You should be A/B testing your offer pages, but in this article I’ll explain one thing you have to keep in mind while you’re doing it.
Apparently I’ve read 13 articles from The Guardian in the past year, so they asked me to pony up. Here’s part of the message.
An erosion of democratic norms. An escalating climate emergency. Corrosive racial inequality. A crackdown on the right to vote. Rampant pay inequality. America is in the fight of its life. If you can, please make a gift today to fund our reporting in 2022.
This copy is written for people who are interested in different things than I am, which raises an interesting question.
If they were only after a sale, a clever marketer might customize this message to appeal to me — assuming they have the tech to know a little bit about me.
In fact, that could be a use case for a CDP. Track a user to see what kind of content he prefers, then change marketing copy accordingly.
If they did that, they might get the sale, but it’s likely I would quickly sour on the coverage and would cancel.
When it comes to subscription products, people used to say that sales were the domain of the marketing department, and renewals were a response to the editorial. That’s obviously not true if the marketing copy misrepresents what’s being sold!
Let’s say the site is 5 percent about something I care about — let’s take Neanderthals as an example. (I’ve always been interested in Neanderthals.) If they change the marketing copy to make me think there’s lots of content about Neanderthals, it might win me over.
I’ve wondered about this when it comes to A/B testing. Most marketers know there are tricks you can use to increase conversions. If you change the color of the button, the way images with eyeballs are oriented, the language on the call to action, …. It’s almost endless how many things you can test, and they often have statistically significant results. A clever marketer can turn a loser of a landing page into a high performer!
Which is a good thing, right?
Well, is it? Are they really buying the product?
If it’s true that nodding my head yes when I tell you something makes you more likely to agree with me, do you really agree with me?
I’m sure you get the point. There’s a difference between persuasive, accurate, clear and honest copy, on the one hand, and parlor tricks. (Note: the parlor tricks might be useful for one-time sales, but not effective for lifetime value.)
The marketer in me says I should find a way to test this, but can you imagine the operational challenges? Every time you make a variation to make a page more profitable, you’d have to note those orders separately and track some metric as a proxy for whether those people actually like the product.
Could you possibly keep track of all that? Maybe, but would it be worth it? Probably not.
Rather, I think it’s better to just keep in mind that A/B testing can influence the buy decision, but it doesn’t affect the stay decision. It’s important to use effective techniques to get the sale, but it’s also important to be very honest about what you’re selling.
Having said all this, how do you keep customers once you’ve tricked convinced them into buying! I’ve been thinking a lot about that recently. If you want to discuss it, give me a call.
