Be careful what you ask for. How 2 words dropped CTR 36% | Hacker News Reader

Be careful what you ask for. How 2 words dropped CTR 36%

blog.appsumo.com

47 pointsjustinchen21 comments

At the end the author makes the conclusion that this "sexy change" was sub-optimal but the problem was a copywriting issue. A large red button that says Buy It? How could anyone not guess that wouldn't get clicked much. People don't want to immediately buy something without reading about it first, especially when coming from an email.

Also: why a huge red button instead of green or any other color? Red has many negative connotations (and is also used to signify the end of a process) so it was probably the worst possible color choice. I bet if it said See It and was green or blue the CTR would've eclipsed the unstyled email very easily.

The problem wasn't going from an unsexy email to a sexy one, it was the boneheaded copywriting errors and color choices. He should find a new designer.

> why a huge red button instead of green or any other color?

Red has been known to outperform green: http://blog.performable.com/631526233/

That's true... but generally what it comes down to is how much the button stands out from the background rather than the colour itself.

On a green site, green buttons will get lost, but red buttons will stand out from their surroundings. It's all about contrast.

In our recent emails we changed the wording to “See it.”

What wording? Buy Now? It's not clear.

Yeah, it took me a while to figure out what he was saying too. He changed the "Buy Now" button to "See it" so people would be more likely to explore before making the decision to buy.

haha. true. updated.

That's an interesting point. Interesting, because upon seeing their "new" design I went "Hmm, I'll bet it was Buy Now, that has negative connotations."

But then again, I don't know anything about design so I purposely ignore it... My sense of asthetics is esoteric at best.

So perhaps i was just being cynical.

I would have liked to see more data. Step 1 - change text email to graphics every thing else stays the same. Step 2 - Change the wording. Step 3 - Leave step 2 wording and go back to ext email.

This way you can compare like for like and figure out what is actually causing the increase/decrease for the CTR.

As it stands, the article didn't include enough data to see if their changes actually fixed the situation.

Sebg

True. The overall CTR dropped. The main difference b/n the two emails was having that button. Stats showed a low # of clicks to that button vs. other links.

The heading changed. A blurb was added. The content was re-ordered. Price is now in bold. No longer any mention that the deal expires.

Believe or not, I didn't even realize the emails were advertising the same product.

Thanks for the reply. I agree with shanonm on this one in that while it looks like the "main" difference between the two emails was having the button there were actually a ton of changes made. Because the one thing that can easily be measured is how many times the button is clicked, that is the first place to start.

That said, it also looked like that was the only way to see if the changes worked. So the button itself could have made your percentage change down, but the rest of the ad make it change dramatically up, so overall the effect is up. But, if the lesson you learned was button = less CTR and didn't test the rest of the changes, it might hurt you in the long run.

Yeah the best thing to do would be to 50/50 test the email to measure the true impact. But I think the conclusion that "See it" is better than "Buy Now" is certainly reasonable.

The first thing I thought when I saw the two images was about mobile - it seems like the first version was a lot more friendly for mobile devices than the second design. Since more and more people are clicking through promotional emails like this on their mobile devices, that could also play a role in CTR.

Interesting. Did the clicks convert differently?

I think the sweetspot is a well designed email/site that is consistent with your brand AND converts well. This wasn't a fair fight between the two designs!

The clicks did convert as once we changed the button to "see it" that email outperformed the previous one.

> I asked a design buddy to sexify it

Here's your problem. Instead of doing proper A/B testing, you asked some clueless designer to "sexify it". Instead of thinking, he/she made it look like a typical spam email.

Also, most modern email clients block images by default. So your email probably looks like crap to most recipients.

The whole blog reeks of a thoroughly muddled understanding of how people buy stuff on the web (or anywhere else). I'm not going to send you money because you're 'sexy'. I can just type "thong" into Google and look at sexy all day long for free.

However I MIGHT send you money if I thought you could improve my life in some way. And if you convinced me that you could make ME sexy, I'd probably send you a lot of money.

Whether or not you or your ads are sexy is of no interest.

> Instead of doing proper A/B testing

I disagree. It is not easy, if not impossible, to go from ugly to beautiful, with A/B tests. A/B tests can make judgments on things, but they aren't going to make it much more beautiful. Making it more beautiful, and doing A/B tests, seems like the best approach.

Who gives a fuck if your email is beautiful?

The goal is to make money, not to be pretty.

But many times, making something look better gives people a better understanding and more confidence, therefore you get more money.

And many times it makes it less readable.

Beautiful and efficient are orthogonal terms. They are pretty much independent.

You still need to decide what's important though. A pretty page doesn't always perform better, and can appearance changes after the fact negate any A|B testing you just did. You can't dissociate A|B test from appearance.

Reply on news.ycombinator.com