How to Reverse Engineer LinkedIn for Predictable Virality

By October 2, 2017Growth Hacking, LinkedIn

Over the last several months, we’ve seen the rise of viral LinkedIn posts.

Millions of views, comments, and website visitors.

Are you getting your slice?

To get ours, we took a scientific approach.

We immersed ourselves in the data.

Rather than relying on hunches, we wanted to know at what point you can consider a post viral.

How many Likes or Comments does it take?

And in what period?

With enough data, we discovered the magic number – here’s how:

Step 1: Discover a Correlation

We tested whether comments correlate to views. We missed the mark.

Then we tested whether Likes correlate to views. We hit the bullseye.

At BAMF Media, we already A/B test our own LinkedIn posts. We know when we’ve hit virality to the point where if posts are not getting strong enough engagement, then we take them down.

If it’s within the first hour, we still have time to adjust.

Rewrite the first line. Draw in the emotion sooner. Emphasize the pain harder.

This process is raw. It’s intuition combined with our best practices and relentless obsession for excellent copy.

It’s not a science, but it works.

Here are our results so far.

The number 142.

That’s how many views a post will get for every Like – on average – with a 90.04% confidence.

You can see how strong the correlation is in the line graph below:

LinkedIn posts have a limited window. Most only go viral for three days.

72 hours in three days. 704 likes / 72 hours = 9.78 Likes/Hour.

Let’s assume that people are asleep 8 hours a day. That leaves us with 16 hours.

16 x 3 days = 48 hours.

704 likes / 48 hours = 14.678

If you are aiming to get 100,000 post views, make sure that you get 15 Likes in the first hour -minimum.

Without taking any other velocity engagement metrics into account (which we don’t have the data for yet), you need, at least, 15 likes in the first hour of a post if you want to get 100,000 views.

How about 1 million views?

Assuming people don’t sleep (LinkedIn is a global network):

1,000,000 / 142 = 7,042 Likes / 72 Hours = 97.80 Likes within First Hour.

Taking into account the views per like rate increase as posts gain velocity and LinkedIn’s algorithm picks it up, then this number is around 50 – 60 Likes.

If you want to get 1 million views, keep A/B testing your post until you get ~ 50 – 60 likes within the first hour.

Our views per like ratio is 90% accurate when analyzing posts after they’re live for three days. Finding how to weigh likes and engagement more accurately in the early hours is our goal moving forward. 

2. Create a Predictable Model

If you’re interested in running a linear regression model to find slope and correlation between engagements and post views, then there are many tutorials readily available via Google.

Here’s an inside look at our spreadsheet where the magic happened. (Note: The data set has been reduced for simplicity.)

INITIAL GOAL: Create an accurate estimate of “Post Views” even though it’s not public.

NEXT GOAL: Developing a chrome extension that provides accurate predictions of whether a post will go viral within the first hour based on past engagement stats. To do this, we need to account for velocity – the rate at which LinkedIn increases your reach when your content is “fire.”

Profile Name Date Comments Likes Views Estimated Views Views/Likes Views/Cmts Est. Views Cmts.
Josh 3w 1,073 32,342 5,286,170 4,592,564 163 4,927 3,156,766
Josh 1w 327 5,339 1,110,415 758,138 208 3,396 962,034
Houston 2w 111 3,903 361,784 554,226 93 3,259 326,562
Houston 1w 71 1,622 230,715 230,324 142 3,250 208,882
Josh 22h 93 938 95,476 133,196 102 1,027 273,606
Houston 2w 16 298 55,496 42,316 186 3,469 47,072
Houston 2w 8 104 10,131 14,768 97 1,266 23,536

Key Stats:

Avg Views/Likes Avg Views/Cmts Confidence (Likes) Confidence (Cmts)
142 2,942 90.04% 67.02%

 

How to BAMF Your Posts

Even if the magic number of Likes in the first hour is between 50 – 60, it doesn’t mean your post will go viral. You need excellent copy to attract engagement.

Ask yourself:

What’s your hook?

What’s the learning lesson?

If you can transfer enough emotion to the reader, then you’ll get engagement.

If all you have is a magic number – that won’t move the needle.

So before you think you have it all figured out, buckle down, and write excellent copy with an aim to hit the viral benchmark.

If you’re looking for more LinkedIn growth hacks, stay tuned and get notified of our Product Hunt launch by subscribing to our weekly updates here.

houston

Author houston

More posts by houston

Leave a Reply

Share This