top of page
Ian Barnard

The worst social media metric (according to Facebook)

A few days ago I had a mini work crisis.


A digital agency was preparing to submit a social media strategy and dropped the words every marketer should hate hearing - "we need to focus on engagement."


Now, I know in my heart of hearts that engagement is a terrible metric.


It's vague (are we talking likes, clicks, shares, comments, or a combination of them all?), difficult to action (how exactly do you make something more engaging?) and has been proven many times over to have zero correlation with raising brand awareness, changing consumer perceptions, or selling products.


But for some reason it sticks around, like a bad smell refusing to die.


I needed something authoritative to kill the engagement angle fast, but didn't know what to use. Panicked, I turned to Twitter for help - and it delivered.


Overview



Engagement should not be a campaign goal


A few people sent me a deck created by a manager at Facebook. It's all about digital metrics: which ones to use, and which ones to avoid.


The basic gist of the presentation is "use metrics that are clearly tied to your business goals", which is solid, standard advice.


However, it goes one step further and destroys the idea that engagement in any form is a meaningful digital metric.



Clicks don't deliver brand results

Clicks aren't a good proxy for brand results
Slide 16: Clicks aren't a good proxy for brand results

There is no significant correlation between click-thru rate (CTR) and any Nielsen brand effect metrics, such as ad recall, brand awareness or purchase intent.


In other words, the number of clicks an ad gets has no relation to being more memorable or effective at selling a product or service.



Clicks don't predict or deliver sales

Clicks are not a proxy for sales
Slide 17: Clicks are not a proxy for sales

There is no correlation between click-thru rate (CTR) and return on investment (ROI).


Ads that get clicked more don't sell more.


Clicks are not a good proxy for Business Performance
Slide 19: Clicks are not a good proxy for Business Performance

Facebook internal data shows that 90% of people who viewed an ad then bought, did not click.


Ads do not have to be clicked to be effective.


Read that again. Facebook themselves are telling you that clicks are worthless.



Download the report


So there you go: three slides that destroys the idea that engagement (of any kind) is a good metric to use for marketing campaigns.


Download the full report and use it the next time you need to convince someone that engagement is a bad digital metric.



Here's another Facebook report, that spells it out even clearer.


"Engagement should not be a campaign goal. If you aim to generate demand for or change perceptions about your product, engagement metrics should not be used to assess your advertising effectiveness. They are helpful, but not sufficient given people can be persuaded by content even if they don’t engage with it." - Facebook for Business



What social media metrics should I use?


If using engagement is a bad idea, what metrics should you use instead?


As Facebook suggests, you need to use metrics that are tied to your unique business needs. So if you're selling widgets, then widget sales attributed to social media activity would be a good place to start. Look at your buying funnel and choose metrics that align with each stage - this will be different for every category and business type!


However, there are a few general metrics that can apply to most situations...



% Audience Reached


Reach is crucial to the success of any campaign.


A good rule of thumb is that a campaign needs to reach 80% of your audience a few times a week/month in order to have an impact. You should know roughly how many people are in your target audience, (if you don't, I show you how to figure it out here) so dividing the total impressions of your campaign by the audience will tell you how many of them likely saw it.


Remember, you need to reach them at least once but probably not more than 3 times, so keep an eye on your campaign frequency capping. Using a reach map can help visualize your audience reach.



Share of Search


Pioneered by Les Binet and James Hankins, Share of Search is quickly becoming the new way to measure the effectiveness of advertising in the digital age.


In a nutshell, Share of Search measures the total organic searches made for a specific brand divided by the total searches for all brands in that category.


It uses freely available Google search data to predict brand growth and measure the impact of advertising campaigns in near-real time.


James has an excellent blog post that explains how to do it, and you can watch Les explain the concept in greater detail below.



If you have other presentations, reports or stats that kill the idea of using engagement as a social media metric, send them over and I'll add them here!



P.S. If you found this helpful and are feeling generous, give this a like on Twitter - a like or retweet is hugely appreciated!




Comments


bottom of page