12m against €18m EBITDA says the asset base is real (a third of operating profit is machine wear — this is not software); and at the comps multiple of 8× the enterprise is worth €144m, of which €109m would go to the sellers.

What the IO actually says

The indicative offer letter is short, and almost everything in it hangs off the sheet above:

1. The number — usually an EV range ("€135–150m on a cash-free, debt-free basis"), not an equity number: you price the business, and the bridge is applied at closing.
2. The basis — which EBITDA you priced ("8.0× FY2025 adjusted EBITDA of €18.0m"), so everyone can see your arithmetic.
3. Assumptions — normal working capital at closing, the adjustments being substantiated in diligence, no surprises in the data room.
4. Financing and process — how you would fund it, what diligence you need, your timeline.

Non-binding means exactly that — but your reputation compounds across processes, so the number must be one you could actually close at.

What this deliberately leaves out

Everything here prices a business as a static object. The actual PE trade — buy with leverage, grow, pay down debt, sell — is a dynamical system, and its return decomposes into three cleanly separable terms. That is the next post, and it comes with a model you can drag.