START MY FREE TRIAL NO CREDIT CARD REQUIRED

How to Get More Retweets in 2017: 6 Data-Backed Ways [Infographic]

  • 0
  • July 12, 2017

get retweets

We all want to get retweets. But you know that. That’s why you’re here.

But how? What can you do differently to increase the number of retweets you get?

Well, we’ve analyzed over a billion tweets in the past. So, we thought, why not let the numbers do the talking? Today, I’ll share with you our latest study on what our big data says about how to get more retweets.

The Methodology

Considering Twitter’s algorithm change, we wanted to provide the most up-to-date data as possible. So, we examined a sample of nearly 500,000 tweets with over 3.5 million impressions from the last three months.

Each tweet in the sample got at least one retweet. We examined six aspects to see what the data had to say about what tweets that get retweeted have in common.

Here are the results:

get more retweets infographic 2

Share this Image On Your Site By Copying and Pasting the Code Below:

1) Tweet Links

The overwhelming majority of tweets in our sample (91.7%) contain a link. This make sense because Twitter is an information sharing platform. People go to Twitter specifically to discover great content. It only stands to reason that when they find it, they want to share it.

2) Tweet about Twitter

Not a shocker, but people on Twitter tend to retweet info about Twitter. 6.7% of all retweeted tweets mention Twitter compared to 1.6% that mention Facebook, and only 0.8% that mention Instagram.
Click here to subscribe

3) Tweet with Hashtags

Nearly 70% of the tweets in our sample contain a hashtag. With impressions being down from your followers because of Twitter’s new algorithm, hashtags on Twitter are more important than ever to help get your tweets seen. It stands to reason that tweets with hashtags get more retweets.

4) Tweet “You” not “Me”

“It’s not me; it’s you.” That’s backwards in a break-up speech, but just right for Twitter.

Tweets containing first-person pronouns like “me, I, our, we” etc., accounted for only 8.7% of the sample. On the other hand, tweets containing second-person pronouns like “you” and “your” accounted for 29.2% of all tweets in the sample.

No mystery here. People don’t care how great you are or what you can do. They want to know why it matters to them. Tweets with copy that speaks to people and addresses their pain points out-perform tweets with self-promotional language.

Using 'you, your' pronouns gets 4x more #retweets than 'I, me, our' pronouns! #TwitterTipsClick To Tweet

5) Don’t Tweet Emojis

This is the most disparate statistic in the study. Only 4.1% of tweets in the sample contain an emoji.

That said, take this one with a grain of salt. It’s likely that many people just don’t use emojis in tweets (yet) and that has skewed this statistic.

Exactly how much including an emoji affects retweets is hard to say, but the data suggests it may hurt. I’d advise testing this for yourself and seeing if you notice a pattern of tweets with emojis under-performing in terms of retweets.

6) Images in Tweets are Optional

This was by far the most surprising result of the study. Tweets without images slightly edged out tweets with images in terms of retweets, 53.9% vs. 46.1%, respectively.

This goes against the grain of conventional social media marketing advice. You’ll see a million stats about how tweets with images get so much more engagement. Well, it may be time to re-examine that idea.

My advice is to experiment with leaving the image out of some tweets and see how they perform. You might be pleasantly surprised.

Wrapping Up

Well, there you have it – real data on what’s working and what’s not to get more retweets now.

So, what do you think? Anything in the study that surprised you? I’d love to hear from you in the comments section below!

Author David Boutin

David is the Social Quant content Gatekeeper AND Keymaster, as well as a customer relations specialist.Follow and Connect with David on Twitter. 

More posts by David Boutin
  • Jessika Leigh

    Of course all of it is shocking, but especially #6 because it’s been nailed into our heads from various sources that images are a must at this point

    • Hi Jessika,

      Indeed that was the most surprising piece of data! Worth testing out some tweets without images 🙂

      Thanks for reading and commenting!

  • Bayorh Ojo

    thanks for sharing

    • Hi Bayorh, sure thing! Glad you found value in the post 🙂

  • David @ Taam Insight

    Hi David,

    I have a fundamental methodology question. How did you come up with the pie charts – what was the data handled? For example, tip #5 “Don’t tweet Emojis” begs the questions of whether you considered the % of retweets out of the 500k and then simply counted the % of retweet with emojis for those 500k tweets? I hope not. Because what if there were only, say 10k tweet with emojis in the fist place. Then OF COURSE the % of tweets w/Emojis retwieeted will be minuscule compared with the all set of 500k weets!!!. But if, again, for the sake of argument, 90% of the 10k tweets w/emojis are retweeted then it’s a completely different story.

    The same question holds true for pretty much all the tips? How where the 500k segmented.
    I’m sure you diidn;t make that rookie mistake but I’d like you to make sure 🙂 (here;s an amoji for you).

    Thanks!

    -David Batel
    @taaminsights.

    • Hi David,

      Per the methodology, all of the 500k tweets we looked at got at least one retweet and then we studied what elements they had in common. In the text of the post I do acknowledge that the emoji stat is likely skewed for the exact reason you stated. But, again, we looked at a pool of 500k retweeted tweets to see what they had in common for this study. Hope that answers your question 🙂

      Thanks for reading and commenting!

  • Chris D Dixon

    David there is another aspect you might consider which is TIMING of Tweets. I post around 120 times a day and have found that timing is super important. Twitter is much different than Facebook. The simple fact that many of us have huge Twitter following with multiple different time zones does play into posts being seen. I have 46,000 followers and my analytics show me that I average 300-500 impressions within the first few minutes of a post. What that tells me is just 300-500 of my followers are online at that specific time. Twitter is somewhat like radio as the posts you tweet out when someone is online are more likely to be seen than posts sent out when they are not online. I average 1.2m impressions a month which I think is decent. I also incorporate hash tags, pics, and all of them have links to articles like you suggested above. Check us out at @concealncarrynt

    • Hi Chris,

      That’s very true and I’m sure it will continue to be. But with Twitter algorithm change, have you noticed a decrease in impressions over the past few months? (We have.) I think as they change the algorithm more, you’ll see that your tweets with engagement will snowball more as they’ll be in the “In case you missed it” section people see when they open Twitter. It will be interesting to see how far the algorithm changes go in terms of visibility of tweets. I think we’re just seeing the beginning.

      Thanks for reading and commenting!

      • Chris D Dixon

        David I included my account analytics from the last 28 days. I average 1.2m impressions a month and this reports shows 1.0m. I must admit that my Twitter activity over the last 30 days has decreased some as well on my part. The biggest thing i notice about Twitter is sudden drops in followers. While I do understand there are a lot of fake accounts out there this puzzles me. In the last 3 days i have dropped 900 followers and gains 300 followers for a net loss of 1.3% of my followers. I have followed cycles like this in the past but they seem to happen about 3-4 times yearly. https://uploads.disquscdn.com/images/c0efdb104bebd3a450c054608ab36ec1de2cc14c92cb68e4f434f44f7a63b12b.jpg

        • Yeah, we see those drops every once in a while too. Best guess is it’s just Twitter cleaning out fake/deleted/suspended accounts but who know?! 🙂

      • Mohammad Tanveer

        Interesting.
        Since I launched our social campaign on Twitter in May, I have noticed a sharp decline in the number of impressions.
        From May to Early June they were very favorable, but since then– impressions are a third of what they were for us. Hmm.

  • Liz Hurst

    Fascinating data, especially the images info. Although, I think I will continue to add images to most of my posts as I think they look better. Thanks for the link!

    • Hi Liz! Thanks for reading and commenting. Having an image doesn’t seem to help or hurt (at least when it comes to retweets) so image away 🙂 – that was a surprising stat though!

  • Great post David. These are some very interesting stats and I really love the infographic.

    I’ll be testing the “you” / “me” in tweets. Thanks for the tip!

    Here’s what I found from my accounts:

    * My text tweets (especially motivational ones) without images are getting more likes and retweets than any other form of tweet

    * Images with too much text don’t get shared as much. I see a trend by the likes of Neil Patel, Inc.com etc posting tweets with images but no text in the image. I’m slowly coming round to this idea. I think it has a little something to do with micro content (where the image should say it all). I also think the younger Millennials prefer images this way

    * Some unfollows aren’t actually unfollows. They are suspended accounts. When they get their account reinstated some of them automatically follow you back. I tested this with a few accounts and it seems like the ones that get reinstated quickly start following you again. (not proven)

    * The emoji question is an interesting one. I generally don’t put them in my tweets but use them in my replies. However, Gif tweets don’t get much likes or retweets but when used in a reply they do (go figure)

    * A pinned tweet that is a video gets more engagement. Also, a tweet where there’s an image of a person looking directly at you via the lens gets more engagement.

    * I get a lot of likes for internal conversations with another account. I don’t know how that happens? I’m guessing they put my twitter handle into search to see my conversations

    Thanks. My 2c

    • Thanks and great observations, Desmond! I love GIFs and have had mixed results with them but agree they work great for replies.

      Very interesting that you’re seeing good results from text tweets without images. That was the most surprising result of the data for us.

      Great point on the images without text – if they are compelling, that may also lead people to actually read the tweets to find out the context of the image and that may help if you’re great at writing Twitter copy.

      I think you’re right about the unfollows not being unfollows – we’ve seen some of that as well.

      Let us know how the you/me test goes for you. Thanks again for reading and commenting!

  • Amitabh Songara

    Thanks for the data. I think the new analysis may change the way of using twitter. I usually include image and links and yes I have seen a good number of engagements.

    • Hi Amitabh,

      Glad you found value in the post! I do think we’ll see some big changes in Twitter in the coming months. Should be interesting. Cheers!

      • Amitabh Songara

        Agree. Every new experiment brings new changes.

  • Tweets without images is something I’m going to test! Last year we were told that images and visuals were important – which they are but this was more for engagement and I’ve seen the rise of images in my feed. Now to test for retweets! Thanks, David.

    • Hi Vatsala, thanks for reading and commenting! The image thing may be a backlash and it could be that tweets without images are actually starting to stand out more. Worth testing. 🙂

    • I would definitely test this Vatsala. I use them for motivational and positivity type posts and find they do really well.

  • Wow! Eye-opening indeed. Sad to say sayonara to my emojis and photos!

    • Thanks for reading and commenting! I wouldn’t be too quick to pull the plug on emojis. As stated in the article, that data may be skewed because simply not that many people are using emojis (yet) but it is something to keep in mind and test for yourself. 🙂

  • Thanks for digging deep in the new Twitter rollout!
    I just went back to apply all your findings in a tweet regarding this article – let’s see how it goes! 😉

    • Hey Agnelika, thanks for reading and commenting! Let us know how your experiment goes. I think this is info that should be generally applied and tested with your particular audience but still would be interested to know how it works out for you 🙂

  • David, I love the one about Tweeting about Twitter and not Facebook or Instagram. Those are for those networks…If I wanted to be on Facebook or Instagram I’d want to be there 🙂 Great reminders and I was surprised about the emoji’s – I would have though the opposite. Thanks for all the info here, great infographic!

    • Thanks Lisa! Glad you like the post! Couldn’t agree more about Facebook/Instagram 🙂

      As far as the emojis, I’d test for yourself and see if it affects things for you. Cheers!

  • Very interesting. I work at a charity, and we’re always grateful to glean these kernels of gold.

    • Thanks, Natalie! Glad you found value in the post 🙂

  • Tyler McGraw

    Did you look at how many retweets each of these 500k tweets got? Simply looking at the number of tweets containing an image or emoji or link doesn’t really mean much. Sure most tweets don’t contain an emoji but that doesn’t mean tweets with emojis don’t get more retweets. The same can be said for the 5 other categories.

    • Hi Tyler, no we did not make how many retweets a factor for this study, all tweets that got at least one retweet were examined. The idea was to look at tweets that were shared at least once to see what commonalities they shared. A deeper dive into how many times they were shared would be interesting and we’ll keep that in mind for future studies.

      Thanks for reading and commenting 🙂

  • The tip to mention Twitter in your tweets when relevant is very very true. I’ve found that discussing how any particular social channel works on that network is a useful tactic for driving conversation. Useful article!

    • Hey Brian! Thanks for checking out the post and commenting!

      Indeed it makes sense that people are interested in the platform they’re on 🙂

      Have a great week!

  • Melanie Downey

    I’m surprised about number 6, but also glad. Big time saver. 🙂

    • Hi Melanie, thanks for reading and commenting! I would definitely test number six and see how the results are for you. We were surprised by that one too 🙂

  • Pretty interesting to see the result of using emojis. Thanks David for sharing! these are really some valuable data. I was wondering, is there a specific time of the day that is best to tweet?

    • Thanks for reading and commenting, Emmerey!

      There is a lot of advice floating around about best times to tweet but I would simply experiment and analyze then let the numbers do the talking.

      That said, you might be surprised by the amount of times you can tweet in a day and how effective more tweets are. We cover that in another post: https://www.socialquant.net/tweet-activity/ some interesting stuff in there too 🙂

  • This is an interesting yet important concept for social media marketers like i am and hats off to your comparative strategy.

    • Hi John, thanks for reading and commenting! Glad you liked the post and found value in the data 🙂

  • Using relevant hashtag is one most important point to get re tweets because it is all about targeted audience

    • Great point, Jane – couldn’t agree more! Thanks for reading and commenting 🙂

  • Great analysis, next time I will be following these best practices. Thank you

2K Shares
Tweet1K
Share71
Pin115
Share121
+1101
Email
Get Started Now!

Thank you for signing up. You need to authorize Social Quant, Inc. to access the account. We need this access in order to be able to manage your following.

This application will be able to:

See who you follow.
Follow in accordance with your keywords.
Unfollow in accordance with your keywords.

This application will not be able to:

Update your profile.
Post Tweets for you.
Access your direct messages.

Authorize App »
 
“Since we started using Social Quant on our Twitter profile, our follower count has skyrocketed 40% in 40 days!! And most importantly, engagement on our tweets has risen by the same amount — which means Social Quant is finding exactly the right followers for our brand.”
— Andrew Warner, Mixergy
Get Started Now!
 
“Since we started using Social Quant on our Twitter profile, our follower count has skyrocketed 40% in 40 days!! And most importantly, engagement on our tweets has risen by the same amount — which means Social Quant is finding exactly the right followers for our brand.”
— Andrew Warner, Mixergy
Get Started Now!
Complete Sign Up » « Back
 
“Since we started using Social Quant on our Twitter profile, our follower count has skyrocketed 40% in 40 days!! And most importantly, engagement on our tweets has risen by the same amount — which means Social Quant is finding exactly the right followers for our brand.”
— Andrew Warner, Mixergy