Trust is cheaper than control, until it isn't

Trust is free.

Right up until the invoice for it arrives.

I've been doing the maths in my head for weeks. We're working with more freelancers now, and every time I hand something over I face the same choice. Do I trust, or do I check?

Checking feels responsible. Trusting feels lazy. I'm starting to think it's the other way round. Most of the time.

The cheapest tool in the company

Control has a cost nobody writes down.

Every review is an hour. Every "can you send me what you've got so far?" is an interruption for them and a task for me. Every second opinion I give on something that was fine is time I didn't spend on a client, or on the product.

And there's a hidden cost. People who feel watched stop thinking. They do exactly what you said and nothing more. Why would they? You're going to change it anyway.

Trust, on the other hand, scales. It costs me a conversation at the start and a look at the end. In between, I get my time back. And the other person gets room to do the job properly, which is usually why I hired them.

For a two-person company that grows on clients' money, that's not a philosophy. It's arithmetic.

Where the arithmetic breaks

The problem is that trust is cheap on average and very expensive in particular.

Most tasks can go a bit wrong and nobody gets hurt. A colour is off. A text needs another pass. You fix it, you move on.

Some tasks can't go wrong at all. Anything that touches what a client charges or what a customer pays. Anything a client launches on a fixed date. Anything where the only person who can fix it is the one who built it.

If you trust those the same way you trust a banner, you're not being generous. You're being careless with someone else's business.

Trust is the cheapest tool you have. Just don't use it where a mistake costs more than the time you saved.

So, which is it?

I don't have a clean rule. I have one question I'm trying to ask before handing anything off.

If this goes wrong, who notices first?

If the answer is "me", I trust. I'll see it, I'll fix it, the world keeps turning.

If the answer is "the client", or worse, "the client's customer", I don't stop trusting the person. I just put some control around the work. A clear definition of done. A test before it goes live. A second pair of eyes on the risky bit, not on everything.

That's the part I got wrong for years. I thought trust and control were about people. Do I trust this freelancer or not?

They're about the work. I can trust someone completely and still want to check what touches money. That isn't suspicion. It's the same thing I'd want someone to do with my own code. Honestly, my own code is where I'd start.

The price of getting it right

Getting the balance right is boring. No heroics, no all-nighters to fix something that slipped through.

That's how you know it's working.

Trust where mistakes are cheap. Control where they're expensive. And try not to confuse the two because you're tired or in a hurry, which, I suspect, is when it always happens.

Trust is still the cheapest thing in the company.

It just isn't free.

P.S. If you ever get a "can you send me what you've got so far?" from me, it's not you. It's the money.