Why We're Building Our Own Shop Rather Than Renting Someone Else's

I build the website. When we started talking about how people would buy office furniture online in the UK from us, the sensible advice was to rent a shop from one of the big platforms, pick a theme, and be trading in a fortnight. We did not do that, and I want to explain why, because it was not stubbornness and it was not because I fancied the work.

Furniture is an awkward thing to sell online

Most shopping platforms are built around a t-shirt. A t-shirt has a size and a colour, and that is the whole of it. Office furniture is nothing like that.

Take one office bookcase. It comes in thirteen finishes, four heights, with or without doors, and the shelves adjust. That is not a small colour swatch problem, it is hundreds of real product codes that each have their own price, their own weight and their own photograph. A desk adds width, depth, frame colour and which side the return goes on.

Rented shops handle this by making you flatten it. Either you list every single code as a separate product, which buries the customer, or you pick a few popular ones and quietly stop selling the rest. Both are bad answers, and both are extremely common.

What a rented shop makes you give up

The first thing to go is the specification table. Platform themes have a description box, and a description box is where a weight rating goes to die. We wanted proper tables of dimensions, seat heights, fabric grades and guarantees on every page, and we wanted them to change correctly when you pick a different size. That is the whole point of what we are doing, so it was never up for negotiation.

The second is how price is shown. We publish one price for every business, and we say so plainly. Most platform themes are built to nudge: fake countdowns, struck-through prices that were never charged, urgency badges on things nobody is short of. All of it comes as standard and all of it had to go.

The third is speed. Product pages carrying real photographs of forty fabric options can be very slow if nobody is paying attention. Building it ourselves means we decide what loads first and what waits until you scroll.

What building it ourselves lets us do

Interactive 3D models, for one. Quite a few products can be spun round on the page, so the question of what the back looks like has an answer that does not involve emailing anybody. That works because we control the page rather than asking a theme's permission.

Search that understands furniture, for another. Somebody typing 1600 white bench desk means something specific, and the search has to know that a width, a finish and a range are three different kinds of thing. It is also why the shop has a page for each cut people search for, so 160cm office desks and white office desks are addresses rather than a box you have to find and tick.

And when a supplier changes a fabric range, we can update every affected product properly instead of hand-editing four hundred listings over a wet weekend.

The honest downside

It took longer. A rented shop would have had us selling months earlier, and there were several weeks where that felt like a very expensive principle to be holding. Building your own means you own every bug, and you fix them at ten at night when a customer finds one.

I still think it was right. The alternative was a shop that could not show what we sell, on a platform that would have quietly encouraged us to behave like everybody else. Have a look round the shop and see whether it shows. If you want the part I am proudest of, open any office chair and scroll to the specification table.