Last year I wrote that the rules of starting a business have changed. Build in sixty days, monetise in sixty, and if it fails you've lost weeks, not years. That piece was about speed. This one is about the bit I got wrong.
Search my name on Companies House and you'll find a handful of dissolved companies. ScaleTrain, Spark.Insure and Unbound.Money among them. Each one a spin-up. Each one incorporated as its own limited company on day one. Each one dissolved by voluntary strike-off. No creditors, no drama, £1 companies wound down cleanly when the idea didn't earn its keep. That's the spin-up methodology working exactly as designed.
But the incorporations were still a mistake.
One company, many spin-ups
Here's what I do now, and what I'd tell anyone starting their first spin-up.
Create one limited company, or get formal approval to run under someone else's. Then every spin-up becomes a trading name of that parent. DeckPrompt is a trading name of X8IQ Ltd. So is GoUK.app. Neither needed its own incorporation, its own confirmation statements, its own accounts, or its own strike-off paperwork when the time comes.
What the parent gives you is real cover with none of the admin. Your insurance protects the spin-up, provided you tell your insurer about each trading style. Your terms and conditions and legals extend to it, because the contracting entity is the parent. One set of accounts. One confirmation statement. And when a spin-up dies, you retire a brand, not a company.
Do it properly, though. Put "DeckPrompt is a trading name of X8IQ Ltd" in the footer and on the invoices. Tell your insurer. And know the limits: regulated activity can't hide behind a trading name, and the moment a spin-up takes on a co-founder, outside investment, or serious customer money, it's earned its own Ltd. Incorporate at the point of momentum, not at the point of idea.
The other lesson
One of my dissolved companies taught me something different. Prosper was started during Covid with people I'd never met in the real world. Clever people, good intentions, and a set of problems that no document could fix, because the foundation of any founding team is trust built in person. If you can't sit in a room with your co-founders before you commit, don't commit.
Cheap to start, cheap to stop
The dissolved companies on my Companies House record aren't something I hide. They're the receipts of a methodology: test fast, kill fast, keep the learning. But test under one roof.
Stop incorporating your ideas. Incorporate your winners.