Screen Shot 2013-04-02 at 9.58.29 AMOne of the first things I’ve learnt, close to 2 years ago when taking the plunge into Social Media with Buffer, was that things aren’t yet very defined.

There guiding metrics and studies are really just in their beginning phase and a lot is still quite vague. To help make things a tiny bit more clear, I thought it might be helpful to collect 10 of the most interesting social media studies and see what we can best learn from them.

So without any further ado, let’s dig in and talk about the most important social media stats out there:

1.) Over 120 Billion Facebook Impressions later – what we learnt

In a very interesting study BlitzLocal looked at close 120 billion Facebook impressions and tried to make sense of it all.  Some of the most interesting facts came around lengths of posts. The researchers concluded:

“Longer posts tend to perform poorly. The ideal interaction being driven by posts is between 100 to 119 characters. Questions tend to drive interaction up by 10 to 20 percent.”


Key takeaway: Whilst most of us know to keep postings short, getting actual data behind it is useful. Another key insight from the study was to prioritize picture postings over links. Find hack number 2 here to do this more easily.


2.) Almost half of online customers expect brands to provide customer service on Facebook, but only 23% provide it

Here is something we’ve struggled a lot with ourselves at Buffer in the past: To provide great customer service on Facebook. I’m not sure why, but it’s always proven to be trickier than Twitter or email.

And yet, seeing that the bar for most people out there is extremely high definitely serves as a good motivator. Interestingly, much less people (only 17%) expect great customer service from Twitter.


Key takeaway: What this takes me back to is Gary Vaynerchuck’s quote:

”I genuinely believe that any business can create a competitive advantage through giving outstanding customer care.” ~ Gary Vaynerchuk

Meaning that, because so few people provide great customer service through Social Media, there is a huge opportunity for you to stand out and really wow the customer. At least for the next few years.


3.) The science of social timing: When, how often and where should you post?

This is something often and wildly discussed in the social media world. And the results don’t always add up from all the different research studies. Here are some of the most interesting results that I found:

1.) Dan Zarrella and KISSmetrics (study):

  • Best time to post on Twitter (for retweets): 5pm
  • Best time to post on FB (for shares): 1pm
  • Optimal amount to Tweet: 1-4 Tweets per hour
  • Optimal amount to post to FB page: 0.5 times per day

Screen Shot 2013-04-02 at 9.43.08 AM

2.) and Raka (study):

  • Best time to post on Twitter (for clicks):  1-3pm
  • Best time to post on FB (for clicks): 3pm

Screen Shot 2013-04-02 at 9.43.28 AM

3.) Argyle Social (study):

  • Best time to post to Twitter: 9am – 7pm
  • Best time to post to Facebook: 9am – 7pm
  • Best day of the week to Tweet: Any weekday gives 14% more engagement than weekend
  • Best day of the week to post to Facebook: On weekends you get 32% more engagement

Screen Shot 2013-04-02 at 9.42.44 AM

Key takeaway: With some of the smartest minds in social media having taken on the science of timing, one thing seems clear: No one arrives at one optimal solution. The best way we’ve found things to work is to experiment lots and see what works best for you. Making sure you get some actionable social media analytics seems more important than a general optimal timing.


4.) Click-Through Rates May Matter Even Less Than We Thought

Now here is an interesting one, that isn’t strictly speaking part of the other social media stats, but an overall online marketing study. The key insight from the study is this:

“The correlation between a click and a conversion is virtually nonexistent.”

The correlation between someone clicking and actually filling out a form and doing something was 0.1. What did however create a high conversion was to measure hover with the cursor. That resulted in a 0.49 correlation of actually converting and completing the call to action. On top of this, pure impressions also created a 0.35 correlation to conversion.

Key takeaway: Keith Pieper, CEO of Pretarget who created the study took this one for himself:

“My key takeaway is that optimizing to viewable impressions or hover time is a better proxy for a brand advertiser than a click-through rate.”


5.) 70% of Consumers Trust Brand Recommendations From Friends

This is something I feel most of us have known and every so often, research pops up to confirm this again. And yet, I feel it couldn’t be any more important to emphasize this, especially in the age of social media.

To have it all in overview, here is how it ranks from top to bottom:

Screen Shot 2013-04-02 at 8.54.48 AM

Personally, what I found most surprising from the study was this:

“Europeans trust all advertising communications less than Americans”

Key takeaway: Shifiting from the traditional “push” advertising model to a “pull” model that uses inbound marketing and friend recommendations as the core way to market is still very hard to grasp for most companies beyond the startup world. And yet, as HubSpot’s state of inbound marketing reveals, this is more important than ever. Go create awesome content for people to share and recommend!


6.) Social Networks Account for 20% of Time Spent Online ( more Facebook statistics)

I loved reading this study, simply because if you are in the social media trenches every day, it’s much harder to feel the growth and the movement that’s actually happening.

To put ComScore’s results even more impressively about social media usage:

“Social networking sites now reach 82 percent of the world’s online population, representing 1.2 billion users around the world”

Key takeaway: What the study didn’t highlight as much, was that “other content” still makes up around 30% of all online traffic. Other content contains to a large amount traffic from blogs and websites. Replacing your Facebook presence as your home base is something to avoid at all costs and a focus on a better funnel is more important than ever.


7.) Shelf Life of Social Media Links Only 3 Hours (bitly social media stats)

One of the most interesting studies from the recent past comes from analytics and shortening service Bitly. Any link you share on Facebook and Twitter will be mostly irrelevant after 3 hours. Especially our recent findings on Facebook statistics seems to also confirm this.

The most interaction will actually happen within the first 60 minutes, as Bitly shows in their graph:


Key takeaway:  The one key action point to take from this research I found is frequency of posting. Instead of fussing over the best times to post, finding high quality content that you can share consistently throughout the day is more important than ever


Quick last fact: “Women spend 30% more time on social networking sites than men”

A last most interesting insight from ComScore came about the time spent on social networking sites per gender. Their end result was that women are 30% more active, across not just time spent, but posting, clicking and overall engagement with content online.

Screen Shot 2013-04-02 at 9.28.14 AM

I don’t dare to interpret this, it’s insanely interesting to say the least.

For more exciting social media statistics check out this post on the latest social media stats.

Writing this article showed me once again how fast social media stats are changing everything around us. Have you found any interesting data insights yourself? I’d love your insights on this.

Looking for a better way to share on social media?

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

Start a 14-Day Free Trial
Written by Leo Widrich

Co-founder at Buffer.

  • I am so much confused,for bloggers which is the best time to tweet or share on facebook

    • LeoWid

      Hey Andrew, that’s a good one, it’s definitely very interesting that all the different studies come to different conclusions. I think the key takeaway is to simply test it out yourself and see what the best time for your followers and facebook friends and fans is! 🙂

  • “Optimal amount to Tweet: 1-4 Tweets per hour” – Isn’t it too much? I would certainly remove such user from my following list 🙂 I think it is annoying amount of tweets, isn’t it?

    • LeoWid

      Hi Peter, that’s a great point, I think it can easily get overwhelming. For the @bufferapp Twitter account for example, we found that 10 updates a day is a good amount. 🙂

  • Thanks for sharing this! I always enjoy when B2B receives a mention in regard to social media.

    • LeoWid

      Hi Colleen, thanks for the kind shout and no worries at all, so glad you liked this one! 🙂

      • Raju teja

        This is the most helpful idea share with us about in Facebook and Twitter. Thanks!

  • Perfect timing! I was searching this information this morning. For me the best time to share on Facebook is around 11am.

    • LeoWid

      So glad we could help with that! 🙂

  • Thank you for this very insightful article. Interesting information here. I plan on getting this app due to the article and the hacks presented.

    • LeoWid

      Yay, really glad the info was helpful, and awesome you’ll give Buffer a try! 🙂

  • I’m absolutely LOVING this article (as I do with ALL of your studies and insight). Anyhow, my personal “KEY” takeaway is most definitely: “Instead of fussing over the best times to post, finding high quality content that you can share consistently throughout the day is more important than ever.” —- This is the essence of why I LOVE Buffer. I can simply and EASILY schedule quality content as I come across it… and then aggregate to SET key times throughout the day. Amazing, practical, and GENIUS. Y’all rock.

    • LeoWid

      Amazing, it’s so great to have you stop by here Kacie, thanks for all the love and the kind support. Your encouragement is exactly the reason, why writing more articles isn’t hard! 🙂

  • Killer article!

  • Really great post, Leo. Thanks for compiling all the studies. Definitely bookmarked this one. I’m particularly surprised by number seven. has been looking at that data for a few years now and it seems that there’s a pretty steady downward trend in shelf-life of links shared on social. Do you think this decline is related to people simply sharing more, more people sharing or both?

  • Learnt is a word ?

    • trolley

      Haven’t your learnt that already?

  • Interesting post but are those times GMT, Eastern, Pacific??? I think that is a rather key thing to know.

    • That depends on your users, the time should be the local time of your targetted users’ location I would say.

  • Super insightful. Definitely going to take these things into account while working through social media sites.

  • Fantastic read! I just discovered the Buffer blog and am greatly enjoying it! Thanks for sharing this Buffer team!

  • great info, thanks

  • michaelgocia

    That’s really great and very true stuff . You have studied very practically and presented in an impressive way, I will surely try those for my social media profiles . Thanks for sharing

  • traderkos

    Very informative although I have to believe that Twitter (times) depends greatly on ones target audience.

  • wow, thanks. This is really helpful

  • This is awesome – an aggregation of some of my fave research, thanks!

  • Nice compilation ! Up to now, my community management has been led by the previously mentionned kissmetric study, as well as this one that could also help to engage more readers:

  • Very cool numbers, and a few I haven’t seen before. Just a caution about customer service and social. The research is misleading because they aren’t asking quite the right question. In fact all things being equal (in terms of wait time, etc), customers PREFER email and phone. Some data from one of our polls at (poll is still open if you want to vote.

  • I appreciate these stats using the details will improve the engagement on my Facebook business page and my Twitter account.

  • I also think its important to monitor your own stats to see what works best for you. Although I find the timing studies very useful, I find they are best used as a starting point. Ultimately one needs to make adjustments to suit their own business and audience.

  • Nice information you have provide in your blog especially the statistical data. Great work.

  • Great post, love the analysis about best times to post and shelf-life as well as good visualizations here w/ the exception of the 3D bar chart. Overall not bad!

  • Great article, but studies also show that “Primetime” Eastern Standard Time Facebook posting also boosts engagement, specifically the sharing rate. What are your thoughts on that?

  • tilan


  • Joel Rogers

    That is really neat data I also so some interesting social networking statistics at SocialDataExcavation.Com They have designed a service that is great for harvesting Facebook, Twitter and google plus data for a really low rate I pulled some data for under $10 bucks.

    Take a look SocialDataExcavation.Com

  • Cristian Pozo

    Women are social creatures! They love to talk! It’s logical

  • janeleonard

    Great post, Interesting point about the life of links. This was tweeted today and I wonder if the long tail can also apply to posts.

  • Merryn Padgett

    Great post. I conducted an experiment recently and found that the type of content I’m sharing also contributes to engagement (clicks, shares etc). Thanks for sharing.

  • Rob Newman

    I have noticed the men tend to “Like” more on FB than women, and women tend to share and comment more on FB than men.

  • tracysestili

    That stat is from 2011.

  • Ilya

    I would love to see more distributions rather than averages…

  • Awesome writing style!

  • I think these percentages varies from country/culture to another! Everyday i read different metrics and percentages and i don’t what to believe anymore? As for number 7, i think the link Shelf life is for ever if you have a well/strong structured blog/website 😉

    Anyway, Nice post 🙂

  • From my recent study and past 3 years of experience in twitter marketing i found only one thing. That if you are not able to get retweets in two digit in he first 6 months then you need to change your strategy. I started using this and found them very useful.