I'd charge more from day one.
The first service business I launched, I priced it at what I thought people would say yes to. Which, it turns out, was about half what I should have charged. Low prices don't make people trust you more — they make them wonder what's wrong with you. I've seen this pattern repeat itself with almost every founder I've worked with since.
There's a particular kind of Kent tradesperson who quotes low because they're worried about the competition on MyBuilder or Checkatrade, and then spends two years working twice as hard as they need to for half the margin they deserve. The irony is that the customers who find you on price alone are always the ones who cause the most friction.
If I were starting again, I'd write down the price that made me feel slightly nervous, and then charge that. You can always negotiate down in the first few months if the market pushes back. You almost never get to negotiate up once you've anchored your rate.
I'd get a real website up inside the first fortnight.
Not a beautiful website. Not a website with five pages and an animated hero and a custom font. A real one — a page that explains who you help, what you do, what it costs or how to find out, and a way to contact you. That's it.
When I launched one of my earlier businesses, I spent six weeks fussing over the site before I'd spoken to a single potential customer. Six weeks. The irony is that the version I ended up with after actually talking to people looked almost nothing like the version I'd obsessed over. The market had opinions I hadn't anticipated.
Today, I could have a credible site live in two to three days using tools like Framer or even a well-configured Squarespace — registered on a .co.uk domain for around £10 a year, connected to a free Google Business Profile, with a Stripe payment link embedded if you're taking deposits. There's genuinely no reason to wait. The website doesn't need to be finished; it needs to exist.
I'd register the limited company earlier than felt necessary.
I stayed sole trader longer than I should have on two of my businesses. The admin feels like a deterrent — filing with Companies House, setting up a separate business account, thinking about VAT — and so you put it off. But the moment you're quoting to other businesses, or taking on work above a few thousand pounds, the Ltd structure sends a signal that you're serious. Some clients simply won't instruct a sole trader for work above a certain size.
Companies House registration costs £50 online and takes about fifteen minutes. A basic business account with Starling or Tide is free to open. The first year's accountancy costs, if you use someone straightforward, needn't be more than £600 to £900. Set against the credibility it gives you with clients and the personal liability protection it provides, that's not a difficult calculation.
Mind you — I'm not an accountant, and every situation is different. The point is: don't let the administration feel bigger than it is.
I'd build a simple follow-up system before I needed one.
The gap that kills most early service businesses isn't getting the first enquiry — it's what happens to that enquiry over the following ten days. A quote goes out. The prospect says they'll think about it. You get busy with existing work. A fortnight passes. You follow up once, feel awkward about it, and let it go.
I've been guilty of this myself. In one business, I did a rough audit of enquiries that had gone quiet and worked out I'd probably left £12,000 of work on the table in a single quarter simply by not following up more than once. That's not a sales problem — that's an admin problem. A simple system fixes it.
Today that system could be as straightforward as a Notion board with a "needs follow-up" column, or an automated reminder sequence via something like Make.com that nudges you (or the prospect directly) three days, seven days, and fourteen days after a quote goes out. None of this needs to be complicated or expensive. It just needs to exist before you're too busy to think about it.
I'd pick one channel and actually work it.
Early on I made the classic mistake of being vaguely present on four platforms and properly committed to none of them. A Facebook page, an Instagram account, a LinkedIn profile, a Google Business Profile — all of them updated occasionally, none of them updated consistently enough to build any real traction.
What I'd do differently: pick the one channel where my specific customer actually spends time, and work it properly for six months before adding anything else. For most local service businesses in Kent, that means Google Business Profile first — because it's where people search when they have a problem right now — and then one social channel depending on your audience. A kitchen fitter in Canterbury gets more from a well-maintained Google profile and a handful of genuine reviews than from daily Instagram posts. A personal trainer in Deal probably gets more from a consistently active Facebook group than from anything else.
Consistency on one channel beats mediocrity across five. Every time.
I'd talk to potential customers before I built anything.
This is the one I wish someone had told me clearly at the beginning, because I learned it the hard way on more than one occasion. The thing you think people want, and the thing they will actually pay for, are almost always different in at least one important way.
Not dramatically different, usually. But different enough that the website copy you write before talking to anyone will miss the words your customers actually use. The service package you design in isolation will bundle things people don't care about and leave out things they'd pay extra for. I've watched founders spend three months building the wrong thing because they never sat across from a real customer and asked: "What's the actual problem here, and what have you already tried?"
Ten conversations, before you spend a pound on anything, will tell you more than three months of planning. That's not a figure of speech. I mean literally ten conversations, with the specific type of person you're trying to help.
What I'd keep exactly the same.
Here's the thing, though: most of the instincts that drove the early momentum in each of my businesses were right. I'd keep the habit of starting before everything was ready, because the businesses I agonised over were always slower to find their shape than the ones I launched scrappily and adjusted as I went. I'd keep the approach of doing the first few jobs myself even when I could have delegated, because that's the only way to understand where the friction actually sits. And I'd keep the discipline of talking to customers constantly — not just at the beginning, but throughout.
The mistakes were mostly about admin, pricing, and systems. The instincts about what to build, and who for, were usually decent. Most founders I talk to are the same way. The idea is often sound. It's the scaffolding around it that needs work.
The version of me starting today would move faster and worry less.
The tools available now — for building a site, taking payments, automating follow-ups, managing bookings, staying in front of customers — are genuinely better and cheaper than anything I had when I started. A service business that would have taken six months to get properly off the ground in 2005 can now be trading, with a real online presence and a working payments setup, within a fortnight.
That's not a reason to rush past the important thinking. You still need to know who the customer is, what they'll pay, and how they'll find you. But the practical barriers — the website, the card payments, the basic automation — those are solved problems now. The version of me starting today would spend less time on those and more time on the conversations that actually matter.