Engagement – it’s what we’re all chasing after on social media.
If no one engages, then it’s like a tree falling in the woods with no one around to hear it fall.
Most bloggers and marketers know that getting people to click on your links is difficult — especially without running ads.
It’s perhaps hardest to get organic engagement/clicks on Twitter.
Twitter’s “feed” is probably the most crowded out of all the social sites. With 319 million monthly active users, trying to keep up with the feed and get engagement is like drinking water from a fire hose!
Today we’re going to test whether posting a photo with a link or using links setup with Twitter Cards results in more engagement.
On our Agorapulse Twitter account, we post almost 100% of the time with images on our link posts. We’re hoping that’s the right strategy but if this study tells us otherwise, we may switch!
There are many studies out there that suggest tweeting with images get more engagement.
Our friends at Buffer did a quick study on their own account and found that tweets with images received 150% more retweets.
Buffer also concluded that tweets with images got more clicks and favorites (likes).
Twitter found in its own research that photos average a 35% boost in Retweets.
Stone Temple Consulting did an in-depth study surveying 4 million tweets and concluded that tweets with images garnered more than double the retweets and favorites.
All of these stats are related to simply posting a tweet with text, a raw link and an image.
Such as this tweet below:
In this tweet clicking the photo takes you nowhere, except to maybe a zoomed in version of the photo depending on your device.
There is also a way to post a tweet that will show more information about the link post and when you click on the image you’re taken to the link. Such a tweet is using what is called a “Twitter Card”.
There are many different kinds of Twitter Cards — you can read about the different types on Social Media Examiner — the Twitter Cards I’m referring to are basically what link posts look like on Facebook — with a Featured image, Title and Description. Followers can click on any part mentioned and be taken to the link.
Unfortunately there isn’t much research out there letting marketers know if using Twitter Card formatted links get more engagement than posting simply the image with the raw link. Just a lot of theory and “best practice” tips.
That’s why I’m here doing this test!
Tweeting with an image will result in more engagement (impressions, clicks, retweets and/or favorites) than tweeting with a Twitter Card.
My thought is the Twitter Cards give up too much of the information on the blog post and followers are less likely to click to read more.
To obtain useful data for this study, I’ll be doing the following:
The Twitter accounts involved in the study:
All links used in the test will be from Agorapulse blog posts.
The posts with Twitter Cards will look like the screenshot I shared previously.
Tweets with an image and the raw link text will look like this:
To get results that we can easily apply and compare, I used data that Twitter Analytics provided for each account.
You get to that by logging into Twitter, clicking your avatar on the upper right, and then selecting Analytics.
I then chose “Tweet Activity.”
Once on this view the data I compiled was “Impressions,” “Engagement,” and “Engagement rate.”
Impressions: Number of times users saw the Tweet on Twitter.
Engagements: Total number of times a user has interacted with a Tweet. This includes all clicks anywhere on the Tweet (including hashtags, links, avatar, username, and Tweet expansion), retweets, replies, follows, and likes.
Engagement rate: The number of engagements (clicks, retweets, replies, follows and likes) divided by the total number of impressions.
I manually went through every tweet and put the raw numbers into a spreadsheet for those tweets with or without Twitter Cards then got an average per account. Then I averaged the accounts together so we can draw a conclusion.
The main number I’m focusing on is Impressions — the equivalent of Reach on Facebook — which we are all very familiar with.
According to the raw data, tweets with an image received an average of 92.67 Impressions while tweets with Twitter Cards received an average of 81.80 Impressions. This was compiled from over 400 tweets.
This means tweets using the Twitter Cards had a 11.73% decline in Impressions.
This supports much of the data mentioned earlier that tweets with images got more engagement.
However, my results found that tweets using Twitter Cards had a 39.68% higher Engagement rate, .34 versus .64.
So while less people saw the tweets with the Twitter Cards more of them actually engaged with the tweets.
My original hypothesis was: Tweeting with an image will result in more engagement (Impressions, clicks, retweets and/or favorites) than tweeting a link with a Twitter Card.
I almost feel like we ended with a tie here on this one!
Tweets using Twitter Cards got more engagement, while tweets using simply an image and raw link got more Impressions.
Is it more important that more followers see your tweets? Or more important that they engage with your tweets?
I’d argue you want more engagement.
With that being the case I’d recommend posting using the Twitter Card format in hopes of getting more clicks, retweets, likes and comments.
The biggest problem I had during this study was scheduling the tweets using 3rd party tools – which I highly recommend to stay organized – it won’t show you the Twitter Card format.
Instead, as I showed in the video, you’ll need to uncheck the “include photo” and you’ll be shown something like this:
This is a tweet scheduled using Agorapulse and removing the photo. As you can see I have no way of knowing it’s going to tweet with the Twitter Card.
Even when you simply try to tweet this directly to Twitter using the Tweet button on the blog post, you aren’t shown it’s going to format with the card:
Here is the above tweet live on Twitter:
As you can see it posted with the Twitter Card, but I didn’t see this until it went live on Twitter.
The same result would have occurred if I had tweeted the link directly on Twitter rather than using the share option on the blog post.
This is a major shortcoming when you’re trying to properly format your tweets, but there is no real way to know this is going to occur unless you test a link from the blog first.
Most major blogs have Twitter Cards setup due to the advantages it gives them.
I would recommend setting up your own blog properly with Twitter Cards if you haven’t done so already.
Subsequently I would recommend tweeting with the Twitter Card formatted tweets, rather than an image with the raw link text in the tweet itself.
Moving forward our social media team may need to adjust how they tweet so we get the most engagement possible! I hope you do as well.