A/B Testing Marketing

Your digital campaign strategy doesn’t matter

  • September 19, 2018

You did it. After weeks of brainstorming, board meetings, and amendments, you finally submitted the final version of your marketing campaign plan. The projections are perfect, the budget and resource allocation is tight, this is the campaign where you show management how digital marketing is done.

Guess what? You’re wrong, and that’s not a bad thing.

No plan survives the battlefield

Planning things is nice, it’s comforting because it makes you feel in control. But what actually happens on the field is closer to chaos than to order, because we are humans working with others humans. Because priorities get shifted, trends come and go, and audiences don’t always react the way you want them to. So how do you build a framework that survives the battlefield?

A framework for digital marketing campaigns

A framework for digital marketing campaigns replaces the campaign plan, but it is not a totally different document.

Simply put, it’s a strategy that add to the plan by incorporating a strategy to cope with failure.

What if your ads don’t get good results? What if your landing page isn’t converting? What if your leads aren’t responding? What if they’re dropping off right before the sale?

You’re prepared, because you’re A/B Testing your whole campaign.

A/B Testing as a standard

In short, A/B testing is the practice of creating a version A and a version B of a piece of content, launching both on a small scale, measuring the performance of A and B against each other, then dismissing the least performing one to focus your budget on the other.

So when your whole campaign plan is ready and you finished all of your campaign creatives, it’s time to multiply the work.

You made a banner? Make a few more. You wrote an ad per ad set on Google? Make them two or three. You made one landing page header? Do one more. Are you targeting one audience? Make segment them into 6 target audiences.

At first, your budget is spread thin across a very long campaign, but by removing underperforming version you can reshift budget of the highly performing ones.

Measuring performance across all of the campaign touchpoints is crucial: if your campaign is underperforming, there are many moving parts that may be the root, but by measuring the performance of each campaign touchpoint, you can quickly identify the culprit and fix it.

Ad performance, revisited

Now that we’re looking at a full funnel that ends in leads (and eventually, sales). We don’t only look at which ad drove the most traffic for the lowest cost. We also measure and optimize ads based on their end results.


If you stop at the CPC metric, it looks like Ad 1 is doing great, but on a sales performance level the results are very poor. In this case, your targeting may be too broad. Your ad will reach people who are curious about the ad or service, but not really interested.

If you stop at the CPC metric, it looks like Ad 2 is very expensive, but if you measure your end result, it’s generating large sales volume. It’s OK to spend more on reaching the right target if your get much better results in the end.

This kind of thinking may seem technical and complex at first, but with a bit of trial and error, you will see it is accessible. All it needs is the right mindset and process, which you can learn more about on our YouTube channel.