Best Infographics: What I Learned Analyzing 1,000 Infographics
Infographics have been a tried and true online marketing strategy for years now. In order to figure out what separates the best infographics from the rest, I decided to analyze 1,000 infographics. I analyzed these infographics and the surrounding articles programatically (to detect width, height, words, etc.) and with human labor (using Mechanical Turk to determine whether the infographics were list posts and how many charts they had).
While there are many reasons to make an infographic (shares, backlinks, traffic, etc.), I figured that they all revolve around traffic in one form or another, so I thought it was fair to rank them on an easily recognized metric: number of social shares.
Here is a summary of what I learned:
1. There is a direct correlation between the domain authority (DA) of the domain the infographic is on and the number of shares it receives.
2. Number of words in an infographic doesn’t affect the number of shares.
3. The ideal infographic width is around 650px.
4. There is no ideal infographic height.
5. The article surrounding the infographic should have no more than 200 words.
6. If you have charts, only use one or two.
7. Infographic titles should be around 65 characters long.
8. Most shares will be from Twitter and Linkedin, followed by Facebook and Pinterest
Strong Correlation Between Domain Authority and Number of Shares
Domain Authority is a score developed by Moz to predict how well a website will do in search engines. Since sites with higher Domain Authority rank higher in search engines and get more traffic, it makes more sense that they get more shares:
As you can see, there are no infographics with over 1250 shares and a Domain Authority lower than 70. While I’m sure infographics have gone viral on lower Domain Authority sites, it’s a difficult thing to do. It goes to show that with infographics (along with articles in general), where you post is almost as important as what you post.
But, Domain Authority is not enough to get a lot of shares. Content factories like The Huffington Post have plenty of infographics that don’t do too well as you can see in the chart. The chart also shows that great infographics on lower Domain Authority sites can do quite well. In today’s day and age, the bottom line is that you have to create a great infographic and promote it well.
If your sole goal is to get as many views as possible on your infographic, consider trying to get your infographic on a high Domain Authority site (make a guestographic). However, that isn’t always ideal since a lot of the time infographics are meant to draw people to your site.
Number of Words in an Infographic Doesn’t Affect the Number of Shares
There can be quality infographics with many words or few words. This infographic from Shane Barker has less than 200 words and over 1500 shares, while this infographic from Jeff Bullas has over 500 words and over 1500 shares.
I think this happens because infographics only need to make one good point. They’re meant to be bite-sized chunks of consumable information. So, you should use enough words to explain your point and no more. Some infographics require more words to explain their point and some infographics require less.
Ideal Infographic Width is Around 600px
It makes sense that infographics should ideally be around 600px, as recommended by the Infographic Design Team. If your infographic is too wide, then it will have to be shrunk in order to fit in the width of a standard blog. As you can see, it seems like infographic creators have realized this and haven’t experimented too much with different widths:
I’m surprised to see that there aren’t more infographics that are less than 500 pixels wide. As more and more people use phones to access the internet, you’d think that infographic creators would make smaller, more mobile-friendly infographics. I can think of two explanations: either infographic creators are still stuck in a desktop-only mindset or they think that designing for desktops is still optimal, despite the negative experience for mobile users.
There is no Ideal Infographic Height
When it comes to height, the data found no real correlation between height and number of shares:
My guess for why this happens is because we read articles and infographics by scrolling downwards, so it doesn’t matter if the infographic is longer than your screen height. So, if you want to create an infographic with a lot of information then expand the height, not width.
I was a little surprised by these findings because I thought longer infographics with more words would have more shares, as longer content gets more shares than shorter content. My hypothesis for this is the same as my hypothesis for why the number of words doesn’t affect number of shares: infographics are meant to be easy to digest and need just enough information and no more.
The Article Surrounding the Infographic Should Have no more than 200 Words
The common practice when it comes to infographics is to have a few words before it to make sure that the audience has some context for the infographic. But, since they are meant to be shared and posted on other sites, they should largely be self-contained. That explains why the most shared infographics tend to have no more than 200 words:
The takeaway here is to put most of your effort into making and promoting a thorough and self-explanatory infographic. The surrounding article should just be a minor introduction. This article and infographic are a great example of a self-explanatory infographic with a small amount of introductory text. The article had 150 words and was shared over 1600 times.
If You Have Charts, Only Use One or Two
This next chart surprised me a lot. I thought that most infographics used charts, but most do not contain any charts at all:
So it’s hard to derive too many conclusions about charts from the chart above, but what it does say is that charts are definitely not necessary to get a lot of shares. Since infographics are meant to be short, sweet, and to the point. Too many charts might make it harder for people to understand. Hence, the underperformance of infographics with many charts.
Infographic Titles should be around 65 characters long
According to Neil Patel, an article headline should be no more than 65 characters. The data shows that this is the case for the best infographics as well:
A lot of the reason for the 65 character limit is because of the limits of Google’s title size in search. But, this analysis only took shares into account. It seems that short, pithy titles do better in this 140-character world. Titles that are too short seem to not be able to convey enough information to generate more shares.
I decided to look into successful infographics with longer titles like
Most shares will be from Twitter and Linkedin, followed by Facebook and Pinterest
As you might expect, infographics are most shared on Twitter. What I did not expect was that Linkedin gave the second most shares. Facebook gave the third most shares, but Pinterest was not too far behind:
What this means is that you’ll want to spend most of your energy getting shares on Twitter. However, Linkedin is over twice as effective as Facebook. This is quite different from regular articles as an analysis of 1 million articles by Fractl in partnership with BuzzSumo shows. For regular articles, Facebook is responsible for 90% of shares.
Why the huge discrepancy? My guesses: Infographics can easily appear in a twitter feed, unlike in the Facebook feed. While infographics can be purely customer focused, a high percentage of them are in B2B industries, hence the appeal of Linkedin. Even though Pinterest has a lot fewer users, the visual nature of infographics allow them to have disproportionate impact on the platform.
Also, Google Plus is pretty much worthless and not even worth your time.
While a lot of the findings from the data were expected, a lot weren’t. What did you think of the findings? What shocked or surprised you about the best infographics? Do you have any questions about the data or my methodology? Are there more questions about infographics you have that I didn’t answer in this article? Perhaps you’d like a much more comprehensive study? I’d totally love to do one, but since I used mechanical turk to do some analysis (for number of charts along with other attributes that turned out to be unimportant), I was limited. Tell me your thoughts and questions in the comment section.
Also published on Medium.