The last website I was asked to audit was an online shop.
It had a phone number at the top of every page, so customers could get in touch.
Except the number that came up when you pressed it wasn’t the same as the one in the text.
And you could only click on it at one point, a thin strip between two of the digits. Tap the number itself and nothing happened.
The owner had no idea. Why would they? You know your own number. You never tap it.
What one fault costs
Let’s put some numbers on it. These are an example, not the client’s figures.
Say your shop gets 5,000 visits a week. The figure most people quote for a UK online shop’s conversion rate is 3.4%, so call it 170 orders. At an £80 average order, that’s £13,600 a week through the website.
Now add one fault that puts off one buyer in ten.
£13,600 × 10% = £1,360 a week. Around £70,000 a year.
You won’t get an email telling you. No error message lands in your inbox. The orders don’t arrive, and you put it down to a quiet month.
With moderate traffic, a fault like that phone number can cost a shop over a thousand pounds a week in missed sales.
Your customers are on their phones
About 78% of visits to retail websites come from a smartphone, according to Charle, citing Statista figures for 2025.
So your customers meet your shop on a small screen, with a thumb, often in a hurry. When they hit a snag, they don’t write to you about it. They press back and buy from the next shop on Google.
The Baymard Institute has tracked this across 50 studies. On average, shoppers abandon 70% of online baskets before they pay. When Baymard asked US shoppers why, the answers read like an audit report:
- 40%: extra costs too high (delivery, tax, fees)
- 19%: didn’t trust the site with their card details
- 18%: the site wanted them to create an account
- 17%: website errors or crashes
Speed counts too. A Deloitte study for Google found that retail sites which got 0.1 seconds faster on mobile saw conversions rise by 8.4%. Google paid for that study and it looked at big brands, so I’d use it to set a direction for your own shop.
Five checks you can do this week
You don’t need me for these. You need a phone, mobile data and half an hour. Do each one as if you’d never seen your shop before.
1. Every way to contact you works
Tap your phone number in the header, the footer and on the contact page. Before you press call, read the number on the screen. Does it match, digit for digit?
Then tap each end of the number. Does the whole thing work, or one sliver of it?
Send yourself a message through the contact form and time how long it takes to arrive.
What it costs: the customers with a question before they buy. Someone asking about sizes or delivery dates is a buyer. A wrong number sends them to a stranger.
2. Delivery costs show before checkout
Pick a product and note where you first see what delivery costs. If it first appears on the payment page, you’re losing sales there. Baymard found 12% of shoppers left because they couldn’t see the total cost up front.
What it costs: the baskets that reach the last step and stop.
3. People can buy without an account
Add something to your basket and head for checkout. If the site makes you register before it offers a guest option, change it.
What it costs: 18% of the shoppers in Baymard’s survey left for this reason alone.
4. Pages load fast on mobile data
Turn off your wifi. Open your homepage, a category page and a product page. Count the seconds until you can tap something. If you get to three, run the page through Google’s free PageSpeed Insights.
What it costs: a share of every visit, on every page, all day.
5. Nothing breaks on the way to payment
Place a real order with a discount code, then refund it. Watch for error messages, buttons that don’t respond and pages that jump about as they load. Check the confirmation email lands in your inbox and not in spam.
What it costs: the customers who were ready to pay.
Work out your own number
You need three figures from your shop’s dashboard or Google Analytics:
- Visits a week
- Conversion rate
- Average order value
Multiply them together and you have your weekly takings from the website. Then take a tenth of that. That’s what one fault costs you if it puts off one buyer in ten.
For most shops with steady traffic, the answer runs to hundreds or thousands of pounds a week. Write it down somewhere you’ll see it. “I must look at the website” becomes a cost you can weigh against an hour of your time.
What fixing costs
Less than you’d think.
The phone number on that shop was a five-minute fix once someone spotted it. Showing delivery costs on product pages, or switching on guest checkout, is a settings change on most platforms. A developer can do most of these in an hour or two each.
Speed takes longer because the causes vary: oversized images, too many plugins or a cheap host. Allow a day for a first pass on a small shop.
Set a day’s work against a fault costing you £1,000 a week, and the sums make the decision for you.
Where to start
Start with check 1. It takes two minutes, and it’s the one I found broken last week.
Then place a test order. It walks you past most of the other faults on the way.
So, business owners: how often are you checking your website on a phone, the way your customers use it? And would you have spotted that phone number?
Once a month would catch most of what I find in an audit.
If you’d like a fine-tooth comb running over your website this week, I offer a website audit with clear findings, plain English and no jargon. Book a free 30-minute website review and let’s find out what’s costing you sales.
I design, build and look after WordPress websites for small businesses across Yorkshire, from Skipton. You'll always deal with me, start to finish.
Get in touch →
