I’m only half kidding.

According to Mark Twain (or, possibly, the British Prime Minister, Benjamin Disraeli) there are three types of lies:

1. Lies
2. Damned lies, and
3. Statistics

I named this blog post “Should the government regulate statistics?” because I was worried I’d have too many posts entitled Lies, Damned Lies, and Statistics.

In this case, I’m referring to a a 2013 Nielsen study titled,”The Follow-Back: Understanding the Two-Way Causal Influence Between Twitter Activity and TV Viewership.”  I found this study in a story posted on NetNewsCheck by Michael Depp entitled “Ten Local Digital Media Trends for 2015.”

Michael predicts that data will show a positive correlation between social media use and Nielsen ratings.  Specifically:

Want to talk to an audience? You find them on social. Those broadcasters who use these platforms well will invariably find themselves evolving in tune with what their audiences really want and care about. Their programming will be more relevant, their engagement much stronger and their proposition to advertisers hugely more valuable.

I presume Michael is basing his prediction on the data summarized in the Nielsen report which is best summarized:

In twenty-nine percent (29%) of the 221 episodes analyzed tweets influenced TV ratings.

That sounds awesome, doesn’t it!? Unfortunately, this is misleading.

Given the scale of the Twitter and Facebook audiences it’s not surprising in the least that there were palpable bumps in viewership.  Notably missing from the Nielsen study was the magnitude of the effect.  If a host/show producer engages audience on social media what ROI is there?  What is the measured engagement and what did that engagement do to ratings?

Nielsen couldn’t have published their report without computing the engagement so why was it missing from the published results?

Because unboosted, unpaid social media engagement is very low.  You don’t have to look far to find stories about engagement rates below 1% on Twitter.

This past summer, Danny Sullivan wrote a post entitled, “Just Like Facebook, Twitter’s New Impression Stats Suggest Few Followers See What’s Tweeted” in which he describes the behavior of his 390,000 (yes, 390k) followers.

Here’s the money quote from Sullivan’s post:

That 5% engagement rate sounds pretty good, but it’s based only on the 7,195 people who actually saw my tweet. What’s the engagement rate for my overall audience of 390,000? That’s 0.1%, rounded up from 0.0923%.

… and that’s for Twitter.  How about Facebook?  Socialbakers’ analysis indicates that the automotive industry has the strongest Facebook engagement at, yep, waaaaait for it.

0.96%

You read that right.  0.96%.  Why?  I can’t say it any better than the indie pop band, Jukebox the Ghost in an October 2014 Facebook posting.

“Sadly, the use of Facebook for effectively keeping our fans in the loop is no longer a total reality. Out of our 50,000 fans on here, only an average of 6,000 or less see what we announce due to an always-changing algorithm and an organic reach that is hard to predict. Many of you might not even be aware that we’ve fallen off your Facebook feed — We’ve gotten the sense that many of you are not aware how much Facebook has changed in the last couple of years.”

People and brands have put a ton of time, effort and money into building THEIR social network audiences – but now Facebook and Twitter are changing the rules, making all that effort for nought – unless you are willing to pay for it.

Social media’s impact on audience engagement is taking a HUGE step backwards.