How To Test Blog Elements For Conversion Optimization


 

by Rita Kreig January 16th, 2015 



Rita Kreig

Rita Kreig is the Inbound Marketer at Magoosh, an online test-prep startup in Berkeley, CA. She earned her Masters of Pacific International Affairs from UC San Diego, where she also studied nonprofit management. Rita loves all things education, marketing, and community development – be it creating and sharing new content or analyzing metrics. Follow Rita on Twitter and Google+.














You’re not normal.

What I mean is, when it comes to seeing your blog, you don’t view it like a normal, everyday reader. Sorry to be the one to break it to you.

You know how looking at your significant other day in and day out makes you immune to some of his or her more distinctive traits? You no longer notice the scar or the flamboyant hand gesture. It’s human nature to take things for granted.

Let’s go against nature for a minute.

If you’re working on your blog every day, you’re probably relying on your intuition to make key decisions. And, you’re probably defaulting to the status quo when you could be experimenting and A/B testing: question the Status Quo!

(To learn more about the difference and similarities between conversion optimization and usability, click here)

Yes, sometimes the status quo is the way to go, but wouldn’t it be nice to have the data to back it up? Certainty can be so comforting.

So let’s assume for a moment that you’ve become blind to many aspects of your blog, and that your intuition isn’t 100% foolproof. Test these four simple steps, and start looking at your blog like it’s the first time.

1. Treat Everything Like An Experiment


Copy, color, images, locations, sizes … everything on your blog can be changed in an infinite number of combinations. You can’t test them all. BUT you can keep a running list of your top priority experiments, and slowly run through them, one at a time.

What types of experiments? Well, I manage the product blogs for Magoosh, an online test prep company, and we run tests all the time. Some of our experiments have included changing copy on our HelloBar from a student testimonial to information about our money back guarantee, sending interested readers to our free trial rather than to our paid product (sounds crazy, but it just might work), and experimenting with new social widgets in the hope of increasing student engagement.

(For another HelloBar experiment see the first case study here)

Why do all this? Because, again, intuition can betray you. Experiments will help you catch it in the act. Here’s a real-life example of when I used an experiment to outsmart my intuition.

When Magoosh hired an in-house designer (hooray!), I thought, “Hey, let’s have him redo the call to action banner at the top of our GMAT Blog’s sidebar. Photoshop and design skills trump Word Art and copywriting skills, any day!”

clip_image002

clip_image004

I replaced our plain, old banner with a beautiful new one, waited a while, and … no dice. Students stopped clicking through to the product. What the what?! So here’s what I did.

2. Make Sure Your Experiment Is Semi-scientific


As with any experiment, I was very careful to gather data that would allow me to accurately analyze the results. Luckily, with Google Analytics custom campaigns, this is quite doable.

Our original GMAT Blog CTA banner was equipped with a tracked link, which allowed us to create a custom campaign in GA. (If you’re not a whiz with custom campaigns, I highly recommend using the Google URL Builder to get started.) We used this as our baseline data. Then, I created a new campaign for the new CTA banner, so that we could compare the data side-by-side.

We were hoping to see that clicks would increase, and that our custom eCommerce variables (number of transactions, revenue from transactions) would show that students were purchasing after clicking on our gorgeous new banner.

intuition can
betray you
What we discovered was that fewer students were clicking AND fewer were purchasing. Not good.

I continued to run the new campaign until we had a near-statistically significant sample size (sometimes in blogging, you don’t get an ideal amount of data). When the results didn’t improve, I called time of death. (Funnily enough, a similar change on our GRE Blog had the exact opposite results. Different people really do like different things.)

Best thing about blogging? Experiments are easily reversed. I put the old GMAT CTA banner back, and it’s performing better than ever. I love/hate it every day.

3. Let The Data Do The Talking


If a key goal of your blog is to convert readers into clients/buyers/users, then you absolutely have to check on the health of your conversion points on a regular basis. It’s the best way to see where you can maximize engagement and fix existing problems. Improving user experience is really important when it comes to creating trust in your product.

Plus, you spend so much time managing your blog – it really should bring joy to your readers. Otherwise, why the heck are you doing this for a living?!

Use Google Analytics to:


  • See which links send the most traffic to your product.

Then check to see if you can improve the user experience. Are most people clicking on a side banner widget? Test color and copy to see if you can make it more inviting. Are people clicking on a text link within a highly viewed post? Make that content really easy to find, and make that link even more interesting to click.


  • See which content gets the most views.

Then check the bounce rate on that content. Sometimes content gets a lot of views, but people click away quickly. If that’s the case, then you can experiment with ways of improving the content to keep readers interested. Respond to comments, add interesting images, include an informative video and links to related posts. Your readers will be happy to finally find the content they were looking for.


  • See which posts have the highest bounce rate.

Speaking of bounce rate, check which of your content sends readers away. Maybe it’s something that can be reworked, or retitled to be less misleading. Or maybe it’s just a dud and needs to disappear. It’s okay to say goodbye to bad content – even when it was created with high hopes.

Hopefully at this point, you’re not thinking, “Yes, I do this. What else?” (If you are, thanks for making it this far. I know you’re busy.) I have just one more point, and it’s not related to Google Analytics.

4. Borrow Someone Else’s Eyes


Mr. Potato Head-style.

Sometimes you really need feedback from a human person with feelings and opinions. Luckily, people really like sharing those things.

User testing your blog is a great way to get a better feel for user experience . Many bloggers, myself included, fall into the trap of organizing things in a way that makes sense to their brains. But, thinking like a marketer means thinking like your audience. Which, in turn, means getting out of your own head.

You can, and should, definitely perform targeted, professional user tests of specific landing page designs when you can. If you’re experimenting with new creative, you can even try a handy tool like Five Second Test, which gives you valuable information about first impressions.

But, I think it’s equally important to perform informal user tests of blogger experience as well. You don’t need a lot of resources, you just need a handful of people who haven’t seen your blog before. Watch them navigate, see where they click. Assign them a task and see if they follow the user flow you intended.

It’s a great way to remind yourself that you’re designing for other people.

Blog on, my friends. Blog on with fresh eyes.

Now read: Conversion Optimization Steps Every Marketer Needs to Take




 


How To Test Blog Elements For Conversion Optimization


The post How To Test Blog Elements For Conversion Optimization appeared first on Search Engine People Blog.

Search Engine People Blog

(377)