Methodology & about

SimulateFin is a free, browser-based financial modelling tool for software businesses. It exists because the two questions it answers — how long does the cash last and is it cheaper to build this or buy it — are usually answered in a spreadsheet that nobody else can check. Everything here runs client-side; no input you type is transmitted or stored.

Who this is for

Three situations, mostly. A founder working out how long the current cash actually lasts once growth, churn and hiring are all compounding against each other. An engineering leader being asked whether to license a platform or build the equivalent internally. A finance or operations lead sanity-checking a plan somebody else built, who wants to see which assumptions the answer is really resting on.

It is deliberately not a replacement for a proper financial model. It has no tax treatment, no cap table and no scenario-weighted forecasting. What it does is make the handful of variables that dominate a software business — retention, gross margin, payroll, acquisition cost and the cost of the software you run on — visible and adjustable in one place, so the shape of the trade-offs is obvious before anyone opens a spreadsheet.

How to get something useful out of it

Change one input at a time. A model that compounds monthly rates over sixty periods is extremely sensitive, and moving three sliders at once tells you nothing about which one mattered. The most informative single experiment is usually churn: drop it by a point, change nothing else, and watch what happens to break-even, runway and year-five ARR simultaneously.

Then read the conservative and aggressive columns rather than the middle one. The spread between them is the honest output of any five-year projection. If a decision looks right in the realistic column but ruinous in the conservative one, you have not found an answer — you have found the assumption that needs verifying before you commit.

Why five years

Sixty months is long enough for compounding to separate genuinely different strategies — a one-time build cost against a per-seat licence, say — and short enough that the assumptions are not pure fiction. Beyond about five years, the error bars on monthly churn and growth are wide enough that the projection stops describing a business and starts describing the arithmetic of exponentials.

How the projection is computed

1. 1

   Each of the 60 monthly periods is computed in sequence. Recurring revenue carries forward, grows by the organic rate, gains the revenue bought by last month re-investment at your stated CAC and ARPA, and loses the churn rate.

2. 2

   Gross profit is revenue multiplied by gross margin. The operating cost base is engineering payroll and contractors (both indexed by salary inflation), fixed cloud spend indexed by price inflation, variable cloud spend proportional to MRR, and the licence or build-and-maintain cost of your chosen software path.

3. 3

   Operating profit is gross profit less that cost base. A share of any positive operating profit is spent on acquisition; the remainder moves the cash balance. Break-even, payback, runway and minimum cash are read off the resulting path rather than estimated from a single month.

4. 4

   The three scenarios apply fixed multipliers to growth, churn and cost so the columns stay directly comparable: conservative dampens growth to 60% and raises churn 35% and costs 12%; aggressive lifts growth 40%, cuts churn 25% and trims costs 6%.

What the model deliberately leaves out

• Taxes and financing. No corporation tax, R&D credits, debt service, dilution or funding rounds. Profit here is pre-tax operating profit after re-investment.
• Expansion revenue. Upsell and seat growth inside existing accounts are folded into the organic growth rate rather than modelled separately, so net revenue retention is an output, not an input.
• Seasonality and step costs. Growth, churn and hiring are continuous. Real businesses hire in lumps and sign annual contracts in quarters.
• Discounting. Cash flows are undiscounted. For a decision spanning several years, compare the totals here against your own cost of capital before treating them as equivalent.

Glossary

MRR (Monthly recurring revenue)

Contracted, repeating revenue normalised to a monthly figure.

ARR (Annual recurring revenue)

Annualised recurring revenue, conventionally MRR multiplied by twelve.

ARPA (Average revenue per account)

Average recurring revenue per customer account, per month.

Gross margin

Revenue remaining after the direct cost of delivering the service.

Net revenue retention (NRR)

Retention including expansion — above 100% the base grows on its own.

Opex (Operating expenditure)

Payroll, contractors, infrastructure, licences and maintenance.

TCO (Total cost of ownership)

Every cost of a path over the full horizon, not just the sticker price.

Those are the seven that appear most often on the simulator itself. Every other measure the model computes — churn, retention, lifetime value, payback, burn multiple, runway, crossover point — is defined in full, with its formula and the assumptions inside it, in the metrics glossary.

Who writes this

SimulateFin is built and maintained by Assaf Schwartz — the projection engine, the guides and the site around them. There is no editorial team and no sponsored content: every guide is written here, and every figure quoted in one is recomputed from the same engine that powers the simulator rather than copied from elsewhere.

Corrections to the model, a metric you think is missing, or a disagreement with one of the assumptions are all genuinely welcome — the methodology above is stated openly so it can be argued with. Email info@simulatefin.com, or see the contact page.