Compelled to Share

At a tech company where I used to work, I looked into the usage of the share-to-Facebook/Twitter/etc. buttons that were placed on the most visited pages of our site. We didn’t use a persistent share bar, but I still wanted to make sure there was good reason for giving these buttons space at the top of the page.

What I found was that on desktop, the button for sharing to Facebook saw some usage. By “some”, I mean that less than 0.1% of people who viewed a page ended up sharing to Facebook.1 Depending on your perspective, this is either negligible or enough to justify keeping a share button there.

On mobile, out of millions of pageviews in a given time period, shares to Facebook were in the single digits. I don’t mean single digits percentage-wise, I mean that out of millions of people viewing a page, less than 10 individuals used the in-page share button to post to Facebook. And yet that button was fighting for top real estate alongside many other critical pieces of the page.

When it came to other types of sharing, I found that Facebook was the only service people bothered with. Sharing to all other services via the in-page buttons happened at a rate of essentially 0% on all devices,2 making those buttons pointless.

I would never pretend that this one case is representative of the web as a whole. Every site obviously has its own unique set of user behaviors and audience segments that are considered valuable. What I will say is this: I hope every site that has in-page share buttons has analyzed their usage stats to make sure the benefits outweigh the costs in screen real estate, brand advertising, and analytics data given to social media companies.

When it comes to mobile in particular, the site I worked on showed that in-page share buttons were irrelevant. On this I have a stronger opinion as iOS and Android have strong platform-wide design patterns for sharing, whether through share sheets, copy/paste, drag and drop, voice control, or whatever else may come in the future. The data I looked at suggested that these platform controls are what people use, not the social share buttons that junk up webpages the world over.

  1. We used a strict definition of sharing, where we counted a share only if the user successfully posted to their timeline. We didn’t count if the user clicked the initial share button without following through. ↩︎
  2. Even extending out to a generous number of significant digits. ↩︎