The Complete Guide to A/B Testing Blog TitlesAlways be up to Date subscribe to updates - September 2, 2015
With the success of publishers like Buzzfeed and Upworthy, people are spending more and more time on their blog post headlines, and they are also starting to realize how important it can be for driving traffic. While you may craft your post titles with attention, seldom do you TEST blog post titles to improve traffic.
Why is this so Important?
Data from Quantcast shows that Buzzfeed now averages 196.4 million unique visitors per month globally.
The roots to this kind of success were laid in identifying their core audience, the kind of content that would resonate with them, and insanely testing their headlines. Adam Mordecai of Upworthy writes 25 headlines for each post to select the best from the lot. The result— over 200 times more views.
“We’ve seen e-mail subject lines where a one word change increased click-throughs by 46%. Open rates were nearly identical and the e-mail creative was exactly the same for both versions, but click-throughs went up by 46% in the second. If the ad was sent to 2,000,000 e-mails, the winning version would lead to 17,000+ more clicks, all from changing a single word.”
Of course they were email subject lines, but they are headlines nonetheless. As seen above, the benefits of testing your blog post titles may be greater than you think, and it’s really not hard to get started. In this post, we will share with you how you can test your blog post topics and titles to uncover the very best variation that gets you more clicks to your content.
[Tweet “Don’t leave your blog post titles up to your personal intuition – test them! #cro #abtest”]
Test Blog Topics Before You Start Writing
Analyze Your Current Blog Post Titles
Your current best performing posts are a goldmine for generating more winning titles. Take a look at your best performing blog posts in Google Analytics. Pay attention to the headlines of these particular posts.
To do that simply go to:
Google Analytics > Behavior > Site Content > Landing Pages
Then filter the results to only see pages that have “blog” in the URL – to see which blog posts drive the most traffic and engagement. If you see that multiple posts you wrote on a specific topic performed really well, while other topics were not as consistent, then focus more on that high performing content.
Take note of the similarities, and if there are any patterns with the post titles that drive the most traffic. Do the same using a tool like BuzzSumo – put in your blog URL to see how many social shares all of your blog posts have gotten. For example, on analyzing the CrazyEgg blog with Buzzsumo I got this list of posts as the top performing ones:
This list can give a good idea on what kind of posts to create. Enter your competitor urls in BuzzSumo to see their best performing posts.
But… I have a caveat here. When I analyzed CrazyEgg blog with Ahrefs content explorer, I got different results.
I can confirm that the results from Ahrefs are more closer to the truth. The first post you see was written by me and I can confirm that this is the most shared post on CrazyEgg. I believe that Ahrefs offers more accurate data compared to BuzzSumo. Nonetheless, BuzzSumo is a great tool and has many useful features that I will explain later on.
The next thing you can do is try to see if there are any similarities between the titles that are performing really well. If you see that you have multiple posts focused on a specific topic that perform best, focus more on that topic. If you notice a particular blog post format that performs best, or emotional wording that works – take note and incorporate this insight into your content planning.
Let’s say that your best performing post is “Best marketing tools for dentists”. You can write additional posts revolving around the same topic like “Best marketing tools for a new dentist” or “How to market your dental site?” etc.
Share the Core Idea of Your Post On Social Media
You can test titles even before you actually publish the post. Andrew Chen follows a three step formula to determine winning post titles:
- Tweet an insight, idea, or quote
- See how many people retweet it
- If it catches, then write a blog post elaborating on the topic
Ever see his post titled ‘Growth Hacker is the New VP of Marketing’? Yeah, that was first tested as a tweet before he ever started writing a full-fledged blog post. Enough people liked the tweet, and he decided that a blog post about the topic might resonate well with his audience as well.
According to Andrew, the reason this works is because the headline gives a glimpse of the actual value of the content in the post. If you have a large number of followers on Twitter, then you should be able to split-test blog post titles easily to see which one could perform better when writing the actual headline. Don’t have a huge social following like Andrew?
[Tweet “Feedback, even from from just a few relevant people can be significant. #growth”]
The simplest way to test if a topic would resonate with your audience and perform well or not is to simply ask people in your target audience for feedback on your topic ideas. Use survey tools like Yopoll or Qualaroo to survey your audience on what kind of posts they want. Listen to how they describe their problem to craft post titles. Qualaroo even comes with template style questions that you can implement easily.
The best spot for a Qualaroo widget to pop up is when the reader has scrolled down to the end of the post. That shows that he is interested in the content and may want more.
Key takeaway: Before you dedicate time and resources to writing a blog post, find ways to test if the topic will truly resonate with your audience first. Just like you would validate a new startup idea, validate blog topic ideas.
Compile Blog Title Variation Ideas
Research Popular Blog Posts On the Topic
Analyze the blog post titles, and the content that is covered, of the most popular blog posts out there for the topic you are writing on. You can find this out by simply seeing the posts that rank on Google for the topic, or by using BuzzSumo to see the most popular blog posts on a particular topic by # of social shares. Here for example, I conducted a search for the keyword “Marketing”
BuzzSumo offers a lot more insight into the overall type of content on a blog. You can use it to analyze what kind of content is performing extremely well for your competitors. Let’s analyze the Hubspot blog. Analysis shows that their blog posts receive the highest number of shares on Twitter.
Infographics beat all other posts in terms of social media shares, followed by list posts.
They receive the least interaction on weekends. If your site is followed by people at work then it’s best not to publish on the weekends.
Finally, the longer the piece of content the higher the number of shares.
Write down how you could differentiate your title to capture more attention than these competing posts. For example one of the most shared posts on Hubspot, an infographic shows how responsive design works.
Instead of an infographic you can write a post on the same topic and see how it goes. Headlines that manage to capture attention should be ultra specific. You can make the post ultra specific for one particular subset say—
“What Facebook marketers must know about responsive web design?”
It’s specific to those advertising on Facebook —many of whom would give an arm and a leg to increase conversions in the face of mounting advertising costs. The trick is to take a trending topic and give it your unique spin.
Find the Best Keywords to Focus on with Keyword Analysis
Use SemRush or the Ahrefs Positions explorer to do keyword analysis and find the best keyword to use in your title to try to rank on Google (and other search engines) for the topic.
Analyze the similarities between the posts that drive the most traffic
Use these similarities in your blog title A/B tests. Analyze and see which posts are driving more traffic to your competition or to you with Ahrefs’ positions explorer or Semrush and improve your posts based on that. Shown below are some keywords that Upworthy ranks for.
Create a Master List of Blog Title Formats
There are so many blog post formats and ways you can position a post. You can make it a list post with a # at the beginning, a how-to post, a guide, a course, checklist, etc. One format may work great for a particular topic, but have lesser results with another topic. This is why how you position the post format is important to test.
On this post on IncomeDiary, David Aston shares the article headlines that resulted in attracting 1000000 readers to Income Diary. These ideas include:
- How to posts
- Solution Oriented posts
- Headlines that state something
- Controversial headlines
- Short headlines
- Case study headlines
- Intriguing headlines
- Ultimate guides, headlines indicating finality
Not to forget that Infographics and list posts work extremely well for Hubspot.
Key takeaway: Your A/B testing will be much easier if you have ideas for variations ready to test. Make sure that these ideas are informed ideas, and you will have a better chance of finding a variation that beats the original.
5 Testing Ideas to Uncover the Highest Performing Blog Post Title
A/B Test Post Titles on Your Blog
Either use an A/B testing tool like Convert.com to test the titles on your blog against a specific goal, or use a tool like it. How to a/b test blog post titles in WordPress? Heard of this plugin— Title Experiments Free. It’s a nifty little tool to get you started with A/B testing blogpost titles. What’s more, it’s free?
Scroll down to Settings on WordPress and click on Title Exp settings to configure the plugin.
This plugin calculates the click probability based on how many times a title is displayed, and how often it is clicked by the users. Don’t choose the time setting to be greater than 10 minutes as it can considerably slow the site down. You can choose as many different post titles as you want and see which one performs better than the rest.
Test multiple variations to see which drives more clicks. You might also want to track how long people remain on the page after clicking to see if the title is misleading or not. A good title will drive clicks and engaged readers.
A/B Test Post Titles in Your Website Footer and other places
The footer of your website is a great place to promote recent blog posts. Use your A/B testing tool of choice to setup a test on the blog titles you have in your footer or anywhere else on your website. Take this post on Analytics for example, How to Take Conversions to the Next Level with Analytics
A/B Test Post Titles via Email
When you share your new blog post to your email list, setup a few variations of your email that promote a different blog post title in the email and different email subject line. This is super easy if your email marketing tool has A/B testing capabilities built in, otherwise you will need to create two different emails and send one email to half of your list, and the other email to the other half. Then simply compare the results.
How to do that?
Set up a subject line test and create an email on the topic you are going to write the post about. Send the email with the first subject to half your list and the other email to the other half and see which one performs better. I am going to show how you can do it on Aweber but any other ESP should be able to help you do the same.
When you login to Aweber and the dashboard opens up, you will see a big green button called Broadcast. Underneath the broadcast button you will see an option for conducing split tests. You need a minimum number of 100 active subscribers to conduct the test.
When you click on that you will be able to access and determine settings for the split test. For example, the number of messages for the test.
You will now see options for what percentage of subscribers to send the email. The percentage of subscribers adds up to a total of 100% for your messages combined. For example 60:40 or 50:50. For this case it should be 50:50.
With the message editor you can choose which subscriber lists to send the message to. Research from Aweber shows that clear subject lines always outperform creative subject lines. The following clear and creative subject lines were pitted against each other:
- Are Blacklisted Link Shorteners Getting Your Emails Blocked?
- Grow Your Email List 99% Faster: How One Site Did It
- 43 Free Animated GIFs For Your Email Campaign
- Email Timing: A Look At 6 Marketers
- Email Marketers, Here’s What to Watch For in 2012
- What Do Teens Really Think Of Email?
- Customer Spotlight ? SEER Interactive
- AWeber’s AWesome Anthony A.
- Selling Digital: The Perfect Last-Minute Christmas Gift
- Getting Earth-Friendly Beyond Email
- Threadless’ Frequency Alert: Hot or Not?
- Why You Want Your Emails Filtered
Here are the metrics by which clear subject lines outperformed the creative ones:
Test Titles via Social Media Sharing
When you split test posts via social media, it’s important that you do share the posts during the best times. Why? You will get the most engagement for your posts when most people are online. Make sure that when split testing, share the two versions during relatively the same times to measure results correctly.
On this post on Social Media Examiner, Jim Belosic shows how PostPlanner split tested two different posts on Facebook to see which one garners more engagement. The first post had a headline that asked a question.
The second test had a statement with a strong word “Beware”.
As you might have noticed, only the copy of the Facebook is different. The headlines of the posts are the same. But taking hint from which one performed better, you can improve the post title. There are several different variations in which you can post to Facebook. Post with/without image or an image that has the post title or description as MarketingProfs does here.
You can also test and see posts with human faces. This can be further split tested with a face that faces the CTA. Here’s why human faces should face the CTA. A study conducted by usability expert James Breeze shows that images of faces can be used to guide people on a website. The eye tracking study shows that when the image of the baby faced the offer, more people looked at the offer.
Test Titles via Paid Distribution
Test titles on Facebook
1. Build an audience’s list
Who’s the best audience to show these post titles? The people who visit your blog regularly is the best audience to first roll out these tests to. You can do this by setting up a retargeting pixel. Here’s how to do that:
Start by building an “audiences list” of your website visitors.
Go to the Facebook Ads Manager:
Click on Tools on the top right side. Next click on Audiences from the drop down menu.
From the options, Select the button Create a Custom Audience.
A lightbox will appear with three options. Click on the option that says website traffic.
You can set the retargeting from 30 days to 90 days.
In this case we need to retarget the new blog post titles to people who have visited your website. So the first option is the best one.
Click on the cringe icon on the bottom left side to view the pixel code.
This is how the code looks like:
2. Place the code on your site
In order to identify the people who visit your website, you’ll need to insert a Custom Audience pixel onto your site. Here’s how:
Place the pixel code provided by Facebook across your web pages. To do this, copy 100% of the pixel code from the Create Web Remarketing Pixel box, and paste it between the <head> and </head> tags on all pages of your website. Once you’ve created your pixel, you can see its status. It must be installed properly on your website to show a Ready status.
Test titles on Twitter
In this post Amanda DiSilvestro shares how you can refine your post titles based on feedback gained from social media channels like Twitter.
Amanda used Bufferapp to measure the engagement that each tweet generated. She then refines her headlines based on that. The post headline shown below was a result of split testing tweets.
How to do this for your site?
Add an image with your social shares as well that has the title variation written on it. See which variation drives the most clicks, retweets, shares, etc. Be sure to share each at around the same time and day so that results are not skewed. Create different variations for each blog post title you would like to test for a particular post you are advertising. See which variation drives the best CTR, lowest cost per conversion, etc.
Other than Facebook and Twitter, you can also advertise on Outbrain, Stumbleupon, Reddit and other networks. Advertising on Outbrain can be particularly helpful as the posts are targeted to specific audiences and they like to read.
Key takeaway: Let facts inform what final blog post title you should choose. Test out multiple variations, then choose the title that the title that contributes most to the goal of your post – whether that be social shares, organic traffic, or product purchases.
Keep Optimizing Blog Post Titles Over Time
You may have chosen a winner for the new blog post title, but you should always keep a look out for new opportunities. Pay attention to how your blog post performs over time, make adjustments to the title and post content if it is no longer getting much attention.
Use a tool like Moz to see when a blog post is close to ranking for a specific keyword, if that keyword is not in the title, update the title to include the keyword to see if that improves your ranking. Moz Pro account lets you see which keywords you are ranking for along with recommendations from Moz on how to improve the post to boost position.
There are also free tools like SerpLab where you can add keywords and see their ranking. Reshare old content and continue to look for new angles to promote your content. Maybe a new trend has emerged in your industry, and an old blog post can easily be repositioned with a new title, and reshared to “catch the wave”.
[Tweet “You can find trending topics with the help of Trends24, Reddit and #Google trends. #content #cro”]
If there is more information available to update an old blog post and add more value to it, do it! You can adjust the title, reshare, and capture attention for the post all over again.
Key takeaway: Don’t just publish blog posts, do an initial round of testing and optimizing, and forget about it. Check back in as results come in and time passes. You might uncover new opportunities to improve or optimize the post to capture attention again.
Before writing posts, it’s important that you identify your core audience first. Take feedback from them on what they want to hear about. You should have a handy list of blog title variations. To do that analyze your competition and your own post titles to identify which titles are performing well.
Social media can be a goldmine to get feedback from your audience. If you have a big following then test out those ideas directly on your social media following. If not, there’s always paid advertising. Finally, don’t rely on your intuition. Small wins can have a major impact in the long term. So, get out today and start testing your headlines.
If you know someone that would benefit from reading THE Complete Guide to A/B Testing Blog Titles, feel free to share it.