Deciphering the Like: measuring Facebook engagement

Both individual brands and agencies create Facebook pages in the hope of acquiring as many fans as possible through which, they can secure a solid marketing and branding channel. 

However, the success of their Facebook page is too often measured against Facebook likes alone. 

Measurable engagement amongst fans, as well as friends of fans is what really demonstrates a successful campaign. The question is – How can we report a solid ROI on something like engagement and, more importantly, how can it be measured?

Studies have revealed that in some cases up to 90% of fans never actually return to a fan page; instead the majority of interactions are occurring within the news feed by fans and their friends. These friends of fans may continue to see a post and may engage with a brand in the feed; however, many never actually become a fan.

Facebook’s real marketing power lies in the old adage of word of mouth, in this case- the friends, who come with the implicit recommendation of a trusted source. It is this emphasis which is one of the most successful ways to secure new customers, and this is even more prevalent now with the launch of Facebook Graph Search, letting us delve deeper into user interactions.

Facebook Graph Search

This enhanced search feature provides the ability to search the interactions of friends, as well as brand pages, giving extra insight into user activity.  Rather than an over-emphasis on likes, promotion of user interactions needs to be at the forefront.

Facebook Graph Search has the power to return an image a friend has liked, a comment, or even a check-in at a restaurant regardless of whether the user has actually liked a brand page, even if this activity has been missed by friends in the newsfeed.

Formula for calculating engagement

Calculating engagement rate is much more effective at providing insight on performance, however current formulas are far from perfect. Let’s have a look at a few options.

Facebook’s provided “People talking about this” metric or PTAT is generated from users who have created a story from a post, for example commenting, liking, tagging or sharing. It also includes viral shares, so any comments and likes that have occurred from shares by others.  

These numbers are unique, meaning if a user creates multiple stories in a seven day period this will only count as one interaction. There is a PTAT at a post level and at a page level.

Unfortunately when looking at PTAT, it does not consider negative comments, it only considers hides or spam flags delivering the misconception of a positive response. People are on Facebook to connect with friends and family, not to be bombarded with advertising, so branded messages need to be delicate.

I have seen many posts that receive serious backlash due to poor messaging. However, these posts would have received a high engagement rate due to the sheer amount of negative comments- engaging indeed!

Consideration of negative comments aside, some use this metric in a formula for calculating engagement of PTAT divided by the number of fans, this is simple and easy enough to calculate. The problem with this formula being calculated against fan count is that majority of interactions may not occur from the fans of a page; this formula does not consider the interactions of non fans as well. To show a more accurate result we could replace fans with reach, this way we are including fans and non fans within the equation.

Another concern with this formula is that results can be misleading when compared with pages that post more frequently. These pages could record a much higher overall engagement rate purely based on number of posts. For example, a page with 20 posts with one like each will record a higher engagement rate than a page with one post and 15 likes. We can improve this formula by breaking it down by number of posts rather than the running total.

With this in mind, using PTAT as a metric combines paid and earned interactions which can also skew results if looking purely at organic traffic. We need to be sure we are comparing apples to apples when we are measuring performance of individual posts.

So, what is the suggested formula for a Facebook fan pages engagement rate?

Total number of all interaction including likes, comments, and shares, divided by the number of posts divided by the number of users reached.

Facebook Engagement rate per day =  # of interactions / # of posts / reach

Facebook Engagement rate per post =  # of interactions / reach

This approach takes into consideration the total activity of fans and non fans while removing the skew from high frequency posts that inaccurately impact page engagement levels.

Does higher engagement mean success?

A million people in Silicon Valley walk into a bar. No one buys anything. Bar is declared a rousing success. – Jeffrey Zeldman

Now that we have a more accurate way to track engagement levels we need to keep in mind that engagement rate is effective in tracking activity, not necessarily outcomes or the ability to change or influence behavior, and at the end of the day this is this what brands are ultimately looking for. Maybe we should be asking how we can measure influence.

A Facebook engagement rate combined with other analytics such as lead and conversion tracking could provide us with that extra level of insight to help close the loop. Campaigns need to be structured in a way that they receive a high level of positive engagement that can ultimately lead to influencing consumer behavior. One can achieve a high engagement rate simply by posting jokes or playful images.

Engagement without context or influence is purely just a number of interactions. A high engagement rate is great, one that can also influence consumer behavior is even better.

Daniel Christos is technology and innovations director at TBWA\Digital Arts Network