Made-to-measure blinds UK: cost per order down 62.63%

The trap in a cheap-looking category

Impress Blinds sells made-to-measure blinds across the UK. Blinds searches look like a gift on paper: high volume, plenty of people typing, nothing outrageous per click. Spend a week reading what those people actually want and the picture changes.

Most of them are browsing prices for a room nobody has measured yet. Some are halfway up a stepladder with a tape, working out whether a bay window needs three blinds or one. A few are chasing a bargain in a town the business would rather not send a fitter to. Paid search will happily sell you thousands of those visits, and very few of them end with a room being measured.

That gap between traffic and trade is the whole problem in this category. Made-to-measure is not a shelf product. Every sale drags a survey, a slot on the manufacturing schedule and a fitter's morning behind it, so the cost of winning an order has to sit comfortably inside a margin that does not stretch far to begin with.

A bad enquiry costs you twice

In a lot of businesses, a poor enquiry is simply money you should not have spent. In made-to-measure it is worse than that. Somebody answers it, prices it, maybe books a survey for it, and only then finds out the customer was comparing against a two-for-one deal in a supermarket catalogue.

So the number that matters is not clicks, and it is not enquiries either. It is how much of the week's diary ends in fabric being cut. Everything else is activity.

Busy and expensive at the same time

The account was buying real orders. The price of them had drifted upwards, the way it does when nobody is reading the search terms closely. The busiest month on record was also an expensive one: 205.66 orders at A$23.60 each, across 27 days.

If you sell something that has to be manufactured to a customer's measurements, you already know why that combination is uncomfortable. A record month ought to feel like a good month. This one arrived with a cost of acquisition that had to be justified against every blind going out the door. The task was never to find more traffic. It was to stop paying full price for the wrong half of it.

What we took on

Three campaigns, and the windows below were picked out of the account and reported without their dates on purpose. Across the account it served 160,043 impressions, took 11,423 clicks and recorded 1,501.98 orders and enquiries. Enough solid months to judge the account properly.

We rebuilt the account around the enquiries that turn into fitted work, and wrote the ads to the job the customer wanted done rather than to a generic product page. Alongside that ran the slow part: reviewing what people typed to reach the site, then steadily removing the browsing, bargain-hunting and how-to traffic that had been quietly absorbing the budget. No single week of that looks like much. Sustained, it rewrites the economics of an account.

Cost per order down 62.63%

From A$23.60 per order in an earlier period to A$8.82 in a later one — a fall of 62.63%. The cheaper period carried only 14 days and still produced 100 orders, at a 17.12% conversion rate.

The line chart is the part worth sitting with. Dashed segments are the chart being honest about months too thin to plot, and the months either side of them show the cheaper price was not a one-off dip. The bar chart is the counterweight: the earlier period still holds the volume record at 205.66 orders, so the improvement in price did not come from a business that had gone quiet.

Worth being clear about that cheaper period as well. Fourteen days is a short month, and short months can flatter an account. It still produced 100 orders, which is the sort of output a full month would normally be pleased with, and it converted 17.12% of the clicks it paid for while doing so.

Then the quality held

A later period recorded the strongest conversion rate the account has had — 17.47%, from 169.99 orders at A$11.27 each and 44 phone calls across 31 days. Better orders, not merely cheaper ones.

Across the account's wider run, 13.15% of the visits the business paid for turned into a tracked order or enquiry. Roughly one in every eight.

• A$23.60 → A$8.82 cost per order, an earlier measured period to a later one — a 62.63% drop
• 100 orders in the cheaper period, from 14 days, at a 17.12% conversion rate
• 169.99 orders at A$11.27 each in a later period, with a 17.47% conversion rate and 44 phone calls over 31 days
• 205.66 orders in an earlier period, the highest single-month volume, over 27 days
• 1,501.98 orders and 237 phone calls across the account's wider run, at an average of A$15.43 each
• 7.14% click-through rate and a 13.15% conversion rate over that same run

What it meant for the business

One in eight is a ratio a workshop can plan around. For made-to-measure, where each sale carries a survey, a manufacture slot and a fitting, that ratio decides whether the bench is full or the fitters are sitting in the van park waiting on a diary that never filled up.

The more useful change is the shape of the account. The early peak was volume bought at a price that needed defending every month. By the last of the periods we are reporting, a far higher share of a smaller, better-qualified stream of visitors was converting — 17.47% in that month — and that is a much easier position to schedule production from. Same category, same competitors, very different arithmetic.

There is a quieter benefit as well. When the cost per order stops jumping around, you can decide how hard to push based on how full the workshop is, rather than on a guess about what next month will charge you for the same customer.

It also gives you a straight answer to a question most retailers can only estimate. Ask an owner what a new order costs to win and you usually get a shrug and a monthly invoice. Ask this business and the answer is A$15.43 on average across the account's wider run, with the good months well under that. You can price against a number like that. You cannot price against a shrug.

None of this required a bigger budget or a different product. The blinds are the same blinds. What changed is which people saw the ad, and how many of them turned out to be worth a fitter's time.

Get in touch

If you sell something that has to be measured, made and fitted, tell us what a good month looks like in your workshop. We will tell you whether your current account is capable of producing one — and if it is not, what is standing in the way.