While brands posted significantly more on social media in 2020, engagement on the whole was down across brands of different sizes and industries.
Even though engagement was down, people spent more time on social media in 2020, in large part due to closures and shutdowns brought on by the coronavirus pandemic.
To increase engagement in 2021, brands should consider posting less frequently in general, more frequently in the wake of big, attention-grabbing events and utilize video more frequently than photo or text posts.
2020 has been a long, strange trip. What started off as a pretty routine year was quickly upended in mid-March when we all learned about the novel coronavirus, or COVID-19, and just how quickly it could spread. One of the effects it had? It drove people to spend more time on social media.
Given that, one would think there would be a pretty consistent trend that social media engagement would be up across the board in 2020. But yet, there have been some rumblings that isn’t the case - particularly in the fall of 2020, when a lot of events were competing for attention. I’m in a few social media related groups, and this has been a common theme I’ve heard recently. Engagement on my posts is down lately, or my engagement has fallen off a cliff, should I post less often?
Since there understandably weren’t metrics associated with those statements, that inspired me to ask a question: What does the data say about engagement compared to last year? Is it really down, and if so why? If not, what could be driving the perception it is? And what does all of this mean for engagement moving forward?
That’s what I, with an assist from Brooke Reinbold, set out to answer. Let’s dive in.
I chose Facebook as the platform I’d use for this experiment. Although it is time consuming, it’s relatively straightforward to see what posts people engaged with, and in what numbers. This is an important point, as companies don’t just volunteer this information. It’s also widely used - according to Business Insider, an estimated 177.1 million people in the United States are Facebook users, so it should give us an accurate representation of what is going on in the social media industry.
I focused only on organic social media as posted on the company page. For this study, paid advertising was not considered. Obtaining metrics on paid would require a level of access that as an outsider, I did not have. Organic posts on the other hand can clearly be seen and tracked, and since they are posted publicly, there are no privacy or proprietary data concerns. One flaw here is that boosted posts (posts that brands paid to have exposed to more people) were not segmented out here. Without access to back end metrics, there is no way to tell from the outside looking in if a company boosted a post, and as you might imagine some brands are less than willing to share that information. So I took the posts at face value, even though some may have been boosted, which could impact engagement.
Next, the brands that went into the study were selected. While the goal was diversity in size and industry, there were a few benchmarks brands needed to hit to be included in the study. Specifically:
They had to have at least 1,000 followers (this helps rule out low engagement as a function of low audience).
They had to post, on average, at least 8 times per month (or twice a week).
They had to do significant business in the United States. They could be global brands (and some were), but to make sure the playing field was level between larger brands and smaller brands (who did business almost exclusively in the U.S.A.), a significant presence in the country was required.
From there, I selected a set of brands to study. I purposely selected different industries to rule out any effects on engagement that just hit one industry. In order to protect identities I won’t reveal the exact brands that were studied, but some of the industries include:
Sports and Recreational Equipment
Social Media Consulting
All told, it worked out to 12 different brands across these categories, for a total of 6,014 posts between December 2018 and November 2020. The study was started in late November 2020, so the values for December compare 2018 and 2019. All others compare 2019 to 2020. Plenty of data to work with.
From there, Brooke and I manually went in and pulled the data for all social media posts dating back to December 2018 for the twelve brands that we picked. This was all done manually - I recommend having a computer with a fast processor and lots of RAM for anyone crazy enough to try this.
There were five things we tracked with each post:
Date of post
Number of reactions (what kind was irrelevant, an angry face counts the same as a heart)
Number of comments
Number of shares
The type of post
When it comes to post types, they were broadly bucketed into four categories:
Text: A post with just text in it
Video: A post with a video playing that did not link to another page or website explicitly
Photo: A post with a still photo(s) in it that did not link to another page or website explicitly
Link: A post that was either shared from another page or one that explicitly and clearly linked to a website
I acknowledge that a large portion of Facebook posts could be characterized as link posts in some fashion. For this study, separating them out like this makes sense, so we may be able to pinpoint certain types of posts that outperformed or underperformed.
The brands were then categorized as small, mid-size or large brands, depending on the size of their following. Small brands had at least 1,000 followers but less than 10,000; mid-size brands had at least 10,000 but fewer than 100,000 followers; and large brands had at least 100,000 followers. There were four of each in the study.
What The Data Shows
To get started, brands posted more in the last year than they did in the prior year. Like, a lot more:
I suspect the impact of COVID-19 had something to do with that (a topic I’ll return to later), but the uptick really started in December 2019 and kept rising from there.
What group drove this change? From the data, our mid-size brands, already quite active, stepped up their posting:
One interesting thing that caught my eye: take a look at the red line and blue lines, representing large brands. See how starting in March they diverge, with prior years’ heading into a valley while current year keeps going up, peaking in June? There’s a pretty important reason for that, which I’ll get to in the next section.
So now for the million dollar question: what did this mean for how the audience engaged with the brands? To start, I took a look at the median engagement each month for each brand. The median is less prone to wild swings because of outlier posts with huge engagement.
The answer isn’t as straightforward as you might think.
First, in taking a look at the change from one year to the next, there were 61 months across all brands where the median monthly reaction was higher than it was the same month last year. There were 73 months where it dropped, and 6 where it stayed even (two of our large brands had two months each where they didn’t post anything, so those four months were tossed out for this exercise).
Most of the drop came from the larger brands:
I did the same thing with shares. For our brands, the median number of shares per month was up in 57 of the months measured, down in 62 of them and even in 21 (remember, there are four months that were thrown out due to lack of data).
Again, our large brands were the primary culprit, though small brands fared pretty well (though it must be said their bar to clear was considerably smaller when it came to last year):
Of the months that dropped, were there any where the drop was more pronounced? Turns out the month of November was responsible for more drops than any other (which validates a lot of what I heard in social media groups). When you look at year over year drops by month for both engagement and shares, you see a pretty clear pattern that activity drops off as the summer goes on and turns into fall:
Comments are a little trickier to deal with. While they are certainly a key method by which people engage with brands and generate buzz around topics, they can be a little misleading for a study like this. For one, because brands can - and often do - use social media to talk back to customers, a good portion of comments on some posts may be from the brand and not the customer. Customers can also leave more than one comment on a post, unlike reactions and shares, which people can only do once per post (the main post, not reactions on comments, which was not tracked here).
In addition, they aren’t evenly spread out. As you might expect, the larger brands tend to get more of them. Interestingly, the small and mid-size brands didn’t register a ton of these - for the 12 brands I looked at, for three of them the median amount of comments for each month was zero for every month. For three others, there were very minimal changes for just a few months - many other months recorded zero comments as their median for both current and prior year.
So, when I tell you that comments on all posts regardless of brand size were actually up 29% year-over-year, take it with a grain of salt. And when I tell you that for only three months did total comments actually fall, don’t read too much into that.
One last data point I looked at was how many ‘dud’ posts there were. These I defined as posts with no engagement whatsoever….no reactions, no comments, no shares, no nothing. We had 280 posts like this across the two years studied, 184 of them coming in current year. As one might expect, large brands accounted for none of them; the majority was in our mid-size brands.
Taking all the evidence into consideration, I can say that general engagement as a whole was down in 2020.
Now let’s get into some reasons why that might be the case.
What Drove Behavior in 2020…..and 2019
So we’ve established that engagement is down. Let’s dig into what some causes could be.
First, one might think that people are using social media less than they did in 2019. At first glance, they might be right. According to Business Insider, people in the United States were starting to spend less time on social media heading into 2019, with that number projected to plateau in 2020:
As you might guess, these projections were done before the COVID-19 pandemic upended lives and caused lockdowns in March (the forecast was done in April 2019). When this study was re-done in April 2020 in the wake of COVID lockdown, take a look at how things changed with three significant social channels:
Bottom line: once COVID hit, time spent on social media increased. So we can’t then say people spending less time on platforms drove engagement down in 2020.
Another commonly held belief is that the amount of things happening in society has something to do with a drop in engagement. The theory goes that the sheer amount of things that happened caused something of a social media fatigue, where there was so much competing for attention that it caused metrics to drop off.
I don’t think it’s that simple. There’s no doubt that a lot has happened in 2020 (if you’re learning this for the first time, I hope the rock you were living under was really nice), but there were more than just the big ticket items. Below I put together a timeline of some things that drove engagement in the United States in 2020. Most of these you know, but some you might have forgot:
No doubt that was a lot. But here’s why I don’t buy the ‘there was too much going on’ theory: a lot of things happened in 2019 too:
As you can see, 2019 had its share of noise. While there are doubtless micro stories that impact individual behavior, 2019 had plenty of events and things happen that drove people onto social media. On a macro scale, I don’t think it’s accurate to say engagement was down solely because a high volume of things happened, and that created a high volume of competition for attention on social media.
Instead, I think the fact that the major events that happened in 2020 impacted a much larger proportion of people than the major events in 2019, and that may have contributed to the drop in engagement we see. The three major stories in 2020 - COVID-19, race relations in America, and the election - were omnipresent throughout the year (and, except for the election, still are). In fact, a portion of the posts brands made were related, directly or indirectly, to one of these three topics.
Those were engagement grabbers in 2020. Go back to the earlier graph of what months had the most drops when it came to reactions and shares. What I’ll call disengagement was up in January, dropped slightly in February and dropped sharply in March. After staying steady the next two months, people were much more engaged in June before becoming disengaged as the summer and fall went on.
This coincides with the start of the COVID-19 pandemic, when brands posted to let customers know what business would look like with shutdown orders coming down, what customers could do to keep safe and keep buying, and what businesses would do to keep customers safe as shutdowns lifted.
The June drop in disengagement lines up with the nationwide protests and conversation around racial equality following George Floyd’s death. During this time, many brands posted messages stating, among other things, their commitment to equality and their support of the Black Lives Matter protests taking place across the country. There is ample evidence that customers expect brands to take stands on social issues, so it’s no surprise posts doing just that would drive up engagement.
As for the rise in disengagement that happened starting in July? By this point, many states had reopened and people were generally used to living with COVID and what that meant. Businesses had reopened and communicated to customers how they were going to help keep them safe to inspire confidence to return. In addition, most of the protests surrounding racial equality had died down, though the conversation still carries on to this day.
Basically, there wasn’t much new to react to on social media that affected a wide swath of people. That, I believe, drove engagement down until October, when election season was in full swing and people were talking about what they thought, or hoped, would happen. Once the election was decided (though maybe not for everybody) in November, we see disengagement rise again as there’s one less big attention grabber to talk about.
This, to me, is the key driver for the engagement drops we see in 2020. It’s not that there was a lot going on in general that drove engagement down, but that once the aftershock of things wore off, there was a little bit of fatigue that set in on the part of the user. Even the most hardcore fans of brands don’t hit like on every single post that brand makes, share every photo or comment on every post. This is where social media posts in general being up in 2020 worked against brands.
With a higher number of posts from individual brands out there - not noise in general, but increased activity by individual brands - there may have been some fatigue that set in for folks on engaging with individual brands. Here I think is where the ‘there was so much going on it’s hard to cut through the noise’ argument holds some merit. There was more noise, but brands were creating much of that noise through increased posting. While there are certainly valid reasons for this, keeping up a high volume of posting can make posts start to blend in, which makes getting consistently high engagement harder.
To sum it up: engagement goes up when there are big events going on. But once the impact of the event is fully felt by the audience, engagement in general slips. Particularly when the amount of posts by an individual brand is up.
Looking Ahead to 2021
The fair question to ask at this point is: what can we expect from 2021? Will engagement keep dropping? To an extent, it’s impossible to tell what 2021 has in store. There are however some things we know are going to be taking place. Going in chronological order of how I think things are going to play out, there are some big ticket items I see on the horizon….
Donald Trump Will No Longer Be In Office
Regardless of what you think of him, Donald Trump - and the things he did, said or tweeted - were ubiquitous on social media in the last four years. Much of what he said or did, or the policies he pushed, tended to garner big engagement. For example, when Trump was hospitalized while battling COVID-19 in October, people flocked to his Facebook page to engage with his posts. Countless others went onto social media and had something to say about it. The man drives engagement like few can.
That’s why I think engagement from political activities will be down next year. While it’s impossible to know if someone will do or say something politically that will drive people to social media, we will hear from President-Elect Joe Biden a lot less - he has already gone on record as telling the press not to expect him personally to talk everyday, through social media or otherwise - and when we do hear from him, at least on social media, the engagement he gets dwarfs that which his predecessor got.
For those hoping the man in the Oval Office will be quieter on social media, you’ll get your wish, but you won’t get the engagement driving things you previously did, at least in this way.
The COVID-19 Vaccine Rollout
As I type this, the first doses of COVID-19 vaccine have been given to people in the United States that weren’t involved in a clinical trial. This kicks off the largest vaccination effort in U.S. history, one that will continue well into 2021. Expect there to be a big public information campaign to convince Americans to get the vaccine, as 58% of Americans as of November are willing to take it, below the 70-75% threshold public health experts say the country needs to achieve herd immunity though the vaccine.
Expect social media to get involved, and to see engagement rise somewhat as a result. This could come in two ways. First, the social media channels themselves could get involved directly in trying to get people out to get vaccinated. Remember during the election when Facebook not only posted reminders to register to vote, but also gave people resources to figure out where their polling station was? Expect the same here, but instead you’ll see the platform point people in the direction of where they can get a vaccine.
I expect the bump that will come from this to happen around March or April once the vaccine is, hopefully, widely available. I do think brands may post about this topic, almost as a public service announcement (like when posts with CDC guidelines circulated), but that the bump from this event will be modest. There is one more big event that I think will mean more for increasing engagement….
The Return To Normal
Think back to May and June as businesses were starting to re-open. Many brands that needed to attract customers back to physical locations posted all kinds of things about steps they were taking to keep customers safe. There was good reason for this - there was ample evidence that suggests a key factor in customers returning was whether they felt safe (this webinar has that nugget, as well as some other good ones from early on in the pandemic).
Come next year, I think a major story will be pushing the return to normal life. Brands, particularly ones where customers come back in person, will be at the forefront of that. This I think will be the next big item that drives social media engagement - and will create a wave that all brands can ride. Whether it’s welcoming customers back en masse, making things feel normal again or posting photos of a parade (New York City has promised to throw a ticker-tape parade for healthcare workers once the pandemic is done), the return to normal will drive engagement up in 2021.
Tips To Increase Engagement
If you’re looking for ways to increase engagement in the new year, I have two suggestions.
First, think about video posts. If done right, they tend to get pretty significant engagement. For the four different post types I looked at, I took the median amount of reactions per post type, per month. Take a look at what percentage of that sum video was responsible for each month:
Even in the months they didn’t account for over 50%, they still accounted for a healthy chunk of reactions (26% was the low for the 24 months). Now consider that, of all the posts studied here over two years, video posts account for 15% of them. This screams underutilized method to me, and one brands would be wise to take advantage of.
Second, I would consider posting less in 2021. Individual metrics here are going to be what drives this decision, but based on the evidence here, more may not be better when it comes to how much individual brands post. It’s easy enough to say “just post your best content”, but that’s a pretty tough thing to consistently nail. Moving forward, I would be more selective in terms of how often I posted if the goal is to maximize engagement - which it might not be in all cases.
The one counter argument to this would be to increase posting when there is some significant event going on in the world that drives people to react on social media and try to ride that wave. In that case, I would post more frequently to take advantage of the surge we talked about earlier. This can seem like trying to time the stock market, but here you can still reap plenty of gains even if you are a fast follower. Always be authentic and true to your brand of course, but take advantage of the opportunities external events hand you.
This post was created with the help of Brooke Reinbold, currently a sophomore at The University of Michigan.
Oak Moon is a consulting agency based out of Columbus, OH that helps companies market their brands, define value propositions and uncover customer insights, among other services. If you are interested in hearing more or have questions or comments about this blog, feel free to reach out to me at firstname.lastname@example.org.