How to Calculate Startup Runway From Cash, Not Profit
How to calculate startup runway: count only cash in the bank, use net burn over a representative window, and say which plan you assume. A method you can follow.
To calculate startup runway, count only cash that is actually in the bank, measure net burn over a representative window rather than last month, and project the balance forward until it crosses zero. Then state, in one sentence, exactly what you counted.
That sounds simple, and the arithmetic is. The difficulty is that every step contains a choice, and each choice can move the answer by months. This post walks through those choices in the order you meet them, so that the number you end up with is one you can defend when someone presses on it.
How to calculate startup runway: start with what counts as cash
The first decision is what goes into the opening figure, and the rule is strict. Money in the bank counts. Nothing else does until it arrives.
That excludes more than founders expect. A customer who has committed to pay but has not sent the money is not cash yet. An undrawn credit facility is not runway either. It is an option with conditions attached, and those conditions tend to tighten at exactly the moment you need it. A signed term sheet is not money. A grant that has been awarded but not paid out is not money until the date it lands, and public timelines are their own genre of optimism.
None of these is worthless. They simply belong somewhere else. Keep a separate, clearly labelled line for near-certain incoming amounts, with the dates you expect them, and never let that line merge into the balance. Every founder who has been surprised by their own runway did the merge somewhere along the way.
| Item | Counts as cash today? | Where it goes |
|---|---|---|
| Money in the accounts | Yes | Opening cash |
| A customer's committed but unsent payment | No | Separate line, with an expected date |
| An invoice sent but not paid | No | Separate line, with an expected date |
| An undrawn credit facility | No | Outside the calculation |
| A signed term sheet | No | Outside the calculation |
| A grant awarded but not disbursed | No | Separate line, until the day it lands |
Why cash and not accounting profit
The most avoidable way to run out of money is to be profitable on paper while insolvent in the bank. Revenue that has been recognised is not revenue that has been received. An invoice that has been sent is not cash, and depending on your customers it may become cash after a long wait, or never.
Runway is therefore a cash calculation and only a cash calculation. It asks what is actually in the accounts, what will actually arrive and when. If your customers pay slowly, the gap between your accounting picture and your cash picture is not a technicality. It is the thing that can end the company, and it bites hardest while you are growing, because growth uses up cash before it produces any. The companion piece on this is how to forecast revenue for runway.
Choose net burn, then choose a window
With the opening figure settled, you need a rate of spending. Use net burn, meaning what goes out less what comes in, because that is how fast the balance really falls. If the distinction is new to you, gross burn versus net burn explains why the two are not interchangeable.
Next comes the window over which you measure it. A single month is noise. An annual insurance payment, a quarterly tax bill, a delayed payroll run or one large customer payment can distort it completely, in either direction. The guide's default is a trailing three-month average, and it adds that you should know your seasonality if you have any.
There is one exception worth stating. If something structural has changed, such as three new hires or the loss of a large contract, say so explicitly and recalculate from the new base. Letting older months drag the average towards a comfortable answer is a quiet way of misleading yourself.
| Window | What goes wrong |
|---|---|
| Last month only | One annual bill or one large receipt can make the answer wrong by months, either way |
| A trailing three-month average | Works as a default, provided you know your seasonality |
| An average that spans a structural change | Old months pull the figure towards a comfortable answer |
Project the balance forward
You now have the pieces of the minimum model. Opening cash, money in by month, money out by month, and the balance carried forward until it crosses zero. The month in which it crosses is the end of your runway on those assumptions.
Notice that this is a projection rather than a single division. Dividing the balance by a rough sense of monthly spend is exactly the habit that produces the confident, wrong figures this method is meant to replace. By laying out the months, you can see timing effects that an average hides, such as a lumpy payment landing in a particular month. How to set the model up properly is the subject of how to build a runway model.
Calculate it twice
There are two honest versions of the answer, and you should work out both. Current-burn runway assumes nothing changes and answers how long you have if you carry on exactly as you are. Planned-burn runway adds the hires you intend to make and the spending you intend to add, and answers how long you have if you do what you said you would.
The second figure is the one that matters for planning, and it is the one founders tend to avoid, because it is always shorter. When the two diverge sharply, the lesson is not that the calculation is wrong. It is that the plan is a wish with a hiring list attached.
Write the answer as one sentence
The last step turns a figure into a fact the whole company can use. The playbook suggests a fixed form of words: we have a number of months, calculated on net burn over a trailing three-month average, counting only cash actually in the bank, assuming the current plan.
Each phrase in that sentence closes a gap where two people would otherwise mean different things. When everybody quotes the same sentence, the runway stops being something you argue about and becomes something you act on.
Reading the result before you trust it
Once you have a month on the page, resist the urge to announce it. Check it against a few plain questions first. Did anything in the window distort the average, such as an annual payment or an unusually large receipt? Is there a lump coming in the months ahead that the average has smoothed away? Does any part of the opening figure rest on money that has not yet arrived?
If the answer to any of these is yes, the figure is not wrong, but it needs a note beside it. A runway number with its caveats attached is far more useful than a clean one that conceals them. The caveats are exactly the places where the next conversation will start.
It also helps to look at the direction rather than only the level. A figure that was longer last month and is shorter now has told you something, even before you know why. Recording the number each month, with a sentence on what moved it, is what makes that visible. The monthly routine for doing that is worth setting up early.
What the calculation cannot do
However carefully you build it, the result is a forecast. It depends on revenue assumptions that may fail, on surprise spending and on timing that no average can smooth. It tells you what happens if the next few months look broadly like an honest reading of the last few.
That is enough to make decisions, and it is not enough to justify false precision. Quote it as a range with a direction if anyone asks how sure you are, and be more interested in why it moved than in the second decimal place.
A worked checklist you can follow today
- Write down the balance in the accounts, and nothing else, as opening cash.
- List committed, invoiced or awarded money on a separate line with expected dates.
- Add up money out and money in for each of the last three months.
- Work out net burn across that window, noting any structural change.
- Carry the balance forward month by month until it reaches zero.
- Repeat the projection with the hires and spending you plan to add.
- Write both results in the one-sentence form and share it.
That is an afternoon of work, and no step in it needs a finance background.
Common slips when calculating
- Using the best month. The window should be representative, not flattering.
- Counting money that has not arrived. Committed, invoiced, awarded and promised all mean not yet in the bank.
- Skipping the planned version. Current burn is the comfortable figure and the plan is the real one.
- Leaving the definition unsaid. Two correct figures that disagree make a company trust neither.
Once you have the figure, the next question is what to do about it. Setting trigger points in advance is covered in how to set runway alert thresholds, and the first reason any of this matters is explained in what is startup runway.
Frequently asked questions
Should I use last month's spending to calculate runway?
No. One month is noise, because a single annual payment or large receipt can distort it. Use a representative window, with a trailing three-month average as a sensible default.
Does a signed contract count as cash?
Not until the money is sent. Keep it on a separate line with an expected date, and move it into the balance only when it arrives.
Why does the planned figure come out shorter?
Because it includes the hires and spending you intend to add. It is the honest figure for planning, which is why founders tend to avoid calculating it.
How precise should the result be?
Not very. It is a forecast, so treat it as a range with a direction and pay more attention to why it changes each month than to its exact value.
The full method, with the checklists, is in The Runway Playbook, a free 31-page playbook.