Social networks are always making updates and tweaks to their platforms.

And in a recent update, Twitter removed share counts from their Tweet buttons, affecting many bloggers and publishers.

As some data has started to emerge on how the has change affected content sharing on Twitter, we wanted to dive below the surface of this update and see how this change could affect marketers.

Let’s get started.

pablo (1)

Meet Twitter’s new share buttons

Like other recent Twitter changes, this one appears mostly cosmetic on the surface. Share buttons still function exactly the same. They just have a new lick of paint as you can see below:

twitter-share

The key difference is that share counts have been removed from the new buttons as Twitter announced in a blog post on October 6th, and from that date share counts started to disappear from the platform’s social sharing buttons.

Below you can see the updated share button on the Buffer blog:
blog-share

At a first glance this may not sound like an earthshaking development. However, for many the share count data was a useful guide point on post performance and also a way to promote content with readers by showcasing its popularity.

Do share counts influence Twitter sharing activity?

We wanted to dive in and find out if removing the share count has had any effect on the number of people sharing to Twitter?

This is where the new update gets really interesting. Shareaholic pulled together and analyzed sharing activity data generated across the 300,000+sites that use their product.

Here are the results:
shareaholic

The data shows that sharing activity to Twitter has declined by 11.28% since November 20th.

The decline in sharing is in the ‘share of voice,’ the percentage of Twitter shares, relative to other sharing destinations available such as Facebook, Pinterest, LinkedIn, etc.

Are people still posting to Twitter through Buffer?

After seeing this data we decided to hop into our own Buffer data and check the number of post being shared to Twitter through Buffer before and after November 20th and as you can see below, there’s been no significant change:

buffer-twitter

What does this mean for publishers and site owners?

Share buttons have been synonymous with blogs and websites since social media started to break into the mainstream, and since the beginning the share count has usually been at the core of each button (I remember when I first started blogging I’d often turn to the share count as my first validation of a post succeeding).

Out of everyone, the removal of share counts from Twitter’s social buttons will likely have the most effect on publishers and bloggers.

One of the big advantages to having the Twitter share count was the social validation that other people have also enjoyed and shared the post. This is something that Slate’s vice chairman Dan Check explains more on over at Digiday:

“It’s meaningful to us. Having share counts along comment counts is a strong way to underline that there’s a conversation around what we’re written. We want to signal to readers that that conversation is happening.”

How to show social validation without share counts

Studies show that we unconsciously look to others to guide us. When we can see a post has been shared hundreds or thousands of times before, this social validation makes us more likely to share.

And with these updated Twitter buttons, some of that social validation has gone missing.

Thankfully there are a few ways you can show social validation and conversation around a post without relying on share counts.

Embed a Tweet sharing your post

This is something we’ve done at Buffer before and I could see it becoming a more common tactic now that Twitter share counts have been removed. Here’s an example:

The embedded Tweet will show readers the number of Retweets and Likes the Tweet has received and go some way to replacing the social validation lost with the updated buttons. 

Embed a search

If you head over to Twitter Search and run a quick search for the URL of your post, you’ll be able to collate all the Tweets where someone has shared your post. To embed this search within your blog post, simply click ‘More Options’ from the results page and then select ‘Embed Search’ from the dropdown menu.   Screen Shot 2016-01-05 at 11.47.07 You’ll then be taken to a page where you can configure and edit your search: Embed search Once you’ve configured your search, click ‘Create widget’ and you’ll be given some code to copy and past into your blog post. Here’s how an embedded search looks:

Experiment with different buttons and locations

While Shareaholic data suggests that Twitter sharing numbers have dropped since the share counts have been removed, not everyone is finding that the case, as New York Magazine’s Digital GM, Michael Silberman, explains:

“We have some evidence that the premise — share counts encourage more sharing — isn’t even true. And of course on mobile, for our site and many others, you can’t even see the share counts unless you tap to share. (And most sharing is happening in the address bar on desktop or iOS share sheet anyway.)”

On their blog, Crew editor, Jory MacKay, shared some of their findings about share buttons after some experimentation.

Crew found that readers are more likely to share using the buttons at the bottom of a post, compared to the ones at the top. MacKay believes there are a couple of reasons for this trend:

  • The share buttons are more obvious and recognizable at the bottom.
  • People are more likely to share a post that they’ve read through.

In his post, MacKay also refers to the importance of prototypical elements:

In the most basic terms, this just means we like what we know. When we expect to see something on a website in a specific place with a specific style (like sharing buttons at the top or down the lefthand side of a post) we feel good when they’re there and disconnected and unsure when they aren’t.

These findings suggest that while share counts can be useful, you can also boost your sharing numbers by experimenting with different styles of buttons and layouts.

Over to you

The loss of share counts doesn’t feel like a cause to panic about the future on Twitter sharing for publishers.

Even with a potential drop in sharing, we may soon see things pan out and sharing return to normal levels. And with more reason to experiment with new layouts and ways to encourage sharing, it could even lead to new findings and increased sharing.

Have you noticed any difference in the amount of shares your posts have received since Twitter removed the share count? Have you tried any new ways to encourage sharing? Let us know in the comments.

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 Ash Read

Content crafter at Buffer. I’m fascinated by storytelling, entrepreneurship, and travel. When I’m not writing, you’ll usually find me on a football pitch or basketball court.

  • Is it just me, or does that data look like there is a gradual decline in Twitter shares over time anyway? The month before even suggests a slight spike, followed by a return to gradual decline.

    Do you think this might be more an indication that Twitter is becoming less popular for sharing? Have you noticed any significant change in sharing of the Buffer blog to Twitter?

    • That was also the first thought in my mind when I saw the graph.

  • I still don’t understand why Twitter took away the shares. Twitter is basically slowly killing itself.

  • Any guesses on what Twitter’s revenue increase is with their demand that you now pay to see shares? Didn’t they start charging, or did they just partner with another company to do that?

    Either way…any odds on when they might backtrack? I hope they do.

    • Twitter isn’t charging to get share counts.

      You can use tools like BuzzSumo to measure these shares but they can only measure public shares and only count those that use the EXACT same URL which means they miss plenty and also miss the 80% of sharing that comes from dark sharing.

      • So what do you think was the rationale then for Twitter to abandon the counters?

        I was under the impression that Twitter is forcing us to go through their own company, Gnip, to buy services that show
        the counter.

        • Twitter explained their reasoning in a blog post after the initial announcement. Largely it comes down to what they have the availability to support and where they put their resources.

          https://blog.twitter.com/2015/hard-decisions-for-a-sustainable-platform

          • I just don’t by that reasoning, myself. I’d be much more inclined to look at the company’s financials and find some reason there.

            Are they losing revenue? Has revenue stalled? Have new users stalled?

            Furthermore, Facebook and Google+ aren’t having this problem.

            Why is Twitter having difficulty? Gosh, might be something else for investors to take note of.

            All in all, I think there’s more here than what the Twitter blog is saying from October, with an update in November.

          • Twitter is struggling to get new users. That’s why their stock hasn’t done so well and why their old CEO is out. Revenue is also not doing as well as projected.

            How do you believe that this move is tied to their financials other than as they say, allocating people to something like the share counter which was never meant for public use and not something they want to spend the money supporting anymore?

            I guess I believe many when they say they don’t have the resources or need to concentrate elsewhere. Buffer made the choice not long ago to close their suggested posts function. I don’t believe they did it because because they want to sell me on something else or ill-intentions. I think they just had to focus on other things that were more important.

          • I’m going with the argument that Twitter has a stake in Gnip and now makes money from what you used to get for free.

          • I doubt enough businesses are paying Gnip for that information. It’s not that valuable of information in the first place. What does it matter if I have 1 or 1000000000 shares if my page isn’t doing what it’s intended?

            You can also get the share information from countless other sources. BuzzSumo, ShareAll, Social Analytics, and many more offer that info. If Twitter was doing it just to make money from Gnip they’d have been much better picking something that no one else has access to, like impression data.

          • Besides, you can’t pay Gnip if you want to. Gnip has only been opened to a very limited group of high level partners. My company was willing to pay to get Gnip access, but all we can do is get on a signup list (which I understand no one ever hears from).

          • That’s really interesting about it being a “fluffy metric”, Ben. Many feel shares help us understand the performance of a post, though I’m torn in that respect. I’m often curious on the legitimacy of shares.

            Would you be up for sharing what metrics you might look at instead of share data, to determine a posts’ social performance?

      • Actually you are wrong Ben. Twitter IS charging for share counts and the price is very high. We have been in touch with many developers who are telling me those prices

        • Artur Brugeman

          As I mentioned above, you can get share counts for free with third-party services https://blog.bufferapp.com/twitter-sharing#comment-2442082805

          • Yes, but be aware that all these are estimates of the share count, not the actual share count.

          • Artur Brugeman

            Hi, Mark. Could you please elaborate on why these are just estimates?

            I’ve been selectively comparing what these services track, and what BuzzSumo browser extension shows (they have access to Gnip data).

            Results match very closely – many times even exactly. That’s a pretty good result given that counting shares is a complicated thing with shortened or canonical urls etc.

          • I was mistaken! I thought I’d seen Buzzsumo say that they did NOT have Gnip access, but their blog post http://buzzsumo.com/blog/twitter-share-counts-in-your-browser-new-buzzsumo-chrome-extension/ says that they do.

            However, in that same post they note that because of the huge cost, they are only collecting that data real time for the first three days of a tweet’s life. After that, they will only do periodic updates, so between updates the tweet count may be slightly inaccurate.

          • Artur Brugeman

            Correct, that’s an implementation detail of BuzzSumo. NewShareCounts, however, counts all shares in real time no matter the age of url.

            All share counting services are essentially an estimate (even Twitter’s own). They always lag in updates, they regularly miss certain shares (like if target url was temporarily unreachable), etc. Some approaches are a very bad estimate – like counting clicks on ‘tweet’ button. The only thing we’re left with is to choose the ones that offer the best accuracy (even if not perfect).

        • Is it possible to get share counts through Gnip? Yes. Are they pushing people to do so? No. It’s certainly not the purpose of Gnip nor is it something most are going to utilize it for if they do get access. Those that use such access generally have far more important uses for it when paying so much.

          As stated elsewhere in the comments, getting access is almost impossible and only available to the biggest of clients (think Fortune 500) and even then is very difficult. If Twitter were looking to eliminate share count simply to make money they would have made it FAR easier for brands to access it for a price.

          • Ben is correct, and we know this from first hand experience. We were willing to pay the price to get the data, but have been ignored and rebuffed at every turn. So this isn’t simply about selling the data at a certain price.

      • Artur Brugeman

        BuzzSumo does indeed have some limitations, but not the ones you mention. Specifically, they count not just ‘EXACT same URL’, but they also handle canonical urls properly, and shortened urls as well – which means they don’t ‘miss plenty’.

        Real BuzzSumo limitations:
        1. They do count only public shares – but are you sure Twitter counted private ones?
        2. They only show share counts for urls they consider ‘articles’, and don’t for any other urls.
        3. They get slow at updating share counts for older posts.

        Could you please explain what you mean by ‘dark sharing’?

        • Dark sharing is all the sharing that takes place not on social. Stuff like sending links via email, text, on forums, through Facebook messenger, etc. Studies show it makes up as much as 80% of all sharing on the internet.

          • Artur Brugeman

            OK, dark sharing might indeed make up to 80% of all ‘shares’. However, does dark sharing happen on Twitter? We could call private messages or tweets as dark sharing, but I doubt they would make anything close to 80%. I’m not sure if there was any research on this, would be very grateful for any links. My intuition tells me that most Twitter accounts are public and there are much less Twitter direct messages then public tweets, so dark sharing on Twitter is negligible. Hence, tools like BuzzSumo are a good way to properly measure Twitter shares.

  • Just checked Buzzsumo, and the removal of the share number seems to have had no effect on our Twitter shares. Several of our most-shared posts this year on Twitter have occurred since the removal.

  • Jigme Datse Yli-Rasku

    Based on the graph you present of “before” the removal, and “after” the removal, while there is a “sudden drop” which you could erroneously point to being due to the change, what is more important the “trend” really is not obviously (in fact it seems to be not at all) affected. I believe your conclusion as to “source” is very mistaken. Especially since you are attributing a cause to something you can *simply* associate correlation to. May I remind you correlation != causation.

    • I too was confused by Ash’s headline of “it appears so,” especially since deeper in the article he himself concludes that there isn’t consistent evidence beyond the one Shareaholic study that shows any drop off at all.

  • I don’t understand how to embed the Tweet. I use Click to Tweet, but this seems to be something related to Buffer. If I don’t have a Buffer account, can I embed the Tweet to get the counts under it like the screenshot shows? Buffer is not free, so I use Hootsuite.
    Jance

  • Artur Brugeman

    Great post, Ash!

    Just wanted to ask why you did not mention any third-party services that provide Twitter share counts? Like newsharecounts.com or opensharecount.com? They allow you to display share counts on your website for new articles, and in case of NewShareCounts – even show counts for old articles as well. I’d say these should be on this list of ‘what you can do about the lost Twitter social proof’.

  • Hi Ash,

    Thanks for this post though. I wish someone of the sharing tool owners buys the contract from Twitter to display share counts.

    I loved your tips on embedding but don’t you think the search embeds can make the site look dismal?

    I think there are some workarounds to display share counts on Twitter buttons.. here are some of those tools http://newsharecounts.com/, http://twitcount.com/ etc.
    P.S. I do not endorse or business with these tools, just added in for information 🙂

    -Thank you 🙂

  • Interesting to see Twitter make this change that seems counter-intuitive to the growth of a social network. I followed you on Twitter today Ash and voted for Andre Drummond too! Cheers and sharing this post with my network today. 🙂

  • Kate Warnock

    I may have missed, but Ash I would think that bloggers would want to increase their use of services like Click to Tweet or other embeddable sharing code tools. It’s certainly not a counter, but it makes sharing what the user is reading far less frictionless (especially if the blogger is smart about using their best quotes/stats throughout the post – user can tweet what resonates most). If the goal is to get the content read AND shared, perhaps those tools will help offset what’s lost with the social validation being removed?

  • Joe Lean

    Any clues as to *why* Twitter have done it?

  • What is the link to Twitter search?

  • Whenever I read posts I look on the social media share counts to judge the popularity of the article. It’s a fact that people follow mass. If an article got lots of social shares then it is likely for new visitors to share them too. I don’t know why twitter removed the share count…

  • Autumn

    You will have it if it belongs to you http://www.19216811254.com