Contents
The £800 campaign
A few years ago I ran a campaign that pulled in over 15,000 visitors in a month. I remember feeling chuffed with myself, watching the numbers climb, thinking this is the one I show off in the report. Then I looked at the sales figures. Four orders. Four.
The ad spend on that campaign came to close to £800. Work that out per sale and it stops looking like a win and starts looking like something you leave off the report. I did not catch it until after the campaign had already run its course, which is the part that stung.
Why the number fooled me
Since then I track cost per visitor against cost per customer on every campaign I run, not just the ones that feel like they are going well. Traffic on its own tells you an ad got clicked, nothing about whether that person had any intention of buying anything from you.
I think the reason that campaign fooled me for as long as it did is that 15,000 is a satisfying number to say out loud in a meeting. Four is not. If I am honest, I wanted the big number to mean something, so for a while I let it.
The thirty-minute check
I now run a short review once a month, roughly thirty minutes, going through which channels sent visitors, how many of those visitors became enquiries, and how many enquiries turned into paying customers. It is not complicated and it does not need new software. It has saved us from wasting money on channels that looked fine at the top of the funnel but fell apart further down, the ones that bring people to the site without bringing anyone who wants to buy.
Thirty minutes a month is a small enough commitment that I have never had an excuse to skip it.
Two client fixes
A client of mine, an HR consultant based in Manchester, had a headline that read “Empowering Businesses to Reach Their Full Potential.” It sounded fine and said nothing. When I asked her plainly what she did, she told me she helped companies avoid employment tribunals, so we changed the headline to say exactly that. No other changes were made to the page. Her enquiry rate went up by 34 percent within six weeks.
The same client had been sending eight months of paid Google Ads traffic to her About page, and when I audited the site I found 61 percent of sessions were ending there with no next step at all. Lovely team photo, a paragraph about company values, nothing telling the visitor what to do. We added one sentence at the bottom, asking if they were ready to talk about their HR challenges, with a button through to the services page. Sessions ending on that page dropped from 61 percent to 23 percent within six weeks, and enquiries from organic traffic went up by 34 percent in the same window.
Both fixes were weeks apart on the same client, and when I wrote the two numbers down together afterwards I did not love that they matched. The enquiries in both cases were coming through organic traffic, so some of that second 34 percent may well be the same people the headline change had already started pulling in, not a fully separate result. I said as much to the client rather than claim two clean wins, because a shared attribution window is exactly what my thirty-minute review exists to catch. Either way, neither fix brought a single extra visitor to the site. Both changed what the visitors already there decided to do.
One number I check before I approve spend
I still think about that four-order campaign more than the ones that went well. £800 for four sales works out to about £200 a customer, and I did not know that number until the money was already spent.
By Lilach Bullock, AI implementation consultant and fractional CMO.