I love to see new stats and research about how to best share to social media.

If it’s research-backed or numbers-driven, sign me up. These actionable tips are what drive a lot of our experiments at Buffer as we’re keen to see if the best advice from these studies meshes with our experience, too.

And there’s a lot of new info to go off of.

I’ve collected 10 of the latest surprising, revealing studies on social media here in this post, with takeaways and insight into social media timing, Instagram sharing, Facebook users, and more. If you’ve seen a recent study worth mentioning, I’d love to hear from you!

Social Media research

1. The peak performance of social sharing

Late afternoon to nighttime is the best time to reach people on social

Social traffic substantially underperforms overall traffic from about 5 a.m. to noon, and social substantially overperforms overall traffic from about 3 p.m. until 1 a.m.

Chartbeat reported on the data of the sites it tracks, looking at how social media sharing corresponds to site traffic. The general trend seemed to follow: Traffic and social sharing both increase throughout the early morning, peak midday, then lessen into the evening.

The unique finding here was in the subtle difference in exactly where each metric peaks.

Social traffic outperforms website traffic from 3:00 p.m. Eastern Time to 1:00 a.m.

Chartbeat social trfafic web traffic research

2. What the average Facebook user looks like

The very male, college-educated, heavily IT, somewhat liberal demographic

Only two publishers–BuzzFeed and Yahoo!–have more women than men in their audiences at 51% and 56% respectively.

Only two publishers–Forbes and Wired–exceed a 10% likelihood in their audiences working at management level.

Fractl and BuzzStream collaborated on a study of 20 publishers’s Facebook audiences, looking at the Audience Insights for publishers like The Guardian, Wired, BuzzFeed, Yahoo, Huffington Post, and more.

In the case of these audiences, the results skewed heavily in a few directions:

  • 18 of the 20 publishers had an audience that was more male than female.
  • The majority of active users on these pages has graduated from college.
  • All but one publisher had an audience makeup of more IT workers than the U.S. Facebook average.

Facebook Audience Insights for 20 major publishers

Comparisons might be a little tricky to draw between these pages and yours, though the research does point to the value of understanding your audience. My best guess at the demographics of some of these publishers would be that the audience was more female (I was wrong) and perhaps not as IT focused.

 

3. Instagram vs. Facebook

Instagram a more engaged platform than Facebook, Twitter

Instagram leads social platforms for engagement with 2.81% of audiences engaging with a post.

Locowise studied 2,500 Instagram profiles from April 2015 to measure a wide assortment of different engagement metrics and content strategies. One of the big takeaways was how engagement on Instagram far outperforms Facebook and Twitter.

Instagram research - Engagement compared to Twitter Facebook

Average engagement per post on Instagram was 2.81%.

On Facebook, engagement was 0.25%.

On Twitter, engagement was 0.21%.

(For Instagram engagement—as you can see from the graph above—the best results still come from photos versus video.)

Other interesting takeaways from the Locowise study include:

  • Likes account for 96% of all engagements (comments account for the other 3%)
  • Brands post 2.3 times per day to Instagram
  • The largest profiles post 7.24 times per day, the smallest profiles post 1.68 times
  • Average follower growth month-over-month is 1.95%, meaning that if you had 1,000 followers in March, you could expect to gain 19 new followers in April.

4. Interactions and Instagram

More interactions happen on Instagram—5 likes or comments for every 100 followers

The average interaction % on Instagram is up to 10 times higher than on Facebook.

Quintly analyzed over 5,000 Instagram accounts (and broke those accounts into buckets of followers, too) to see the current trends in engagement, content type, and strategy. One of the main takeaways from the study: Interactions are amazingly high on Instagram.

Quintly measured Interaction Rate, which is interactions per post divided by number of followers. They found that Instagram’s Interaction Rate was 4.80 interactions per 100 followers. Facebook’s rate is 0.72.

Further, Quintly also shared the average interactions per post for Instagram photos or videos, along with a breakdown of what you might expect at varying follower levels.

Quintly Instagram report - Interactions

5. Where is social media marketing headed?

Survey says Twitter, YouTube, & LinkedIn

Social Media Examiner surveyed over 3,700 marketers on their social media strategies, goals, and plans, ending up with some truly fascinating results on where social media marketing may be headed.

A significant 66% of marketers plan on increasing their use of Twitter, YouTube, and LinkedIn.

Future use - social networks via Social Media Examiner

Additional cool findings from the Social Media Examiner survey include:

  • Marketers are most keen to learn about Facebook
  • Nearly 3 out of every 4 marketers plans to increase video usage
  • Facebook and LinkedIn are the two most important networks for marketers
  • Most marketers aren’t sure their Facebook marketing is effective

6. On reposting content

How to get more engagement with a second tweet

We’ve written much before about the case for reposing content, sharing an article more than once on social media. A research team from Cornell investigated this strategy, looking at the effect of wording on sending multiple messages through Twitter.

The researchers developed an algorithm that could successfully predict which variation of the same tweet would receive more retweets. (You can try out the free tool that is based on the algorithm.)

Here are the factors that researchers identified as being helpful for reposted content. (The most significant factors are highlighted in bold.)

  • Ask people to share – Use words like “RT, Retweet, spread, please”
  • Informativeness helps – Focus on length, nouns, and verbs (and not so much @-mentions or hashtags)
  • Make your language align with both community norms and with your prior messages
  • Mimic news headlines
  • Use positive and/or negative words (both seemed to work equally well)
  • Use third-person singular – He, she, it, and one
  • Generality helps – Use indefinite articles like a, an

7. Twitter images for smaller accounts

The 9x increase in retweets just by adding an image

In a huge Twitter analysis by Stone Temple Consulting—over 2 million tweets analyzed for eight different factors, including unique things like domain authority and Followerwonk social authority—the authors discovered a few insightful trends, perhaps none more actionable than the power of tweets with images.

5-images-increase-rts-favorites1

According to Stone Temple’s study, adding an image to your tweet doubles the likelihood that your tweet will receive a retweet or favorite.

And for those with low-level social authority—low follower counts, just getting started on Twitter, or otherwise—adding an image to your tweet generates 5 to 9 times as many retweets and 4 to 12 times as many favorites in total.

28-total-rts-images1

From Eric Enge of Stone Temple:

At lower authority levels including an image will get you 5 to 9 times as many Retweets and 4 to 12 times as many favorites than you will if your tweets don’t include an image. Hopefully, you were sitting down when you read that. Note that high authority levels also benefit as well, though for the 90-99 range the gain is relatively modest. For those high authority accounts, people are already hanging on their every word.

8. The top social networks

The surprising result at #1, plus the unique spot for Twitter

The Global Web Index’s most recent quarterly report (a survey of more than 40,000 Internet users) looked at social media usage and came out with a couple keen insights.

  1. More Internet users visit YouTube than Facebook.
  2. YouTube and Twitter have significantly more visitors than active users. 

Social media active use and visits

So in case you had yet to consider YouTube as a possible channel to meet your audience, there seems to be solid evidence here that your audience is quite familiar and comfortable with hanging out at YouTube. (We’ve got some tips on how to make videos for your brand also, if that’d be interesting for you!)

And as for the drop in active users for YouTube and Twitter, I like to think of this in terms of consumption versus sharing. Someone may be on Twitter to hear the latest news, click some links, see what’s happening—they may still be engaged with your Twitter stream without contributing anything of their own to Twitter.

We covered a series of social media personality types awhile back, and these folks seem to fit well into the lurker category—still a valuable addition to your network, just with their own personal tastes when it comes to being involved.

9. How people spend their time on social

Twitter is for news, Facebook is for friends

Another interesting takeaway from the Global Web Index report is in the survey responses about how people spend their time on social media sites. For Twitter, Facebook, and Google+, the report found the following:

  • The most popular activity on Twitter is reading a news story
  • The most popular activity on Facebook is clicking the “Like” button

Here’s how the rest of the activity breaks down. Note how many of the top Twitter activities deal with reading the news or catching up on what’s been happening whereas many of Facebook’s top activities involve connecting with friends.

User activity on Twitter Facebook Googlt+

 

10. Make waves by responding quickly

5 in 6 messages that need responses are not answered by brands

Sprout Social regularly shares insights from its data, making particular note about the way that brands and businesses listen and respond on social media. Their 2013 benchmark study showed great room for brands to improve, and Sprout’s followup study in 2014 had many of the same takeaways.

There is great opportunity for you to stand out on social media by simply replying to everyone. 

The data from Sprout Social showed that businesses are learning how to reply quicker to responses (we’ve mentioned before that response expectations on Twitter typically hover under 60 minutes). However, they’re replying to a smaller percentage of the volume of messages they receive.

Response study - Sprout Social

  • Response rate: 17% (was 21% one year ago)
  • Response time: 5% improvement from previous year

Over to you

Which of these stats stand out to you? 

Is there anything here that seemed particularly surprising or true from your experience? 

I’d love to hear your thoughts in the comments! Feel free to leave any input you might have, it’d be great to hear from you.

Image sources: Pablo, IconFinder, UnSplashNew social media research

Looking for a better way to share on social media?

Schedule, publish & analyze your posts across the top social networks, all in one place.

Get started for free
Written by Kevan Lee

Director of marketing at Buffer, the social media publishing tool for brands, agencies, and marketers. We’ve got a new podcast! 🍟

  • Margaret

    ———- that’s a full enjoy blog.bufferapp…….. < Start working at home with Google! It’s by-far the best job I’ve had. Last Wednesday I got a brand new BMW since getting a check for $6474 this – 4 weeks past. I began this 8-months ago and immediately was bringing home at least $177 per hour. I work through this link,

    go to tech tab for work detail _________ Continue Reading

  • MorrisJOverby

    ………………..Reset your job with blog.bufferapp Find Here ……………………

  • Elicia

    In point 2, you end with a sentence “To check your own Audience Insights, xx.” Did you mean to end the sentence that way, or were you going to replace the xx’s with something? I fill in with x’s also, btw. 🙂

    • Thanks for the sharp eye there, Elicia! Totally my mistake. It’s all removed now. 🙂

      (And awesome to hear you’re a fellow x’er when you write. Yeah, sometimes I get ahead of myself and forget to check back for x’s!)

  • #4 is an interesting one though we should remember that not all engagements are created equal. Yes Instagram has a higher engagement rate but also look at what is asked from the engagement. People tend to click “Like” on darn near everything on Instagram. Handing out Likes like Halloween candy can lessen how much we weigh them. We should also remember our goal with the content we post. If we’re looking simply for a sign that someone saw it, a Like is fine. If we want to drive another action like a link click or response, it may be better to consider other areas to share that content where these actions are easier to encourage (such as a link click on Facebook or Twitter).

    #10’s fact about 5 out of 6 interactions with brands go unanswered is one I cite often when advising brands on social. So many invest heavily in creating a conversation on social but then when their following responds, they do nothing. Buffer does such an amazing job of responding to everything across all channels. I wish other companies would do the same.

    Here’s one I found interesting recently. A recent study by Twitter showed tweets with @ mentions and hashtags see less clicks.

    “For example, when you’re trying to drive visitors to your website, a tweet that doesn’t include a # or @ mention will generate 23 percent more clicks. When the tweet is focused on driving an app install, forgoing a # or @ mention increases clicks by 11 percent.”

    We should keep this in mind when considering what we’re looking to accomplish with our messaging. If engagement is the key then we should look to utilize hashtags and @ mentions as we know from other studies that they increase overall engagement. But if we’re looking for people to actually click our links, we should consider skipping the mentions and hashtags.

    Thanks for putting together this collection of studies. I do something similar each week on our company blog. It’s a great way to keep on top of new ideas and better ways of running our social efforts.

    • Awesome stuff, Ben! Love that Twitter study you mentioned. I’d noticed similar results in my social sharing, so it’s neat to hear that there’s some research to back it all up. Makes complete sense!

  • Thanks for including our Stone Temple Consulting Twitter engagement study, Kevan. We put a ton of work into that sucker! Readers should note that unlike many other similar studies, we rated not the number of engagements but the likelihood of engagement for various kinds of tweets. Subtle but significant difference.

    • Great point, Mark! Thanks for highlighting that – and for the amazing research you shared with us all!

  • PahlaB

    #4 is everything I love about Instagram as both a personal user and as a brand. Interaction is ridiculously simple: Scroll, double tap. Scroll, double tap. Occasionally comment, and BOOM! Done.
    When I post as a brand, I’m never looking for more interaction from my followers. It’s more of a reminder, “I’m here. I’m still putting out content.” When the Likes roll in, it’s just social proof that people are interested in what I’m putting out there, like a popularity contest. Users notice when you get a lot of likes (and when you don’t). Also, I double tap frequently to show the people I follow that I’m listening to them and interested in what they do.
    I think a lot of brands aren’t active on IG because they can’t get more direct responses (website visits and the like), but I will tell you as a personal user that when a brand Likes my photo I have a HUGE fangirl reaction! Building goodwill like that goes a long way with your audience, even if you can’t measure it directly.

    • Hi there Pahla! Thanks so much for the comment. It’s really wonderful to hear your perspective on things, and I’ve no doubt that your sentiments are shared by a lot of people! So interesting to hear your reaction when a brand likes your photo – that’s great validation for brands to be involved in the pages of their fans, I think! 🙂

  • John Chapman

    I can’t agree with the best times for traffic chart. I find for my 10,000 followers it varies from day to day and from platform to platform. For example on a Monday my peak traffic on Twitter comes at 4:00pm EDT; on a Wednesday it’s at 2:00pm; Thursday has peaks at 11:00am, 2:00pm and 6:00pm; Friday has peaks at 10:00am and 4:00pm. Overall peak traffic is at 4:00pm. It seems to me that peak traffic times are worth investigating but it’s a very individual result.

    • Great point, John! Thanks for sharing your experience here. Have you noticed any trends in when your posts are most shared on social media?

    • kvdmm

      I have noticed the same, be it with a (much) smaller amount of followers. Having read many articles about this phenomenon, I reckon that every company has its own perfect times to post on social media.

      • John Chapman

        Is there a best time at all? If your ‘English’ market is global then while the US is going to sleep, New Zealand, Australia, India, South Africa and the UK wake in turn.

        • kvdmm

          I am convinced that there is or are perfect hours to post for a company. I mostly trade in Europe and Africa, so the lack of time zone differences does help for me.
          If I read your first comment, I see that ‘overall peak traffic is at 4PM’. Well, 4PM might be your best time to post, with a possibility to post on (some of) your other stated peak hours.

    • MyraCWashington

      All time hit the blog.bufferapp Find Here

  • JohnASilcox

    Reset your job with bufferapp Find Here

  • I “Like” #9 how people spend their time on social. A client of mine was recently mentioned in the Huffington Post. We posted two times with a link to the post in each with a screen shot as a picture for one as well. Both of these fell into the “read a news article” category and we had almost double engagement! As always, great work Kevan! Have a terrific Tuesday my friend! 🙂

    • Great stuff here, Steve! Love that you’ve found some of the research here to be true of your social sharing as well. Happy Wednesday! 🙂

  • Big A

    The You Tube stat stands out to me. We actually use You Tube to host internal training videos because people are looking for answers and information there anyway. It is an easy ask in a change management setting.

    • Sounds like a really great strategy! Have you found that you’ve picked up some good views that way?

  • Hey Kevan.

    Epic content as ever.

    I like the final point – reply to everyone. You know what? that works wonders on Facebook.

    I also try to reply with a question. I like to call it the politician approach (they never respond with the response you want but more likely one that will benefit them).

    I like to reply with a further question as this engages the user even more and when you ask someone a question (especially an opinion-based question) they’re likely to respond.

    Instagram is an interesting one, right?

    Yes there’s more interaction but it’s not solid, cold hard conversation, is it?! I mean, social media is about building relationships. I was wondering. Does this really work with Instagram? Interaction is good if it’s something you can build on. But I’m not convinced Instagram is that relevant as a conversation/relationship tool. Thoughts?

    • I love the politician approach. Gonna try this one :D.

      • You have one for free! 😉

    • The politician approach is great, Todd! Love that! I know some folks who are always so great about asking questions in conversations, they seem to always find a way to keep the conversation going. That can be a really wonderful strategy for brands starting out on social. 🙂

      And yeah, great stuff about Instagram engagement. I think you make a really great point, the type of engagement you’ll get on Instagram could feel quite different than other places. From what I’ve understood about Instagram, there are a couple challenges there for marketers in having meaningful conversation and in creating calls-to-action or clicks back to a website. One of the real advantages I find with Instagram is that it might feel a bit more welcoming to new marketers since content does stand a good chance of getting a handful of likes each time. This encouragement and sense of connection can sometimes be quite meaningful for folks I think!

      • Be a politician, Kevan!

        Yeah the early interaction on the insta posts is really good but links and conversation are harder.

        Love Instagram all the same but only use it for me not my business 😉

  • LisaRSanders

    Your first choice bufferapp Find Here

  • kvdmm

    “We’ve got some tips on how to make videos for your brand also, if that’d be interesting for you!” (#8)
    Actually, that would be very interesting for me. I work for some kind of webshop, which was set up about a year ago now. We are currently expanding and updating our social media channels, a YouTube channel has already been set up.
    Mr Lee, if you could send me a hyperlink (or put one in the article, whichever option is easier), that’d be highly appreciated!

    • Thanks for the nudge here! Yes, just added a link to the story above. 🙂

      • kvdmm

        Great! Thanks. I just read the other article, and liked it. 🙂

  • Somaked

    < ✜✱✪✪✲✜.+ bufferapp+ *********Reset your job with modvive < Find More='' ……..''

    1

  • Interesting info and fun to read, thanks. Looking from the back end, as a business tracking and analysis, I see best times as a wash in global commerce. There are better days, Sun and Mon are best for us. My biggest take away from the brand side is how often Pinterest is not mentioned. For us, and others, Pinterest is by far the most active, most leads in traffic to our sites and conversion. Where do you see Pinterest in this mix? Maybe product company and retailers bias?

    • Hi David! Great to hear your take on this. Yes, my sense is that Pinterest is a potentially huge outlet for brands – seems to be somewhat underutilized currently, and my guess is that it’s partly a matter of educating people about how it works best. I think I could improve on the info there, too! Once you get into it, I think people find that it’s quite unique compared to Twitter & Facebook. It’s an opportunity to do some fun new things!

  • CharlesMSmith

    Your first choice blog.bufferapp Find Here

  • swaadd

    a
    شركة سواعد

    شركة كشف تسربات المياه بالدمام

    شركة عزل أسطح بالدمام

    شركة عزل خزانات بالدمام

    كشف تسريب المياة بالدمام

    غسيل خزانات بالدمام

    افضل شركة تسليك مجارى بالدمام


    كشف تسربات المياة بالدمام والاحساء

    افضل شركة نظافة
    منازل بالاحساء

    كشف تسربات المياة بالدمام والاحساء

    كشف تسربات المياة بالاحساء

    افضل شركة كشف تسربات بالخبر

    افضل شركة نظافة بالدمام والخبر

    شركة عزل اسطح مبلطة
    وغير مبلطة بالجبيل

    شركة عزل اسطح بالجبيل

    شركة عزل مائى بالدمام والخبر والقطيف الجبيل

    كشف تسربات المياة بالدمام والخبر والقطيف
    والجبيل

    كشف تسربات المياة بالجبيل

    كشف تسربات المياة بالقطيف

    نظافة شقق بالدمام والقطيف

    شركة غسيل خزانات بالدمام الخبر القطيف

    كشف تسربات المياة بابقيق

    كشف تسربات المياة بابقيق

    كشف تسربات المياة بسيهات وعنك

    كشف تسربات المياة راس تنورة

    شركة مكافحة حشرات
    بالدمام

    شركة مكافحة حشرات براس تنورة

    شركة رش مبيدات بالدمام الخبر القطيف

    شركة مكافحة حشرات بالجبيل

    شركة مكافحة الحشرات بالدمام والخبر

  • Complatict

    < ✜✱✪✪✲✜ +bufferapp+*********….. < Now Go R­e­­a­d M­o­r­e

    12

  • Jason Smith

    Thats really a huge list of info graphics and thank you for the specifying the times for each social media. Lets see how the engagement is if we post in the post afternoon sessions. Video Interview

  • GordonMBland

    Your first choice blog.bufferapp Find Here

  • Wuhaiyan

    Finehope Polyurethane Co., Ltd devoted in promoting widely usage of
    polyurethane since 2002. During the past 12 years, we had worked with many
    prestigious companies. Visit this to learn more: http://www.polyurethanesupplierschina.com/products/Polyurethane-baby-stroller-wheels-stroller-big-wheels-China-Polyurethane-Foam-Suppliers-Polyurethane.html#.VYDHxfR72tg

  • Overy1954

    < ✜✱✪✪✲✜.+bufferapp+ ********* ….. < Now Go R­e­­a­d M­o­r­e

    12

  • The reason behind the More user engagement on Instagram as compared to other social media like Facebook and twitter is that, Every Instagram post is Visual, In from of a Photo and Video and These kind of content engages user and gives them quick idea about what is going on. While facebook has a distributed interface. There is plenty of blank space between posts which asks user to scroll and usually user leaves page instead of scrolling down. While twitter lacks in Expressing the user completely as the 140 character limit actually limits your expressions and idea.

  • Pugboy

    One month I had this page open in Chrome! Finally, it’s read. Why did I refuse to close the tab for so long? Kevan Lee knows his s$%t. Kevan, if you see this, thank you for another great job and taking the time to do it scientifically. Many people don’t realize how time-consuming it can be to understand why the results happened.
    Re, best times to post mentioned below… In my experience, it varies much more with niche markets. This is for a variety of factors involved to reach your audience. Factors like: global? Time Zones? When you WANT to reach YOUR buying/consuming audience, etc. PLUS working with algorithms like EdgeRank on fb to do your bidding to some degree. Regardless, there is definitely an average. for the majority of the bell curve in the middle.
    You can derive when your audience is online by using fb Insights, Pinterest Analytics, Twitter “whatever”, Google Analytics and Buffer has great analytics for social. Regardless, I’ve found that part of the puzzle easier to figure out than some others.
    I would focus some small ads, which are targeted to boost engagement at those times, and see what happens.. You can confirm the results by running ads at other times as a control. I like to start all campaigns this way. It tends to save me time and money in the long run.

  • I don’t think YouTube should be considered a social media. Sure, people comment, but it is not Social where people hang out and talk like the others.

  • Robert Wasson

    Great post! Take a look at this article to find more fascinating social media statistics from credible sources. It’s book marked because it’s really my “go to” for statistic related posts and tweets.-

    http://growingsocialmedia.com/social-media-facts-and-statistics-for-2016/