Facebook's 2016 analytics error is now in litigation
Two years ago, Facebook told marketers that the future was in video. Their analytics showed a big bump in engagement (comments, likes, shares, viewing time and other metrics) from video content. Since then, an entire sub-industry has emerged focused entirely on micro-video content. And then we all found out that those metrics were bunk. Facebook is now in court over allegations that they intentionally misinformed the public in order to hype up the need for richer content. For the full story - here's a great take on it: http://www.niemanlab.org/2018/10/did-facebooks-faulty-data-push-news-publishers-to-make-terrible-decisions-on-video/
Dig deeper. It was external industry analysts that figured out that Facebook had bloated its figures, and this wasn't the first time. Understanding how to look beyond the numbers and see insights based on both your gut and your brain is more critical than ever. It's important to learn about data today, but it's more important to understand that true analytics is both math and investigatory skill. Look for the "how" and they "why", not just the "what".
Virtual Reality can create more compassion than other mediums
Stanford released a case study this week about a program they created called "Becoming Homeless". The VR program allows audiences to live virtually as a homeless person - immersing users into the realities of harsh living under the poverty line. Research was conducted in survey format before and after the experiment, and sympathy / empathy increased significantly for homeless people upon using the program.
The findings of this study aren't surprising, but it's great to see that we're building up confidence in what has been historically viewed as bleeding edge entertainment. Just last week we saw how Netflix and others are turning the broadcast industry into a two-way interactive one, and VR will likely be one of the most important tools in the suite. We can expect to see more ramp-up here with both the entertainment field becoming more interactive, while the tools we use (think TV vs. VR headset) enable that experience in a far more engaging way.
99.7% of Net Neutrality opponents were bots
There's been speculation over the past year of the role that bots had to play in the repeal of Net Neutrality in the US. Stanford released a new study this week showing that of 22 million responses about Net Neutrality, all but 800,000 comments about the issue were real. You'll find a great write up of the issue here: https://www.techdirt.com/articles/20181015/13034340843/997-original-comments-opposed-fcc-repeal-net-neutrality.shtml
We're no longer equipped to understand public discourse as just human beings. Without the aid of AI tools that help us detect things like bots, we're at the mercy of dark players in this space. The biggest problem is that these tools are typically not available off the shelf. It can take weeks, months or sometimes years to properly train these tools to identify these elements, while political decisions like these won't wait for the true data.
Our democractic process isn't obsolete, our analytics are.
Subscribe to Jon Litwack
Get the latest posts delivered right to your inbox