I Don't Believe in Cutting Coffee. One Fixed-Cost Audit Freed 4,000 Kč a Month.

There is a half-written blog post sitting on my laptop arguing that frugality is the weakest lever in personal finance.

I still believe most of it. Skipping a coffee does not move a FIRE date. Being worth more between nine and five does, and it isn't close. I've said versions of that in most conversations I've had about money in the last five years, usually with the quiet satisfaction of someone who thinks he's found the adults' table.

Then one Sunday last winter I ran a fixed-cost audit: insurance policies, energy contracts, mobile and internet tariffs, every recurring charge on the bank statement. By dinner I had freed up more than 4,000 Kč a month. About €160. Permanently, and without changing a single thing about how we live.

That afternoon did not turn me into a frugal person, and I don't think it caught me being a hypocrite either. It taught me that I'd been arguing against the wrong thing for years.

Frugality is a tax. A fixed-cost audit is a transaction.

Cutting coffee is not one decision. It's the same decision, made again, every morning, forever. The money you save is real, but you're renting it. The day you stop paying attention, it walks back out. That's the part almost nobody prices in: the ongoing cost of a frugal habit isn't the coffee, it's the attention — and attention is the scarcest thing most working parents own.

A fixed-cost audit has the opposite shape. One afternoon, once. Every month afterwards the saving arrives whether I'm disciplined or tired or on a plane or three weeks into a bad sprint at work. Nothing to remember. Nothing to resist.

Same money. Completely different price.

So the "latte factor is a myth" crowd, which I have been a card-carrying member of, is right and then stops one step too early. We prove that small visible spending is a rounding error, everyone nods, and then we never say where the money actually is. So people go back to feeling clever about their coffee, and the contracts keep renewing.

There's a decent explanation for why the contracts win. Prelec and Loewenstein's work on the pain of paying describes what everyone already feels: a cost hurts in proportion to how visible and immediate it is, not how big it is. A coffee is maximally visible. You hand over money, in public, for a thing you hold. A home insurance renewal is the exact inverse — invisible, automatic, annual, and delivered as a PDF you never open.

We didn't design that asymmetry. But an entire industry has learned to sit on the right side of it.

The money was not leaking out of my weaknesses

The Sunday itself was less heroic than the word "audit" makes it sound.

I'm an introvert who genuinely dislikes phone calls, so I went the fully remote route wherever it existed: comparison sites, online quotes, portals, a spreadsheet. Nobody had to be argued with. That matters, because "call your provider and negotiate" is advice that quietly excludes a lot of people, me included.

The real discovery wasn't a price. It was a category error I'd been making for years.

We had two cars and two houses insured. I had bought each of those policies separately, at the moment we acquired the thing, each time picking whatever looked like a reasonable offer that month. Four sensible decisions, taken years apart, each one fine on its own. What I had never once done is look at the four of them together as a portfolio, which is precisely what I do with every other financial thing I own. I will happily spend an evening on fund overlap in an ETF I've held for three years. I had never spent ten minutes asking whether four insurance contracts belonged at the same company.

They did. The "second car is cheaper" and "add the house and it drops again" offers that most of these companies advertise are not a trick, they're just an offer I had never been in a position to take, because I'd never presented myself as one customer with four things.

That single reframe produced about half the win.

Data: our own household audit, one Sunday afternoon.

Energy contracts were the next 30%, and the mobile and internet tariffs the last 20%. Subscriptions, the category everyone writes about, came last and smallest, which is why the step-by-step subscription audit is a good place to start and a bad place to stop. Note what is not on that chart. Not one koruna of it came from something we enjoyed. Nothing got worse. No one had to be talked into anything, and I didn't have to become a different person on Monday.

The money didn't even go where you'd expect. It didn't go into the ETF. My wife's income dropped when she went on leave, exactly as we'd planned pessimistically at our January financial date, and the 4,000 Kč covered a decent chunk of that gap. So the audit didn't buy us anything new. It bought us not having to change the plan.

For a year in which almost every other number moved in the wrong direction, that turned out to be worth more than a better return.

Nobody builds a contract that expects to be re-decided

Here's the part that pushed me from "huh, neat" to "I was wrong about this category."

The UK financial regulator studied home and motor insurance and found six million policyholders paying high or very high margins in a single year, purely for the offence of staying. Had those people paid the average price for their own risk, they'd have kept £1.2 billion between them. The FCA banned the practice, which the industry calls price walking, from January 2022, and estimated the ban would return roughly £4.2 billion to consumers over a decade. Its own 2025 evaluation concluded price walking has largely been eliminated, and noted in passing that some firms had built deliberate friction into cancelling auto-renewal, like requiring a phone call you couldn't make online. Read that sequence again: a financial regulator had to write a rule to stop companies charging loyal customers more, and the companies responded by making it harder to leave.

Citizens Advice put the UK loyalty penalty across broadband, mobile and mortgages at £1.3 billion a year, and calculated that a household paying it in all three markets loses around £1,144 a year. Separately, they found unused and forgotten subscriptions cost UK consumers £688 million in a year, up from £306 million two years earlier. Not people who couldn't afford to cancel. People who never re-decided.

And the reason we don't re-decide is old, well-documented and slightly insulting. When New Jersey and Pennsylvania set opposite defaults on car insurance in the early nineties, one cheap and one expensive, roughly three quarters of drivers in both states stayed put, regardless of which side of the deal they'd landed on. DellaVigna and Malmendier tracked nearly eight thousand gym members and found the ones on flat monthly contracts were forgoing around $600 each by not switching to the pay-per-visit option sitting right there on the price list.

Those are old studies from another country, and I'd take the magnitudes with a pinch of salt. The direction has never been in doubt.

Two footnotes I owe you. Those insurance numbers are British, and Britain now has a rule against price walking. Where I live, and in most of Europe, nobody has banned anything. I don't find that reassuring at all. If a regulator had to intervene to recover billions, then in the markets where nobody intervened, that money is still on the table and the only person who can collect it is the customer. Which is you.

The one contract I deliberately made more expensive

An audit is not a cull, and I'd have sold this whole idea badly if I pretended otherwise.

Somewhere in the middle of that Sunday I looked properly at our internet connection for the first time in about a decade. It was, functionally, the same line I'd had since roughly 2015. I'd never upgraded it, not to save money, but because I had nothing better to compare it to at home and had simply absorbed the slowness as a fact of life, the way you stop hearing a fridge.

We upgraded. It cost more than the old line and it is one of the best small purchases we've made. I would not go back to 2015 for free.

That's the version of this exercise nobody writes up, because "I reviewed my contracts and chose to spend more on one of them" is a bad headline. But it's the honest one. An optimizer running my Sunday would have kept the archaic connection and banked the difference, and would have been wrong, because the point of having money is deliberate upgrades you actually notice, not a smaller number at the bottom of a spreadsheet. Same reason we each keep a fun-money pile nobody has to justify.

You are looking for money that buys nobody anything. There's more of it than you'd think, and it is never the coffee.

The honest case against me

Three objections, and the first one is good enough that it changed how I do this.

"One act, savings forever" is too strong. It's the weakest claim in this whole article. The FCA found firms rebuilding friction as fast as rules removed it, and prices creep back at renewal because that is what renewal is for. The accurate version is that the savings recur with a low-effort annual check, not that they recur forever at zero effort. I've stopped saying forever.

Frugality does have a behavioural payoff. Watching small spending builds the habit of noticing money at all, and plenty of people found their fixed costs only because they started looking at a bank statement for coffee reasons. I still think the financial magnitude is small. I no longer think the habit is worthless, which is what my half-written blog post implied.

The audit has a ceiling and the ceiling is low. You can do this once. You cannot do it again next month, and there is no version of it that turns into a career. Income and skills are still the lever that decides whether you get to retire early, and one good Sunday does not change that. What it changes is my claim that the entire frugality half of the ledger is noise. It isn't. One part of it pays like an investment — and I'd been throwing that part out with the lattes.

The fixed-cost audit I'll actually run next January

Once a year, around the time our policies start quietly renewing themselves, I'll run the fixed-cost audit again: an afternoon checking whether we're roughly at market price. It doesn't belong in our ten-minute monthly money date, which is deliberately too short to decide anything.

It will not be 4,000 Kč again. It shouldn't be. The 4,000 was a decade of never re-deciding, collected in one sitting, and that particular reservoir only fills once. Next time I expect five minutes of reading and a shrug, and the shrug is the whole point: knowing you're roughly at a standard price is worth the afternoon even when the answer is that nothing needs to change.

If you take one thing from this, don't take the number. Take the shape of it. Go find the money that's leaving your account because of a decision you made three years ago and have never revisited, and leave your coffee alone. The savings you have to keep choosing are the expensive kind. The ones you only have to choose once are the only frugality I'd actually defend.

Mine was hiding in four insurance policies I'd never looked at in the same room.