Most valuation writing is either a formula with no judgement in it or judgement with no arithmetic. This is what we do: where the number starts, what moves it, and the four things that change a price after we've sent it.
A valuation is not a fact about your company. It's a statement about what a specific buyer can do with it, and every buyer's number is different for structural reasons rather than personal ones.
A strategic acquirer values you at what you're worth inside their business, which is why they can pay more than anyone. A private equity fund values you at what they can sell you for in four years, with debt lifting the return. We value you on what we think the business can earn under us over the next ten years, and we plan to own it for far longer than that.
That last framing has a consequence founders don't always expect. Because we're not planning a resale, we don't need a story about multiple expansion. We need the company to be durable. Durability is mostly retention, and retention is why our number moves more on churn than on growth.
The number comes from earnings. We estimate what we can run the company at over the next ten years, discount those earnings back to today, and set an exit value against public market comparables. Because we pay all cash, the test underneath is simple: will the business pay back what we put in while we build it for the long term? Revenue multiples still matter as the benchmark the market quotes, but they check the answer rather than produce it. Every assumption we make gets written down in the letter with a reason attached.
Linear growth over exponential. Sustainable over unsustainable. Calm over urgent. Those aren't slogans, they're the underwriting.
What the company earns today: contracted recurring revenue from the billing system, less what it really costs to run, including a market salary for the founder. Not bookings, not the plan.
What we think we can run it at over the next decade. Retention drives this more than anything, then gross margin, then customer tenure, and it's the part we spend the most time on.
Those earnings are discounted back to today, with an exit multiple tied to public market comps. Concentration, founder dependency, technical debt we'd have to pay down, and restrictive contract terms all raise the discount.
The range goes out with the reasoning attached, so you can argue with the logic rather than the number. If we're wrong about your churn, tell us.
Rarely, and only for these four. A buyer whose number routinely drops during diligence is running a different process from the one they described, and past sellers will tell you if you ask.
Revenue in the data room that doesn't match the bank statements. Not usually dishonesty, usually a definition of ARR that included things it shouldn't. It still changes the price, because the price was built on the other figure.
An exclusivity clause, a change-of-control right, or a customer who can leave on thirty days' notice. This changes what's being bought, so it changes what it's worth.
The revenue export shows one customer at 40% that the summary described as a healthy enterprise base. Not fatal, and it does move the number.
A large customer churns during diligence, or a key engineer resigns. Rare, real, and we'd tell you exactly what changed and by how much rather than quietly re-cutting the offer.
The method above is only useful if you can point it at your company. These four do that, and they all show their working rather than returning a number you have to take on faith.
Put in ARR, growth, retention and margin. It returns a range with the arithmetic visible, using the same inputs we would read on a first call.
Multiples by revenue band, growth rate and retention, so you can see where your company sits before anyone quotes you a number.
Set your revenue and compare a broker, a marketplace, private equity and a direct sale in real dollars rather than percentages.
A broker earns their percentage by widening the field. This works out whether the field is wide enough already for your company.
Every tool on this site returns an estimate built from public comparables and the inputs you gave it. It is a way to check whether a number you have been quoted is sane, not a price anyone has committed to. The only prices that exist are written ones, and ours arrives with the maths attached.
Because a multiple tells you what other companies sold for, and earnings tell you what this one can produce. We model what the business can earn over ten years, with the founder paid a market salary so the figure isn't distorted by how they pay themselves. Revenue multiples are the benchmark we check the result against.
Curious is a long-term holding company that buys and grows software companies with empathy.
Join the Curious Dispatch to receive our latest thinking, news, and opportunities within our portfolio.
