Day 133: Landing Page Split Test Is Live

Yesterday I increased the ad budget again, adding another £20 to try and push sales later in the day.

We ended up at $1,176 in sales, so a solid day overall.

Ad spend came in at £487. I’d set the budget to £460, so it overspent by around £25–£30, which is pretty normal. Usually it’ll rebalance by underspending another day.

That £487 works out to about $653 in ad spend, which means profit was $523. Not quite a full double, but pretty close, and $523 profit in a day is a good result.

Today so far we’re at $451 at 11:30am. It looks low on the chart, but it’s still early, and we’ve seen before that some days crawl and some days take off. We’ll see how it goes.

January Focus Recap

The January plan is still the same two priorities. First, reduce cost per customer acquisition. Second, increase 60-day customer value (what people spend after the initial purchase, where most of the back-end sales usually happen).

To reduce CPA, I’m doing two things: adding more creative and split testing the landing page. The landing page split test is the first one to go live, and that’s what today is about.

The Split Test

The idea behind this test is simple. What happens if I strip away most of the design elements and make the page feel more like pure information, rather than a “designed” sales page.

I’ve split tested lots of other parts of the page already, so this felt like a clean experiment to run next.

Control page (current winner)
This is the existing page, with a bit of colour, a reversed-out heading, and the current layout that’s been working.

Test page (new variant)

This is the stripped-back version. It’s mostly black and white, with minimal styling, and the headline is the first thing you see (I removed the very top header).

I haven’t changed any of the wording. I’ve mainly opened up spacing around headings and sections to make it easier to read and skim. There are still a couple of colour images, but overall it’s deliberately “anti-design”.

Will it beat the control? I’ve got no idea. We’ll need the data.

How Long I’ll Run It

Unless the new page performs disastrously, I’ll likely run this for the rest of the month so we get a meaningful amount of traffic through both pages.

We should start getting early signals within a few days, but I want enough volume to make a confident call.

Next Up

Tomorrow I’ll move onto the ad creative side. I’ll be honest, I’m cautious about touching a campaign that’s working, because it can be surprisingly fragile.

So rather than heavy edits, I’ll probably duplicate ad sets and run new creative in parallel, so I’m adding variety without disrupting the existing setup too much.

Once the landing page test and new creative are running in the background, I’ll start digging into the post-purchase sequence.

That’s the big lever for increasing 60-day value, and it’s pretty clear I’m not doing enough with it right now.

That’s all for today. I’ll catch up with you again tomorrow.