How Much Cash Should a Small Business Keep in Reserve?

I keep seeing the same answer to a question that probably shouldn’t have the same answer.

How much cash should a business keep in the bank?

Three months.

Six months.

Sometimes the range gets wider, which I suppose makes it feel more customized.

I understand why the rule survives. It’s simple. It sounds conservative. You can calculate it before the coffee gets cold.

But I keep getting stuck on the word month.

A month of what, exactly?

A quiet February? A three-payroll month? The month the annual insurance premium hits? The month your largest customer discovers that Net 30 is apparently more of a suggestion?

“Six months of cash” sounds precise until you ask the calendar to explain itself.
If you want a quick baseline, my cash runway and burn rate calculator estimates how long available cash lasts at a constant monthly net burn. It is a starting point, not a substitute for a dated cash forecast.

I’d rather know what the cash has to survive

Imagine two businesses with $300,000 sitting in the bank.

The first bills hundreds of customers every month, collects mostly by card, carries very little inventory and could slow discretionary spending fairly quickly.

The second gets 40% of its revenue from two customers. Payroll hits every other Friday. It has a warehouse lease, debt payments and a large inventory order that has to be placed before the busy season.

Same bank balance.

I wouldn’t feel remotely the same about it.

That’s the part the generic reserve rule misses. Cash isn’t just covering expenses. It’s covering the time between something going wrong and management being able to do something useful about it.

If a customer representing 20% of revenue leaves tomorrow, how long does it take to replace that revenue?

If collections stretch from 35 days to 55, how much cash gets trapped in receivables?

If sales soften, how quickly can spending actually come down?

Not theoretically. Actually.

Payroll doesn’t become variable because somebody changed a cell from blue to yellow.

The expenses that refuse to negotiate are the ones I care about

I’d start by separating expenses management can change from obligations that already have a vote.

Payroll. Rent. Debt service. Taxes. Insurance. Contractual commitments. Minimum inventory purchases. Anything else that keeps showing up whether revenue had a good month or not.

Then I’d look at timing.

This is where a monthly average can get a little too soothing.

A business might average $150,000 of monthly cash outflows and still have a week where $110,000 leaves before the largest receivable arrives.

The average isn’t wrong.

It just doesn’t have to make payroll on Friday.

I’ve always thought cash has an irritating habit of becoming most interesting at the exact moment the income statement looks least helpful.

You can be profitable for the month and still have a very uncomfortable Wednesday.

That’s why I’d rather build the reserve from an actual cash forecast than multiply average expenses by a number somebody heard on a podcast.

Then I’d make something go wrong.

Not everything. I’m not trying to recreate the Book of Revelation in Excel.

Pick the risk the business genuinely carries.

The biggest customer pays 30 days late.

A sales slowdown lasts a quarter.

The new location opens eight weeks behind schedule.

A piece of equipment has to be replaced.

A planned financing takes longer than expected.

Now watch the cash balance.

Where does it bottom out?

How much time does management have before the problem starts making decisions for them?

That number tells me more than “three to six months” ever could.

Some cash is already spoken for, even when the bank app disagrees

This is another place owners can get fooled by a healthy-looking balance.

The bank account says $420,000.

Nice.

Except $90,000 is needed for payroll and payroll taxes. There’s a $65,000 inventory order due next week. Quarterly taxes are coming. The business collected deposits for work it still has to perform. A debt payment is around the corner.

The bank app doesn’t put little name tags on those dollars.

Maybe it should. It would ruin the dopamine hit, but it would be useful.

As a mom, I recognize this accounting system. A refrigerator can look completely full until you discover half of it has already been claimed by specific children.

Business cash works like that more often than owners expect.

The question isn’t just how much money is in the account.

It’s how much of that money is actually available to absorb something management didn’t already promise it to.

I don’t think the safest company is automatically the one with the most cash

This is where I part company with the idea that more reserve is always better.

Cash creates options. I like options.

But cash sitting in a bank account also has an opportunity cost.

Maybe the company could hire the salesperson who unlocks another market. Replace equipment that’s hurting productivity. Pay down expensive debt. Buy inventory at better economics. Make a small acquisition. Distribute capital the business truly doesn’t need.

I don’t think any of those decisions are automatically better than keeping the cash.

I just think “keep more” eventually stops being risk management and starts becoming a capital-allocation decision.

Those are different conversations.

A business with extremely volatile revenue, concentrated customers and limited access to credit may rationally keep a larger reserve.

A business with recurring revenue, flexible costs, diversified customers and an unused line of credit may be comfortable with less.

Neither one is being reckless simply because the number of months is different.

The reserve should reflect the risk the company actually has, not the risk profile of an imaginary average business.

The number should come with a sentence

If management tells me, “We keep $500,000 in reserve,” my next question is why.

I’m not looking for a sophisticated answer.

I’d be perfectly happy with something like:

“That gives us enough room to carry payroll and fixed commitments for four months if our largest customer leaves, which is roughly how long we think we’d need to replace most of the volume without making panicked cuts.”

Now the number has a job.

And because it has a job, management knows when the job changes.

If customer concentration falls, maybe the reserve can fall.

If the company signs a larger lease, maybe it rises.

If a revolving credit facility becomes available, that matters too—although I’d be careful about treating undrawn credit as identical to cash. Banks have an unfortunate tendency to become more interested in risk at roughly the same moment borrowers become more interested in liquidity.

Funny how that works.

There’s nothing wrong with using three or six months as a quick starting point. Even longstanding small-business guidance commonly uses that range while acknowledging that applying one reserve rule to every company can be misleading.

I just wouldn’t let the shortcut become the policy.

The better answer is sitting inside the company’s own timing: when customers pay, when payroll clears, which commitments can’t move, what can realistically be cut, and how long management needs to recover from the thing it’s actually worried about.

Maybe that calculation lands at three months.

Maybe six.

Maybe the right answer isn’t clean enough to fit on a refrigerator magnet.

I can live with that.

I’ve spent enough time around cash forecasts to be suspicious when the calendar suddenly gets very tidy.

If you’re trying to set a cash reserve around the way your business actually collects, spends and absorbs surprises, that’s exactly the kind of question I work through in cash forecasting and FP&A consulting.