A friend of mine runs a small biscuit business out of Kolhapur. Started it with her mother during the first lockdown, mostly to have something to do. By the second year, people were ordering from as far as Pune. By the third, she had a decision to make.

A distributor approached her and said he could get her into thirty stores across Maharashtra. The only condition was that she'd need to extend shelf life from three months to nine. She asked what that would take. He explained. She called me afterward and said – and I remember this clearly – "It would still be called the same thing. It would still look the same. But it would not be the same thing."

She turned it down. He didn't understand why. I'm not sure she could fully explain it to him. But she knew, the way some people just know, that the moment you make that trade you've also made a different choice about what kind of thing you are. And you can't unmake it by keeping the original name on the pack.

I've been thinking about her a lot lately.

There's a reason most things taste the way they do, and it's not because the people who make them don't know any better. It's because doing it better costs something – time, margin, the ability to scale in the ways that make a business legible to investors and distributors and every spreadsheet designed to evaluate whether a food business is working. The economics of food at scale reward a specific set of decisions. And those decisions, made consistently over time, are why a lot of things that used to taste like something now taste like a version of themselves.

This is not a new observation. Anyone who's eaten their grandmother's cooking and then gone looking for something like it in a supermarket already knows. The thing you're looking for isn't there. What's there is something that shares a name and a rough shape with the thing you remember, but the resemblance ends somewhere around the ingredients list. You just tend not to look at that part.

What I find genuinely difficult to explain to people who haven't worked in or around food is how many of these decisions are invisible from the outside — and how many of them are made not in some dramatic boardroom moment but in the middle of a quiet, ordinary procurement conversation where the cheaper option is sitting right there on the table and nobody would know the difference. Not immediately. Maybe not for a while.

Palm oil is cheaper than groundnut oil and more stable on the shelf. Refined flour behaves more predictably than stone-ground. Synthetic flavouring is consistent in a way that the real thing — which depends on season and source and a dozen variables nobody fully controls — simply isn't. None of these substitutions are secret. They're just normal. They're how most things at most price points in most categories end up being made, because the alternative requires a different kind of stubbornness. Not dramatic, not heroic, just the quiet, recurring decision to not take the easier path today, knowing you'll have to make the same decision again tomorrow.

And there's a particular kind of cost to this that doesn't show up in any accounting. It's the cost of being slow when everything around you is optimised for speed. Of being expensive in a market where price is the most legible signal of value. Of producing something that can't fully explain itself in thirty seconds on a shelf, because the thing that makes it what it is lives in a process, in a sourcing relationship, in a years-long decision about what you're not willing to compromise on, none of which fits on a label.

The people who do it anyway tend not to talk about it much. That's one of the more reliable signs that someone is actually doing it, I've found. They're not making noise about their integrity. They're just getting on with the work.

I visited a sesame oil mill in Tirunelveli. Maybe two years ago, a small family operation that's been pressing oil the same way since the early seventies. The man running it now is the second generation. He showed me the press, wooden, enormous, pulled by a single bullock moving in slow circles. It takes four hours to extract what an industrial cold press machine does in forty minutes. I asked him what the difference was, practically. He said the slower extraction keeps the temperature lower, which keeps certain compounds intact, which is why his oil has a smell and a depth that machine-pressed oil usually doesn't. Then he said something I've thought about many times since. He said, "The machine doesn't know what it's losing."

He sells to a small and loyal set of customers. He's not trying to grow beyond what this press can do in a day. A business school would look at his operation and see inefficiency. He looks at it and sees a decision he made a long time ago about what he was willing to be.

Sweet Karam Coffee uses 18% ghee in the Mysore Pak. The category average is somewhere around six to eight. They put it on the pack because it's one of those numbers that does the explaining that words can't quite do. But what that number doesn't capture is what it means to hold that line on an ordinary day when there's no campaign running and no one watching, when the temptation to adjust is just a quiet background hum that comes with every procurement conversation. The price of doing things right is that you pay it the most when no one's looking. That's also the only way it means anything.

My friend in Kolhapur still makes her biscuits the same way. Still three months' shelf life. Still the same butter, the same proportions, and the same process her mother worked out in that first lockdown summer. She told me recently that she's found a different kind of distribution. Smaller, slower, people who want the actual thing. "I had to accept that I'm not for everyone," she said. "And that's not a failure. That's just information about what I am."

She seemed fine with it. More than fine, actually. She seemed like someone who had stopped having a fight with a decision she'd already made.

That's the thing about doing things right. And I don't mean right in some moral sense, just right in the sense of true to what the thing actually is. Once you've genuinely committed to it, a certain number of problems stop being problems. They become the shape of the work. The distributor who doesn't understand becomes someone who's simply not for you. The cost becomes just the cost. The slowness becomes just how long it takes.

It doesn't make it easy. But it makes it simpler. And simpler, when you're in the middle of a long thing that you've decided to do properly, is worth quite a lot.