This is the 5th post in a blog series that uncovers how the major social media platforms are personalized to their end users.
Every marketer using Facebook should know that the newsfeed (where every Facebook users spends the majority of their time) is personalized to them. Many marketers are also aware that there is this thing that Facebook uses called “EdgeRank”, which is the name for how they execute that personalization.
However, did you know that it’s gotten even more sophisticated? Edgerank is no more (the developers dropped that name) but the algorithm is still very powerful and has about 100,000 different weighted factors.
Here are some of the user actions that aids in Facebook’s personalization to their users:
- Every piece of data in that user’s profile
- Who that user interacts with and reads the most
- If that user visits a website with a Like button that interaction will influence content on that user’s newsfeed and right rail
- If that user provides feedback around why they hide posts
But it goes beyond this….
In December 2013, Facebook took their personalization work one step further by hiring Yann LeCun, a well-known researcher in deep learning to head the division. (You’ve seen his work if you’ve used an ATM that can read your checks.) Deep learning focuses on training computers to make sense of what seems like random human generated data, vs. machine learning where the data needs to be labeled prior to being processed. LeCun uses a technology called “convolutional neural networks” – it’s really using unsupervised learning models to let computers function like a brain. This is very similar to Google’s work where they taught a computer to recognize a cat.
“The goal really is just to try to understand how everything on Facebook is connected by understanding what the posts that people write mean and the content that’s in the photos and videos that people are sharing,” Zuckerberg explained on a Facebook earnings call covered by Huffingtonpost. “The real value will be if we can understand the meaning of all the content that people are sharing, we can provide much more relevant experiences in everything we do.”
So what is LeCun AI focus at Facebook? Here’s response from a Wired interview he conducted while at the Neural Information Processing Systems Conference in Lake Tahoe:
But our mission goes beyond this. How do we really understand natural language, for example? How do we build models for users, so that the content that is being shown to the user includes things that they are likely to be interested in or that are likely to help them achieve their goals — whatever those goals are — or that are likely to save them time or intrigue them or whatever. That’s really the core of Facebook. It’s currently to the point where a lot of machine learning is already used on the site — where we decide what news to show people and, on the other side of things, which ads to display.
This deep learning work is definitely impacting the newsfeed – reducing the 1,500 updates an individual user could see down to 30-60 that are most relevant to them. Mike Schroepfer, Facebook’s chief technology officer alluded to this work during an interview with MIT Technology Review.
So what does all this mean to you as a marketer?
- First of all, you need to know your target audience very clearly. Dig into all the data you have and craft content that will appeal to those who follow your page. Post when they are online.
- Give Facebook more data about your website visitors. Install the Facebook follow and like buttons you can get from the site.
- Secondly, with organic reach for brand pages nearing 1%, it’s time to directly use Facebook’s data mining and start running Facebook ads to target your page’s followers.
The days of creating content without clear target personas and looking at the data are clearly over. It’s time that everyone creating content online dig into the data.