In 2017, we launched an all-new annual report, the Social Media Industry Benchmark Report, with the intention of giving you a better sense of how your industry is performing and how your social media compares.
This year, we not only expanded the industries included but also decided to do a deep dive webinar into the 2018 Social Media Industry Benchmark Report.
During this webinar, Co-Founder and Head of Product, Seth Bridges, analyzes the report pulling out key insights. Of course, Seth went even further and also provides brand NEW insights that didn’t make it into the original 2018 report.
During the webinar Seth:
- Explained key findings and all NEW insights that didn’t make it into the report
- Dived into top-performing brands and campaigns
- Shared what’s working and how you can make it work for you
- And much more…
Download a Copy of Seth’s Slide Deck: New Insights and Examples
Read the full Social Media Industry Benchmark Reports:
- 2019 Social Media Industry Benchmark Report
- 2018 Social Media Industry Benchmark Report
- 2017 Social Media Industry Benchmark Report
Seth Bridges: Hey, there. Good morning, everyone. My name is Seth Bridges. I’m one of the founders of Rival IQ, and I am really excited to have you here with me today. As all of you know, we published the 2018 social media benchmark study back in April, and now it’s been a few months. All of you, I presume, have gotten a chance to read it. And many of you had questions for us in terms of more – how else to think about it, how else to look at it, how else to dive in. So, we went back and did a bunch of additional research, dug in. I’m really excited to share all of that with you today. So, before I get going though, I am recording this. And I’m sharing my screen, so you can all see my slides. I’m made that mistake before. And I got the questions panel up here, so please do ask questions as we go.
I’ll save some time at the end to talk about your questions. But of course, ask you go. And if it makes sense to even interrupt where I am, happy to stop and answer then as well. Already a question – will I be sending out slides. Yeah, I will be sending out slides afterward, so don’t worry about taking notes. There will be some links that I’ll mention here as we go. They’ll come out in the email. They’ll come out, so don’t worry about that. One other housekeeping note – this webinar…think about as a preview or a warm-up for next week’s Data-Driven Marketer webinar where we’re featuring Lee Odden. I’ll be interviewing Lee, talking about how he and his team at TopRank Marketing use data to elevate the social media strategies and execution of their clients.
Should be a great webinar. Lee really knows his stuff. He’s a wonderful interviewee, and I’m very much looking forward to that as well. And if you haven’t gotten to read the social media industry benchmark report, it is live on our blog. There’s no gating. The whole thing is there. Go get a copy, search for it on our site. It will come up, no problems. That is going to be the basis to what I’m talking about today. But I’m ripping through the report. Don’t worry. You can go read it. I trust that everyone does a good job reading. We’re going to go deeper today. We’ll barely touch on some of the slides that are there, and we’re going to look at all new content. So, stayed tuned. Again, the slides will come out.
So, I want to just set the table, talk a little bit about the background, the methodology, what are these companies, how many, what are we talking about, kind of get the statistical baseline out of the way. I’m going to talk a little bit about the industries like how they compare. And then where we’ll spend most of our time…I want to dive deep a little bit, look at some new analysis, look at some time series. So, how do these numbers change across the year? And then I found some real-world examples from some of the top performing industries on each of these channels. And I want to look at a few of these sort of micro case studies with you and see…
Hopefully, that will spur some ideas about how to improve some of your own social media marketing. So, let’s jump right in. Background and methodology. So, this year, we did 10 industries with 150 companies per industry. Last year, we only did six. So, this felt like a big improvement, looking at a broader swath of the world. Obviously, we track over 150,000 companies in our database. So, what we wanted to do was find companies that have at least 25,000 fans on Facebook last year, up to a million. So, we’re not doing on the smaller end, and we’re not doing the massive.
This is somewhere right in the middle. We also made sure that all these companies had an Instagram presence for the entire year and had a Twitter presence for the entire year. And that they were active on all those presences. So, this is not necessarily meant to be representative of every business. This is active businesses on Facebook, Instagram, and Twitter pulled from our system. And here are the industries – higher ed, nonprofits, media, hotel and resorts, sports teams, fashion, food and beverage, health and beauty, home décor, and public figures or influences.
You’re going to recognize a lot of these logos. Some of the places where some people have had questions like, “Oh, I don’t see X, my favorite sports team.” Yeah, million cap from the Facebook fans really was one of the criteria that we used to select these companies. So, you’ll recognize a lot of those logos, the Sounders, Avalanche, Oakland A’s, the Buffalo Bill’s, for example, as well across the board. But just keep that in mind as we’re thinking about who exactly all of these companies are.
The entire data set that we’re looking at was about 5 million social posts, and a total engagement of about 2.3 billion. So, this is the observable public total engagement that you would see, for example, in our product in the social benchmarking section. This is like likes, reactions, comments, shares, retweets, etc. across these different platforms. And that is the data set. Now, how does that break down? Facebook was about 1.7 million of those posts. Instagram only 600,000, so about .6 million, and Twitter about 2.7 million tweets. Now, I know a lot of people think about Twitter as this high volume platform. There are no replies in here. We don’t count your retweets.
This is just owned tweets going out for each of these brands. And not everybody is into the automation and scheduling ten tweets a day. It turns out maybe that’s less of a thing these days. But you can see the ratios of how sort of activity goes across the data set. But in terms of the interactions, the engagement, Instagram, the overwhelming contributor to that 2.3 billion with about 1.6 billion comments and likes from the posts from the 600,000 posts. We’ll do the math here in a second just to see how that breaks down. But these are the totals. Now, remember, media was one of our industries. Media an activity producer through the roof.
So, the average isn’t necessarily the average across all companies. We’re very concerned with medians here, trying to sort of deal with some of the skewness of the data. So, the average company in this study with about 1.1 Facebook posts a day, about .7 Instagram posts a day. So, that’s about once a weekday. And about 1.1 tweets a day as well. So, if… You’re already saying, “What do I do for my social media?” This is the median company, the middle. 50% do more, 50% do fewer. But right there is what we’re talking about. And from an engagement rate perspective, the platforms are different.
The audiences are different, the way in which you engage the…the UX around how you engage, it’s different. Why you engage is different. So, not to be surprised, Facebook’s average engagement rate is about .16 per post. So, this is interactions divided by followers, we’re trying to normalize out how big your audience is. We’re also doing that on a per post basis. So, if you post a lot, it’s not like we’re adding up all the engagement. We’re normalizing followers and posts, trying to give us a comparable way to look at this. Instagram clearly much, much higher engagement. Order of magnitude higher – 1.73% on average. And Twitter down at about .046%. So, we’ve got an order of magnitude in either direction, Facebook in the middle, Instagram on the high end, and Twitter on the low end. That’s our data set. Those are just some really high channel comparisons.
Spend a moment looking industry by industry. So, where I’ve stolen just a few slides out of the actual report itself. On the X axis across the bottom, posts per day on Facebook. On the Y-axis, the average engagement rate. Every little dot is one of the industries. So, you’ve got a clutter over there on the left. And I want you to note the one outlier on the far, far right, media. Almost ten Facebook posts a day. But dead last in terms of average engagement rate on a per post basis. So, you can see even if we exclude them, there is not really a strong correlation between posting more and more engagement on a per post basis. This makes sense. We know how the algorithm works generally in that pumping a ton of content out there doesn’t matter. Right?
It’s all about producing high-quality content. And so not surprised to see a lack of correlation here. You can note though that food and beverages influencers, hotels, among the top all producing in that one to two posts a day. Higher ed, one of our generally top social media performers is up there in that kind of top section as well. On Instagram, you see that the posts per day are much more narrowed down. Obviously media, not a big outlier here, not a ton of content being put out on Instagram from the same media companies that we see on both Facebook and in a second Twitter. Sports teams, however, look a lot like media. They produce from an activity perspective, from a post perspective way more posts in general, particularly on Twitter than some of these other folks as well.
But their engagement is much, much higher. I think in general, we see that higher ed, which is our top at over three percent… That’s almost double. In fact, it is double the average industry here. That’s a whole special topic that we’ll talk about in other webinars. Higher ed is just an amazing community. But sports teams…people love their sports teams, and it appeals to a broad set of folks. And you see that here even though they produce a lot more content on average, that engagement rate much higher. Notice though nonprofits, hotels, food and beverage, also still kind of on the upper half of our performers as well. And then Twitter. So, this is where we see, again, media, really high production outlier.
Same with sports teams, but they are much, much higher…more than two X. In fact, they’re kind of measuring up at the bottom of the other industries that we looked at — beauty, fashion. But again, look at hotels and resorts, food and beverage, higher ed, nonprofits. All those industries just kind of keep floating up independent of which of these channels we’re talking about. And so today as I dig in more, I’m going to make sure to pull some examples from some of these more top-performing landscapes. I see a quick question here. Let me check. Make sure it’s not something horrible like my screen isn’t sharing. Okay, does food and beverage include restaurants. Food and beverage…these are broad.
So, they could be producers. They could be restaurants if they’re large enough to have 25,000 or up Facebook fans, depends when we look at restaurants that are kind of…particularly if it’s a multi-site situation like a nationwide chain for example. You’ll often see one kind of global handle as well as some local handles. We have a mixture in this data set. It’s also not US specific or even North American, or even English specific. And so we’ll often see outposts of particular hotels that do really well in a particular place in Canada, or in Asia, or in somewhere. Those kinds of companies will end up here as well. We tried to sample without bias from the data set that we have because we wanted to get a clean look at how the channels are performing on average.
And didn’t really want to introduce selection bias in that way. So, that’s our industry by industry comparison. Again, if you want to get to your industry or dive in on all the specific post type metrics, hashtag metrics, etc. on an industry by industry basis, that’s…you definitely want to go look at the report and go find those specific numbers there. I’m going to spend the next…rest of our time together going through Facebook, and Instagram, and Twitter. And I want to ask a question or look at some of the…how these numbers have changed over time, and then try to pull a little bit on that and figure out are there examples of things that are bucking the trend, or aligning with the trend, where we can learn something.
So, on Facebook, what’s the conversation been for a long time? Maybe we’re even done having this conversation. But okay, organic reach declining, gone, not gone exactly but declining. And does that mean declining engagement? I think everyone out there is going, “Yeah, yeah, okay. We’ve heard this. Yes.” But we didn’t include any time series information. We didn’t show any trends over the year. But here, we went back and did additional work to be able to bring you some of this data. So, Facebook, average engagement rate per post… This is one data point per month through the entirety of 2017. So, we’re starting at over .16%. And then month over month, trending down.
Just a slight little lift in July before we headed down even further to below .1%. So, we saw a drop off of about a third from the beginning of the year to the end of the year. And if you saw your Facebook organic performance declining throughout the year, obviously you were not alone. Again, this is representative of the median company. And so the question you might be asking is like, “Well…a couple question…why? Are people posting less, posting more? Did that have any impact? And then are there any companies that are perhaps bucking this trend?”
So, the first look I want to give you is, again, something that we’ve never published before. It’s apart of the study. But let’s break that time series down by industry. And I sort of faded out most of the industries here. But if you just kind of squint through for a second, you get a general feel that yes, they are all down and to the right, which is generally never a good word to hear. I did highlight food and beverage as one of the companies that performed near the top of all of our industries. And you’ll see that everybody is going to have some noise.
But generally actually flat through the year, maybe down from the summer a bit. But in general, flat, flattish. Which when the global line looks like this, hey, sometimes stable is a good thing. So, were people posting less because of this? And on average, the answer is no. We see that 1.1 posts power day, generally consistent throughout the entire year. A little bit quieter in the winter, particularly in December. Also a little bit of a blip there in July as well. And again, this is across 1,500 companies. This is a thing that in the summer, there are just maybe five to ten percent fewer posts on a monthly basis, at least in the middle of the summer there. At least in the northern hemisphere.
So, as we saw, is everyone declining? No. Are most people declining? Yes, absolutely. I went on a hunt. Pick one of the top performing industries like food and beverage and try to go find companies that saw either flat or growth throughout the year. And I will tell you, it was a little bit of a hunt. But my job is to find you examples that sort of buck the trend. And let’s see what we can go learn. Hi-Chew. How many of you are familiar with Hi-Chew? This is some chewy, taffy, sugar candy thing. My gym has these.
It seems very contrary to fitness and then eating kind of sugar bumps. But they’re delicious. My kids love them. I love them. And I was happy to find them here that they were, A, in our study, and B, that they were one of the companies bucking the trend. So, food and beverage, they make…this is a processed candy in a bag. You buy it at the grocery store or Walgreens and Target as I learned from their social media. They published in the entirety of the year, 205 Facebook posts. So, well less than one per day. And they saw on average a .35% engagement rate per post.
Remember, the industry average was much, much lower on the order of .16. So, these folks are doubling up. But look at the time-series graph pulled straight out of Rival IQ here on a month by month basis. They’re actually up, up and to the right. And yeah, it’s noisy. But again, everything is going to have noise in it. Over the year though, the latter half of the year versus the first half of the year. They’re definitely, definitely up. And now we’re going to go and dig into why or why I think why.
And I always find it instructive to go look at the top performing content for the company and just ask a few questions – what are they doing. So, visually is a great place to start. And thee are the top performing posts from Hi-Chew for the year of 2017. You’ll notice very consistent from a coloring perspective. Obviously, it’s very heavily influenced by their actual packaging. But I can tell you the color of the actual candy. The upper two left posts were boosted almost certainly. They do not have a… They have lots of engagement, which is why they’ve risen to the top. Our system noted them as flagged. When you inspect it, it certainly seems like they were promoted, which is not the right mix of engagement on those posts.
But they also don’t look anything like the rest of these posts, which feature product, tight crop, pretty well produced, and good consistency. Whereas if I… And we look at those posts, and they’re like thousands of thousands of engagements on a per post basis. Now, I went and looked at the worst. Like, show me their worst content for the year. Just pause and look at that for a second. It is different. In fact, it doesn’t look… It’s like two different social media managers were in charge of the best and the worst – consistent, tight crop, featuring the product… These are not that. First of all, there are some people in there. I like people, but clearly, these are not the people shots that people resonate with.
There’s stuff that’s not product. There’s a Christmas… Whatever. It doesn’t even look the same. And we’re talking single digit engagement on a bunch of these. And it’s not like these at the beginning of the year or these were… These were spread throughout. And so okay, visually we can see there’s a difference. But let’s go jump in a little bit more. So, here was one of those top performing posts. “We are falling for new sweet and sour mystery mix…” Okay, so they have a new product launch. They clearly have an engaged generally just on the average. But read down on the copy a little bit. “Try the mysterious flavor and take a guess to when the ultimate Halloween hookup…”
Okay, and they have a link that takes you to a place where you can register for this contest. And okay, I’m not surprised. Contests are a thing that some companies do on social media to drive engagement, particularly when you can do things that drive virility like comment and share. And clearly, the audience here likes to do that a lot. So, you might think, “Oh, well, they’re just doing all contests. I don’t want to do all contests. That’s not my thing.” But it turns out only about 10% of the posts of those 205 posts that Hi-Chew did were contest posts. And on average, the companies in food and beverage were at about 7%.
So, yes, Hi-Chew is doing almost 50% more content posts than the average company here in our food and beverage landscape. But it’s not like it’s all their content. It’s not even half their content. It’s not even a third. It’s 10%. I can tell you though they do the post and announce the winner later. Those are horrible. They don’t get any engagement. So, if you’re running contests, maybe trying to get the win on the contest part is a good thing. So, Facebook, Hi-Chew, contests. I don’t know if you’re running contests. Think about it. The way that I do a lot of this analysis… Again, we’re either in Rival IQ or Tableau from Rival IQ. And I’m out looking for words like win, guess, comment, tag, comment, winner. These sorts of words are good keywords that you can start to use to filter down and find those contest posts as an example.
Okay, Instagram. Instagram had…remember, order of magnitude engagement on Facebook in the 1.7 range. So, here we go again looking at average engagement rate on a per post basis across all of the 1,500 companies. And you see that it was trending down a little bit there in the first half of the year, kind of stabled out in the summer. And then we see a bit of a rise there through Q4. Back up to levels that look like what we were seeing at the beginning of the year. Now, let’s do the same dance. We’ll break it down on a landscape by landscape basis, on an industry by industry basis. And here again, I’ve let most of the landscapes fade to the back.
I’ve brought hotels and resorts forward. That was another one of our top performers. And you can see that first of all from the back, everyone again was experiencing… This is the beauty of averages when you break things out. Most landscapes were experiencing the same thing – slight downward trend through the beginning of 2017. Obviously, as Instagram has put that algorithm into place, fewer people are going to be seeing your content. I’m sure all of you are seeing that in terms of your reach as well. And if your kind of content quality is consistent, and your new followers that you’re adding look a lot like your old followers, you should expect to see engagement rates consistent.
Of course, if you are changing up how you’re doing things, attracting new or different audiences and engage differently, you might see something different. But look at the far right of the graph here. Everyone lilting up. That’s higher ed there at the top. Like really driving hard through the end of the year in terms of increasing engagement. But almost every company, every landscape here, saw that bump from out of October into November, and then flat through December. From a content production perspective, on average, we’re still looking at about once a weekday.
Now, I know we can all schedule posts now on Instagram through a variety of mechanisms. That was not necessarily a thing that people were doing before. But .7, that corresponds to about once a weekday. And we definitely see on many, many, many business’ Instagram handles…if we look at posting on a day by day basis with Sunday to Saturday, you get consistent production of the week at one and then falls off on the weekend. And you see that here flat at .7 per day. So, thinking about that engagement rate though…
Facebook was declining. Instagram a little bit down, stable, and then maybe even back up here at the end of the year. So, I’m checking out some questions here. Just affirmation. This is wonderful. Thank you, Alexander. So, stable engagement rates. Good, so now my job was not to go find somebody who was say just stable or down, but let’s go find some people who are really killing it from hotels and resorts. So, who did I find? The Fairmont Chateau, Lake Louise. This is a boutique hotel in the Fairmount family located near Banff, maybe even in Banff National Park. So, this is in Alberta. This is the province just one to the east of British Columbia in Canada.
See, I live close to Canada, so I know a little bit about it. And the Fairmont did 326 Instagram posts last year. So, that’s nearing one a day. So, we already know they’re doing more than average. They’re not… This is not some of the beauty brands we see – 5 a day, 6 a day, 10 a day. You know, super high content production. But look at their average engagement rate. 4.2% on a per post basis. And then their time series… They actually saw a little bit of a decline through Q1. But then almost flat or up every month through the end of the year, and then seeing that sharper lift as the year kind of ended there in Q4. Remember, 1.7 was our overall average. 4.2…
This is higher than our average university, which is already our top engaging industry in our social media studies that we always do. So, let’s do the same dance now that we did for Hi-Chew. Let’s look at the visual content, dig in, see what we find, think about how we can apply it to what we’re doing. Okay, just take it in for a second. I’m going to look at over. I got the big screen over here. It’s beautiful. It’s beautiful content. This is, again, all their top performing by total engagement. A strong consistency from a coloring and a quality perspective. Not every shot is without a person, but most of the shot are without people. There’s obviously a very famous window you can see this kind of… The hotel must look out in some beautiful, I’ve never been. I don’t live that close as it turns out.
But just I love the content. And you just love the consistency. But now let’s go look at their least well-performing content. And it’s different. Right? And I’ll flip back. Take in the color, the consistency, the framing. You see even similar shots where the fire pit is looking out on the lake with the mountains. It’s awesome. Great. We jump over here… The first thing that jumps out at me is people. And I like people. I like people. I got kids, I’m married. I like people. But one, two, three, four… There are people in more than 50% of the shots. And I don’t know that that content…
Random people in your feed maybe is not something that you react as strongly to as a beautiful shot of some place that you’ve been, or would like to be, or is near you, or that you aspire to. Their content certainly proves this. Now, this is their worst performing content. The engagement rates here are still 1.5%. They’re still average. So, most companies would kill to have their poor content be better than most people’s kind of average or good content. But take it in. This analysis… Show me visually. Look at the campaigns. Look at how this stuff is framed. Look at how it’s produced. We can see consistency. We can see execution. But you have to dive in. So, we’re going to go from the good, and now I’m going to jump ahead here. We’re going to dive in. Look at that shot.
Turns out it’s totally a composed shot if you read the comments. The author is like, “The sky wasn’t real. It was like daylight, and I put a sky behind it.” But it still looks amazing. And this person, DelJay Photography, this was their shot. And you’ll see… Read down the comment there. Fairmont at Lake Louise says, “Incredible. We’d love to report with credit.” And of course a comment or two later, the author says, “I’d be delighted for you to repost.” And then what we’re seeing actually here is Fairmont’s repost. And you’ll notice they’ve got the camera emoji, giving the credit for the photo to Deljay Photography. Oh, okay.
So, now you got me. Now I’m wondering, do they actually have a staff photographer, or are they reposting other user generated content? And it turns out that, again, using sort of keyword search and looking for all the phrases that would indicate that this picture was not from them… So, there are a couple different picture emoji’s, the one with the flash, the one without the flash, pic by, picture by, photographer, quote, credit… There’s lots of words that you can use. But we’ve got a pretty flexible search infrastructure. And so it turns out that 78% of the Instagram content that Fairmont puts on their page is reposts. So, curated user generated content.
There is strong travel vibe in Alberta. There’s #ExploreAlberta, #TravelAlberta. Clearly, the travel kind of bureau for Alberta has strong campaigns and runs strong campaigns. The hashtags have high engagement. If you go do some Instagram searching on that hashtag, you’ll see what I mean. And of course, because this is all their content, 4.6% engagement rate, it’s slightly higher than their overall average, which is about 4.2. But the thing I wanted to call out as the comparison is that hotels and resorts generally do well. Again, people like to think about going on vacation. These are destination places. They invest in photography.
They invest in production, something that you want to like on Instagram. But on average, only 18% of the content in this landscape was reposts. And if you’re on a budget, and you have a strong community, or there are people in your community that would like to get exposure from your brand, lots and lots of inspiring kind of Instagram influencers or photographers would like to be in this boat. It’s not like, “How do you get in touch with these people” On Instagram. You can see it right here in the comments. That’s how these communications are happening. That’s how the permission was given and granted. It’s clearly doing very well for Fairmont. Okay.
Let’s move on to Twitter. Twitter, everyone’s favorite. I spend a lot of time on Twitter. Somehow I’ve lost my slide that has my overall average. I’ll tell you the story. We can see through the data when I send out the deck later on. Make sure to get that slide back in there. But Twitter, again…an order of magnitude lower than everyone else, which I’m .05, .04% from an engagement rate perspective. We obviously know there’s lots of content on Twitter. Again, the algorithm there, I find that it makes my life more enjoyable on twitter. I see the content from people I tend to engage with. There are differing opinions. But Twitter was down a little bit through Q1, flat until late summer. And then a downward trend for the year.
But as we can look here, that was not necessarily the case for every landscape. Higher ed is one of our ones there are the top. But I focused here on nonprofits. They also saw a bit of kind of down from Q1, flat. A little lift in September, but then back down toward the end of the year. From a content production perspective though, we actually do see that between the first half of the year and the second half of the year, there was a drop off somewhere in the 1.2 kind of posts a day range. 1.1, 1.2 down to a little south of 1.1. So, 15% or so on average. I know here at Rival IQ, we’ve actually paired back what we’re doing on Twitter. We used to schedule a lot of posts, and now we’re trying to be a little bit more authentic, a little bit more engaging and not just reposting the same stuff all the time. You see it in the data here as well as what a lot of other brands, and personas, and people are doing on Twitter.
So, I’ve heard this, you’ve probably heard it, too. Is Twitter even worth it? This is a question that people ask if you’re a business. And before I even show the data about somebody who’s doing very well from the nonprofit space, why do people use twitter? Do people spend time on twitter? Yes. Are audiences growing on audiences? Yes. I don’t have the slide here for you. But are they growing as fast as on Facebook or Instagram? No. But did the feed make people able to see content that they engage with more frequently? Absolutely. Do real conversation happen on Twitter? Yes.
If you have buyers that are potentially the kind of person who’s on Twitter either demographically or because of what you sale or who you sale to, it doesn’t matter if the engagement rates are low. The engagement rates are low because not everyone is seeing all your content. But if you’re getting good engagement on the content that you have, forget about all the followers you have. It doesn’t matter. This is why engagement rates are hard. You want to be able to dig in deeper on your Twitter analytics, go to analytics.twitter.com.
Look at your impressions, look at your engagement rates from who actually sees your content, and then you’ll get a little bit better idea about the quality of your content. Now, the organic reach of it…that’s a problem that every platform is going to keep having. But if you do customer service, if you’re out there interacting, engaging, getting love from people on Twitter… We get lots of love from people on Twitter sharing our content, reacting to our content. There’s a been conversation about like baseball the other day, and some stats we published.
If we weren’t on Twitter, if we weren’t thinking about Twitter, we’d be missing out on a lot of that. Okay, I digress. Let me jump back in. Children’s Healthcare of Atlanta. This is the children’s hospital in the great Atlanta metro area. And they do a fantastic job with their Twitter handle. Their engagement rates…they don’t even look like Twitter engagement rates. They look like Facebook engagement rates or even better. And so I think it’s instructive to get into why.
First of all, look at the content production level here. So, 205 tweets throughout the entire year. So, that’s fewer than one per day. And they’ve done about 225, 26 retweets of others. So, if you’re kind of following the stream without replies, you’re looking at about 4, 450…about once a day. And that’s clearly what they’ve committed to from a social media perspective on twitter. Again, I told you, kind of down in Q1, flat, and then tailing off there late summer through the end of the year. That is not the pattern that we see here at all from Children’s Healthcare.
In fact, you see them sort of settling in at the .2 range. So, that’s four X what we see on average from our entire study. And they got this huge pop in August from a campaign that they did. I think we’ll actually get to see one of those tweets here in a second. So, let’s do the game. This is our third channel. This is our third chance at the game. Here is all their top performing tweets for the year. And Twitter people think about as a less visual medium. Of course, things are changing over time. We now have nice Twitter card previews if you’ve got your kind of image metadata set up right.
We get nice link preview now without having to necessarily share an image. It’s not a not visual platform. But this is a hospital. They’re not necessarily selling anything. I live in Seattle. If my kids are really sick, I go to Children’s. I don’t…it’s not really a choice. That’s where I’m going. I guess this is similar in Atlanta. But look at the content here – lots of people. So, Instagram for Chateau Lake Louise, there was…the people were bad. Here, we’re showing off people, and it’s not necessarily all patients. Though there are some patients mixed in there. But depending on your level of sort of pop culture ability, you maybe recognize some of these people here. I recognize some, not all.
So, let me give you a little overlay here. Turns out those top two pics, some professional wrestlers from Ring of Honor came. That was that August spike. 4% engagement rate on these tweets. Came by to spend some time with the kids. Of course, social media manager on point, taking care of that. You see this – the Atlanta Falcons. Those are not pictures. This was a mention of those. You got Nick Jonas, Ludacris, Coach Vince Dooley, tagging in the Braves. So, well over 50% of their top performing content included a mention, and it included a mention of a local…either local superstar from some genre or superstardom. But it involved a mention more important.
And so let’s just jump into one of those performing tweets. So, “Thank you @ringofhonor wrestlers for spending the day with our patients!” Clearly, a lot of burly men with belts came by to spend time and included a patient or two in the picture as well. They mentioned Ring of Honor. Ring of Honor has hundreds of thousands of twitter followers, apparently including Barack Obama. Okay. Who knew. But turns out…remember 205 tweets throughout the year, 73 of them had mentions. Their tweets that have mentions, two X the engagement of their tweets that do not have mentions. Turns out twitter is a very social platform. Most of us don’t get so much engagement on our handles that…
I’ll speak for myself. I can leave notifications on my phone. It doesn’t blow me up. Obviously, bigger brands have different ways to deal with these things. But if you’re a professional social media manager, people mention you, it’s a cause you’re supporting. You’re of course going to try and amplify that content. Leveraging the amplification power of your network or the people that you’re collaborating with or co-marketing with, absolutely is a way to find success. And we see that here for Children’s as well.
So, I’m about out of time for the prepared content that I have, but I do just want to summarize all of that into three quick takeaways for all of you. And then we can spend a little bit of time getting into questions. So, first, you have to know your platforms. You have to know who your audience is just generally but particularly if you have demographic differences between your audience on different platforms. You really have to know and understand those. You have to understand the unique strengths, the ability to do things like mention on platforms and how that will cause behavior change in the people you’re reaching.
And you’ve got to leverage the unique strength of each of those platforms. But more importantly, think back to the Chateau Lake Louise slide right there – good content, what it looked like from a visual kind of curation perspective. And think about how that compared to their least well performing content. Every day is a new day. You have a chance to test campaigns, to test ideas, to test different strategies. “We’re going to go all UGC. We’re going to feature our staff. We’re going to feature our product.” Whatever it is that you’re going to do, you have an opportunity every day to test, to measure, refine, and get to a better place.
Now, if you’re doing this on platforms that have stable or growing engagement rates, great. You’re not even climbing uphill. Facebook, we’ve seen it’s down. It’s still going down. I guarantee when we go and do a refresh and look at early 2018, we’ll continue to see that as they pair back organic reach as well. But if you know it’s flattish…Instagram is flattish…then you’re dealing with one less factor, test, measure, refine. You can do this. We can all do this. I say it about email marketing. I’ll say it about blogging. I’ll say it about anything you do. Everything time you do something else gives you a chance to build a data set and go back and find the best and the worst. Try to do more of the best, a little bit less of the worst.
So, let’s take a little time for questions here. I’m going to go find my question panel here. And see what people have to say. How do I make this thing bigger?
Okay, so through our research, are fashion and luxury brands performing well on Twitter?
I honestly have not memorized the entire study. And so we can go back and look to see what’s going on. I have that slide in here. Let’s see. Let me see if there’s a fast way to do this. I’m good with computers sometimes. There. Okay, so I’m back to one of the early slides that we looked at that was Twitter activity and engagement. And the question was about fashion and luxury brands. So, here are fashion landscape, again the 150 companies that we had selected.
And they are separate from, again, health and beauty, which was like wellness, makeup, etc. They’re actually one of the worst performers on Twitter. And they’re clearly not investing very heavily in it either. So, we’re at well under a Tweet a day. We’re probably at average a tweet every other day on average with the engagement rate also below average – worse than sports teams, only better than media. So, again, on average, we don’t see that these fashion and luxury brands are doing very well on Twitter. For the sake of real time…looking at it…we see that on Instagram, they don’t do much better.
Nor does health and beauty even though…which we know health and beauty brands are very heavily invested in social media marketing as well as in influencer marketing. And so here, there’s a little bit of a question for me. And I don’t have the answer, so I’m not going to be able to do this in real time. But how much of the…how many of the brand impressions and the reach that some of these brands are getting are actually not through their own channels but rather are through some more heavily influencer marketing campaigns that they might be doing? I don’t know the answer to that. Could be studied. I just can’t…I’m not smart, as it turns out. Or capable in terms of doing it in real time. Let’s see. Let’s look…
So, somebody else has a question here about generalizing the benchmarks to my country, which I know has different culture and limitations. So, again, everyone needs to know their…needs to know their audience. And things that we see that are happening at a platform level, it generally doesn’t matter about your country. Generally. I’ll qualify that in a second. But if we know that on average, Facebook is decreasing organic reach for example, of Instagram has a lot more content producers, which because of the algorithm has a factor on reach…
If more people are producing content and more persons like ourselves are following more brands, following more friends, that’s just…but you have a fixed amount of time to spend. You’re going to see a smaller percentage of the things. And the more that these algorithms get feedback from you, the more they’re going to prioritize the content that you typically engage with or watch and de-prioritize that other content. So, if we know that the algorithm is the algorithm generally, then we’re all finding that. It doesn’t matter the country that we’re in.
Of course, different countries obviously have different social media adoption levels on different platforms. There are countries in which Snapchat is really big, and there’s countries in which it is not. Same goes for Instagram. I think Facebook has a got a pretty broad worldwide structure. Obviously, we’re not looking at China at all here, for example. So, there are discrepancies there. But now you’re getting back to questions about everything from how do people want to be hearing from brands, the words that you should use, what’s okay, what’s not okay from a content campaign creativity perspective.
And there’s… Good luck if you’re not having someone on your team that is that person generally or understands that person or understands that target audience. I think we for example concentrate much of our marketing on the United States, more broadly in the English speaking world. But obviously there are strong cultural differences between us here and the United States, and people in Australia, or Canada, or United Kingdom, Ireland, etc. And you have to be careful to not overstep your bounds and to not tread into areas that could be fuzzy.
Let’s see, a couple more questions here. What agencies and advertise… No, so agencies and advertising companies…we actually…they’re explicitly not in that media. This is media in television, print, radio. There are probably a few personalities from that medium as well in there. But generally, it is organizations that product or share news. News or interest stories, etc. We actually had a separate category for agencies, a bit large. Did not include it in this year’s study. But it’s something that we’ve looked at doing. And maybe next year when we… Every year, we try to kind of make the study either deeper, broader, or both. So, when we head into next year, that is certainly I think a thing that we’re thinking about.
So, the next question – any ideas on top posts for higher ed posts with people without…?
So, let’s see. We’ve done a couple webinars on higher ed, mostly because every year…and we’re about to do the next one…we produce an entire study looking at all 355 universities that are part of the NCAA, division one, division two. And we look at some of the top performers on social media. And then we’ll have webinars that are just like this, trying to break down why some of these universities are so good. And I will tell you that the thing that stands out to me as someone who spent way too long in academia is that they have really broad communities.
You’ve got students who are potential, prospective students. You’ve got current undergraduate students. You’ve got current graduate students. Which it turns out are very different from your undergraduate students in many ways and very similar in other ways. You have your alumni. You have your faculty. You have your staff, your professors, etc. So many different sorts of constituencies and audiences. But this is one of the reasons that higher ed does so good on social media is that you’ve got a breadth of audiences that are all very excited. Sports fans…
Heck, you’ve got people that are fans that never even went to your university, haven’t give you a dollar except maybe in football, baseball, volleyball admissions. But they all have an affinity to your school and are looking to engage and be excited about what it is you’re doing. So, things that we’ve seen go well – focusing on your calendar. So, back to school commencement, winter break, just happenings on campus. Those things always do very well. Celebrating community.
These are things that also do really well. So, you have students that are doing amazing things off campus or on campus, whatever. But you’ve got students that are Olympians. You’ve got students that are winning major awards, inventing things, doing research, etc. The same goes for your faculty and staff. Celebrating the things that are happening for your community on social media always is a winner. Always, always, always is a winner. Focusing on the people, as it turns out, as much as I was making fun of Chateau Fairmont with their people posts. Community is what higher ed is all about.
And it turns out that those posts do really well. I think as apart of the some of the links that we send out in the email. We’ve got one or two webinars that cover this topic very specifically as well as I would drive you to our last year’s higher ed report where we do many case studies of things that go really well for higher ed social media as well. Let’s see. One more question in here. Thank you. Mentioned that social media should be involved in the community of our media. For example, being involved in meetings…
I think what they’re asking is…I mentioned something about social media being about involvement and engagement. This is one of the reasons I love Twitter is because it feels to me like the social place for my professional life. Most of my twitter following is social media, digital marketers, SEO’s, etc. And they’re people, and so they’re having people level conversation. They’re having work level conversation. I tend to enjoy that, so that’s where I spent a lot of time actually having back and forth, and commenting, and questioning, etc.
Facebook for all of us could be some of that. But if you’re a brand, that’s going to be the professional connection. And so I think like many people, Twitter is as much of a professional network as is maybe a LinkedIn might be. And so making sure that you’re not just pushing…this is not just like a link farm. You’re not just like, “Go to my content. Go to my content. Go to my content.” You should be taking the opportunity… Twitter is one of the easiest way to reach someone who you might not otherwise be able to reach. And because it’s easy. And if you are consistent, and you’re helpful, and you say the right things, you have a chance to have that conversation in a way that you might never have had before. Let’s see.
Are you finding that posts on Facebook and Instagram that have text on them perform poorly?
So, I do a lot of these kinds of deep dives. Yes. Instagram generally… I’ll talk about Instagram specifically. I find that on Instagram, we… Your traditional B to B content, which is like, “Come look at my white paper. Here’s a pithy quote or an inspirational quote in a box.” It’s not visual, and it’s not compelling. That kind of content just does not do very well on Instagram. It seems to do better on Facebook for the right audience in the right way. So, let’s switch to Facebook for a second. Facebook doesn’t want you to leave Facebook. So, if you spend all your time trying to monetize a customer by driving them off site, that kind of content is not going to get a lot of reach, and it’s probably not going to get a lot of engagement either.
But content that is native to Facebook and has…keeps people’s’ attention, keeps them on the platform, has high engagement, that kind of content is going to get more reach. And so strategically, there’s nothing wrong with trying to send people off Facebook every once in a while. But certainly engaging, keeping them on the platform is what Facebook is trying to drive most. And so there was kind of a mommy blogger that we studied a while ago, and she talked about breaking her content down into three parts – one purely on driving engagement for the content, one on highlighting content that she likes from other…curating other people’s’ content, and then a third kind of going off site.
A lot of her onsite content was inspirational quotes like, “You can do this. You’ve got this,” sort of support for parents and mother specifically. And some combination of the right quotes at the right time, the right look and feel, very sharable content, and a lot of people feel very inspired to share that, which of course drives your reach very heavily. So, that’s words on images that can do well in certain cases. But if you’re thinking about Instagram, and you’re a business particularly on the B to B side maybe try to pair down a little bit in terms of kind of the text-heavy on images. Let’s see. There’s one more question here. Another question. I know that many influencers, brands accounts that want to be famous, gain followers use robots a lot.
They like, post, comment. So, nobody likes robots. I think everyone, even people who are not in this industry, are starting to realize that bots are annoying. Bots noise things up. And the platforms are starting to try to clean things up. They’re not… Twitter did a big cleanup. We know Instagram has always had a strong effort to try to crack down on bots. But it happens. My recommendation, my advice, and certainly what we do on our own social media is we don’t use bots. We try and actually not even use auto responders at all to message DM’s.
They’re just not authentic. And personally, for me and personally for my business, we’re always going to bias towards authentic. And sure, we use listening. We absolutely…in platforms where it’s possible. We want to know when people are talking about us and sharing our content and if it’s appropriate to be able to jump in and be helpful or to say thank you. But yeah, I’m not super into advocating anything about bots. And so how much does video drive traffic and engagement? This is another question here. So, we know that video engagement has tended to be pretty high on Facebook.
Now, for a long time, that’s because native video was getting disproportionate organic reach to the other content types. I believe that they’re starting to ratchet that back a little bit, something we’ve started to study. I don’t have anything to publish yet. But Facebook video has definitely been a big driver of organic reach. Now, Buffer…everyone knows Buffer app, the social media management tool. They did a study, as did Wistia. So, Wistia’s study was really good. It basically looked at offsite performance. So, click performance to their site when they used video as their creative versus an image.
So, like you post a video, and then in your copy, a little message and arrow like, “Here’s my link. Click this link.” What they found for all their video posts was that it didn’t do very well in terms of actually driving clicks to their site. But it did great from a content engagement perspective. And so I forget the exact final ruling, but I think it was something in the vain of if we’re trying to dry people offsite, we’re actually not going to use video. We’re going to use video to drive content engagement on the platform. But if our true mission is to get you offsite, we’re not going to use video as our creative. Specific recommendations for video on social.
Bumpers matter, particularly on Facebook. That first one, two, three seconds…if people are scrolling, and they don’t stop at least with auto play, you’ve lost them. You see people with beautifully produced videos, and they have a seven second lead in, a ten second lead in, a black screen that… Not for Facebook you don’t, not if you want it to do well. So, you’ve got to remember to do social specific edits of your video content to make sure you grab folks. And I’ll say it, but everyone…captions. So many people watch with sound off. It’s a thing. Just deal with it. Do the captions. So many tools. YouTube makes it easy. Facebook makes it easy to do auto-captioning and clean up of those files.
Do you recommend follow/follow back at the first activities?
If that’s your thing, great, I think. I personally don’t do that. There are lots of tools that can let you do that. Of course Twitter is not super happy about automated bot activity. In fact, they’re trying to crack down on that more and more. So, I kind of put this into my own general mantra of I don’t mess with automation bots too much because at the end of the day, if you’re putting out good content, and there’s a reason that people want to follow you, great. And if not, then you’re not really measuring how well you’re doing.
Cool. Well, that’s all the comments, and that’s all the time. I really appreciate everyone coming to hang with us here and for sticking around to the end. I’ve had a lot of fun. I really love getting all the questions, so thank you to all of you. Rachel, and Angela, and Hamid, and Alex for all of the questions that you’re asking. It actually does make it a lot more fun for me. I love doing this anyway. But hey, engagement, interaction, that’s what this is all about.
I hope to see you next week with Lee Odden here on the Data-Driven Marketer. Have a good day, everybody.