How We Increased Our CTR by 267% on Facebook

How We Increased Our CTR by 267% with This Simple Trick

authorMáté Rauscher October 6, 2015

It was late night and rainy weather in Amsterdam when we finished our meeting. We decided to push a little harder this "content-marketing-thing". Even though we had blog and posts we struggled to get traffic, so we had to come up with something. Somewhere at the end of the meeting I asked my colleague and friend:

"Why don't we try out facebook advertising?"

We had a facebook fanpage with nearly 1200 likes, so that seemed like a good idea. However, we barely had any experience with boosting posts, and he said;

"That should not be too expensive. Let's give it a try!"

We jumped on it, right then and there... and ohh my God, we did so much stupid stuff. But you shouldn't, so here's a guide to avoid the mistakes we made.

Test, test, test

Target audience

That's something we knew we had to do, but that's also something that is easier said than done. Basically, we came up with different target audiences based on location and interest. We set up the campaigns and waited for the results. After a day or so we'd see the results and eliminate the poorly performing audiences.

That sounds great but here's what to avoid:

  1. Don't test too many audiences at once if you want to keep on budget. Instead, test the two or three most promising ones and narrow it down after a day or two.
  2. Decide if you want to have: a) a high volume of traffic on your site which generally has lower quality or b) a lower volume with higher quality. The only way to do so is to measure everything that's happening on your blog and website.
  3. Test quickly. Waiting for a bigger sample size isn't worth it, because you'll end up spending the whole budget just on testing.

Headlines

You might have heard - just like us - that the X way to write things, or Y is the perfect headline, but the reality is that there's no such thing as "the-perfect-headline". It can differ based on the topic, the audience or other things happening in the world. Therefore the only way you can figure out which is the best is to test it.

In the beginning we would craft a headline and stick with it - that is not effective. Now we've figured out a better process using the following steps:

  1. When writing a blog post I have little trick - which I've stolen from the smart guys at OneMonth. When you are writing a post you have to come up with 25 different titles. I know it sounds weird, but it it really works. You might be able to follow a framework with 10 or 15 but after that you have to be creative.
  2. I complemented this process by writing them in a Google Docs, sharing with the team and asking for opinions.
  3. Everyone can pick 2-4 favorite titles, then we choose the winner but also test out the second and third place runner ups. The ultimate winner is determined based on it's performance over a few days.
25 Headline Test in Google Docs

Images

Visuals are an important part of social media. This is what catches people's attention. But this is also the point where we made our biggest mistake. Especially with Facebook, you can: a) upload an image and write a link or b) post a link and a preset image will automatically load.

But. There's a big difference between these two.

The image you upload separately might look cool, but if someone clicks on it, it will only show a bigger image in Facebook's image viewer.

screenshot-2

We ended up getting a lot of clicks, not on our site but on the image. That was definitely not the goal.

So we learned our lesson and started to post links using the preset images. Those got more clicks, sure. (Update: I guess facebook also realized this issue so they made it easier to post these link-type images. Just copy and paste a link to share and you'll see the new feature.)

The magic trick

Changing the link's preset image is a point where most of us never get to, but then we discovered our magic trick. We started to test different images and measure their click trough rate.

Disclaimer: Being a designer helped a lot at this part.

After a few iterations we realized a 267% increase in our CTR by using a certain type of image.

screenshot-1

Why this type worked so well?

I believe there are at least 4 reasons.

  1. There's a specific font and color palette which matches our brand.
  2. Using the rule of thirds on the image makes it easy on the eyes.
  3. The title is right there on the picture, so you don't have to read anything else. This is a huge part, keeping in mind that you only have a fraction of a second to get the readers attention and get them to click.
  4. The image can set the tone for the post - so everybody knows what to expect.
Promo Image That Worked

Takeaway

Boosting posts on facebook to accelerate your content marketing engine can be an effective method, if you use it right. Here are some of the key recommendations that worked for us.

Test everything and test quickly. Use images, not as a separate image but as part of your link, and use our magic trick if you wish.

If you're not familiar with design tools, feel free to reach out and I'll be happy to send you the template I use. If you liked this strategy and it worked for you, please write about it in the comments below. Also, write down if you have other power tactics that work. To get more useful stuff like this, subscribe to our newsletter. Also, if you want to learn about App Store Optimization, we just started a free email course, make sure to sign up for that.

13 Email Hacks Not Just for Startups

suggested reading

13 Email Hacks Not Just for Startups

facebooktwitterlinkedinemailpinterest