Social Media Tactics and Strategy: What We Can Learn from Chess and Nonprofit Association Social Media Data
by Steven S. Vrooman
When I spoke on social media at the IAEE #expoexpo, I tried to emphasize strategy. My thinking on this was informed by my love of chess and by the sayings of one of my all-time favorite players:
The problem with an Analytics approach to social media is that it is tactical and reactive. You spew content into the world, measure what works, and then try to make more of it. That's okay, I guess, but it speaks to a lack of vision or mission and a lack of understanding of your audience.
To put it another way, there is a price to pay for outsourcing our social media strategy to your followers. When you use them as live guinea pigs for user testing, by the time you've figured out your message, they've moved on. How can you be sure they will bother coming back if you are not a celebrity? You can't.
In Tartakower's version of things, what he is talking about really is responsive versus generative thinking. If you hang a knight in a chess game or the other player hangs a knight, well, you respond to that. That's the baseline of chess competence. You respond to the needs of the position. There is something to do.
Strategy is what do do when there is nothing obvious. You need to push the position, to create imbalances you can take advantage of, to set the stage for some later tactical fireworks. You plan. You generate vision. Purpose. Meaning.
Social media to me often feels like chess in this way. And there are other similarities. It feels, occasionally, like a competition for eyeballs. You are responding to the shifting nature of the communication landscape and hoping to move first into an opportunity, maybe even to create a new opportunity.
So I am interested in strategy. I will outline some of that thinking in my next blog post, but you can also get a taste of it in my podcast episode of my talk at INBOUND 17:
I'm tired of the blog posts and conference breakout sessions devoted to tactical-only considerations, as if that is the only way to deal with social media. So we run from meeting to meeting or reading to reading, desperately trying to outcompete Facebook's feed algorithms, perhaps hoping to be lone wolf Bobby Fisher against the collective Cold War era Soviet chess machine?
Still, tactics can't be ignored. And I spent some time in some tactical analysis of 10 social media presences of associations attending the IAEE Expo (which is like an association of associations). They remain nameless in this analysis, save for a few examples below of good practice. I figured they'd be okay if someone saw what they were doing right.
The rest of this post will take you through a lot of that tactical material. But you will find sneaky bits of strategy in here, as well.
To start with, here's the coding system I worked out to help you interpret what will be a set of heat tables:
I developed this coding scheme iteratively, and some of it is very idiosyncratic to this particular data set. For example, pdfs are not commonly tweeted, but they were in this selection. There are many other bad practices that would go into the red category, and a few more good practices that would go in the green, but this is a start. I have been developing this general structure of analysis of good and bad social media work over the past few years as I present on social media and develop custom analyses for the groups I speak to. New things come up and surprise, but the things that work generally remain constant, and that is what this chart is based upon.
Let's move through a few key lessons learned from this data.
Here's the overall picture of the 10 associations on the right. All have Facebook and Twitter presences. 6 also have Instagram, as well.
Just looking at this image, which has been coded according to the scheme noted above, you can see at least one thing very clearly: no one is using a consistent strategy across all the platforms. This is interesting for a variety of reasons, but primarily, I get the sense that this was not done strategically or even tactically. The 3rd through 5th associations on the left have a vague pattern of similarity between Facebook and Twitter, but then Instagram is a totally different animal. This is a reflection of how easy it is to automate dual postings on Twitter and Facebook with services like Buffer, but how hard it is to do that on Instagram.
After my IAEE talk, one of the questions in the mass of people who came up to talk afterwards was this question of whether or not they should repeat content across platforms. It's a good question, as we've all probably seen those moments on Facebook when people asked each other to stop cross-posting tweets and then later, Instagrams, onto Facebook, since we all are connected with each other on all platforms and are tired of redundancy.
I had two responses to that. First, looking at the pattern of response when the same content is used across platforms in the sample above reveals no clear difference in terms of effectiveness, which is especially important given the data above, where it looks like the Instagram accounts above are filled with more personal pictures and memes than the other platforms. In the following example, which was par for the course across this data set, you can see that the patterns of interaction are similar across platforms. It looks like the pictures get more traction on Instagram, but in comparing these two posts with the platform averages (the numbers below) for this association's posts, you can see that it is not THAT much better on that platform.
In fact, the pattern of responses here is really quite similar. The dark green bar is a contest put on by the association which was advertised by a child's crayon drawing. The lighter green bar is a picture with a minor celebrity at one of the association's events. Although I coded the child's drawing as more effective with the dark green, in general, in this case the celeb won the day to about the same degree across the platforms.
Thus, my second response to the repetition question, if response structures are the same across platforms, is about that redundancy and fatigue issue. I suggested that if they are posting, say, a picture of the association staff looking at the eclipse outside their headquarters (I saw a lot of these, and I generally like that idea), that a) they should repeat that across the platforms, but b) they should use a slightly different shot for each platform, and c) they should cross advertise platforms in the post contents or comments. For example, "Here's Maria and Kim using their eclipse glasses for safety. We put a picture of Steve on Twitter doing the same thing, but his glasses are upside-down. Shh. Don't tell anyone." This gives value in seeking out the full story across platforms, and it reduces redundancy exhaustion. Let's face it -- if you are taking events pictures, you have more than you will use. You already have the resources to do this, perhaps even at this exact minute on your phone.
When I spoke on social media at the IAEE #expoexpo, I tried to emphasize strategy. My thinking on this was informed by my love of chess and by the sayings of one of my all-time favorite players:
The problem with an Analytics approach to social media is that it is tactical and reactive. You spew content into the world, measure what works, and then try to make more of it. That's okay, I guess, but it speaks to a lack of vision or mission and a lack of understanding of your audience.
To put it another way, there is a price to pay for outsourcing our social media strategy to your followers. When you use them as live guinea pigs for user testing, by the time you've figured out your message, they've moved on. How can you be sure they will bother coming back if you are not a celebrity? You can't.
In Tartakower's version of things, what he is talking about really is responsive versus generative thinking. If you hang a knight in a chess game or the other player hangs a knight, well, you respond to that. That's the baseline of chess competence. You respond to the needs of the position. There is something to do.
Strategy is what do do when there is nothing obvious. You need to push the position, to create imbalances you can take advantage of, to set the stage for some later tactical fireworks. You plan. You generate vision. Purpose. Meaning.
Social media to me often feels like chess in this way. And there are other similarities. It feels, occasionally, like a competition for eyeballs. You are responding to the shifting nature of the communication landscape and hoping to move first into an opportunity, maybe even to create a new opportunity.
So I am interested in strategy. I will outline some of that thinking in my next blog post, but you can also get a taste of it in my podcast episode of my talk at INBOUND 17:
I'm tired of the blog posts and conference breakout sessions devoted to tactical-only considerations, as if that is the only way to deal with social media. So we run from meeting to meeting or reading to reading, desperately trying to outcompete Facebook's feed algorithms, perhaps hoping to be lone wolf Bobby Fisher against the collective Cold War era Soviet chess machine?
Still, tactics can't be ignored. And I spent some time in some tactical analysis of 10 social media presences of associations attending the IAEE Expo (which is like an association of associations). They remain nameless in this analysis, save for a few examples below of good practice. I figured they'd be okay if someone saw what they were doing right.
The rest of this post will take you through a lot of that tactical material. But you will find sneaky bits of strategy in here, as well.
To start with, here's the coding system I worked out to help you interpret what will be a set of heat tables:
I developed this coding scheme iteratively, and some of it is very idiosyncratic to this particular data set. For example, pdfs are not commonly tweeted, but they were in this selection. There are many other bad practices that would go into the red category, and a few more good practices that would go in the green, but this is a start. I have been developing this general structure of analysis of good and bad social media work over the past few years as I present on social media and develop custom analyses for the groups I speak to. New things come up and surprise, but the things that work generally remain constant, and that is what this chart is based upon.
Let's move through a few key lessons learned from this data.
Should I Post the Same Content Across Platforms? Yes (Sorta).
Here's the overall picture of the 10 associations on the right. All have Facebook and Twitter presences. 6 also have Instagram, as well.
Just looking at this image, which has been coded according to the scheme noted above, you can see at least one thing very clearly: no one is using a consistent strategy across all the platforms. This is interesting for a variety of reasons, but primarily, I get the sense that this was not done strategically or even tactically. The 3rd through 5th associations on the left have a vague pattern of similarity between Facebook and Twitter, but then Instagram is a totally different animal. This is a reflection of how easy it is to automate dual postings on Twitter and Facebook with services like Buffer, but how hard it is to do that on Instagram.
After my IAEE talk, one of the questions in the mass of people who came up to talk afterwards was this question of whether or not they should repeat content across platforms. It's a good question, as we've all probably seen those moments on Facebook when people asked each other to stop cross-posting tweets and then later, Instagrams, onto Facebook, since we all are connected with each other on all platforms and are tired of redundancy.
I had two responses to that. First, looking at the pattern of response when the same content is used across platforms in the sample above reveals no clear difference in terms of effectiveness, which is especially important given the data above, where it looks like the Instagram accounts above are filled with more personal pictures and memes than the other platforms. In the following example, which was par for the course across this data set, you can see that the patterns of interaction are similar across platforms. It looks like the pictures get more traction on Instagram, but in comparing these two posts with the platform averages (the numbers below) for this association's posts, you can see that it is not THAT much better on that platform.
In fact, the pattern of responses here is really quite similar. The dark green bar is a contest put on by the association which was advertised by a child's crayon drawing. The lighter green bar is a picture with a minor celebrity at one of the association's events. Although I coded the child's drawing as more effective with the dark green, in general, in this case the celeb won the day to about the same degree across the platforms.
Thus, my second response to the repetition question, if response structures are the same across platforms, is about that redundancy and fatigue issue. I suggested that if they are posting, say, a picture of the association staff looking at the eclipse outside their headquarters (I saw a lot of these, and I generally like that idea), that a) they should repeat that across the platforms, but b) they should use a slightly different shot for each platform, and c) they should cross advertise platforms in the post contents or comments. For example, "Here's Maria and Kim using their eclipse glasses for safety. We put a picture of Steve on Twitter doing the same thing, but his glasses are upside-down. Shh. Don't tell anyone." This gives value in seeking out the full story across platforms, and it reduces redundancy exhaustion. Let's face it -- if you are taking events pictures, you have more than you will use. You already have the resources to do this, perhaps even at this exact minute on your phone.
Should Our Content Always Be "Professional"? No. (Sorta).
This was a key issue. In fact, a lot of people came up afterwards and told me they agreed with what I was arguing but that they'd never convince the C-suite to go along with it. The people in charge wanted it "professional" and no cat pictures.
I hear this a lot.
This is a false dichotomy. I know I wear cat suspenders to these events, but that doesn't mean everything has to be GIFs of cats jumping in boxes or conference schedule PDFs. There is an in-between.
I showed this example in the presentation. I was doing a webinar on game design and play in the classroom and did this Facebook live bit. I also did one later, showing the camera setup and other things in the studio. I told some jokes and had some fun.
Doubled my audience.
Have fun. But have "serious fun"? Don't waste people's time, but, on the other hand, your social media stream should not look like copy from the back of the brochures you have sitting on the endtables in your waiting room, either. This is advertising people are opting into. Reward them for opting in.
Here's how that works. This post crushed it for the IAAPA:
They took a popular meme image, memed it up and then dropped their logo onto it. Here's how the likes (red), shares (orange) and replies (yellow) fared in comparison to their typical results:
Big bars. I think the shape of the graphs for Facebook and Twitter are also interestingly congruent. Yet another piece of evidence that there is less of a magic formula of certain content for certain media than we tend to think.
Another way of thinking about this question of "the professional" vibe is demonstrated below on this table of Instagram likes and replies for another association:
Nope.
People would rather see pictures of regular people from the conference or backstage images (the one dark green line), in the aggregate, than the more polished ones. These conference pics, by the way, for this account, were very very casual. These were not pics of speakers on stage. These were clusters of people talking around the coffee service, etc. Those are the things we want to see.
I want you to put the social back in the social media.
Should I Draw a Line Between My Professional and Personal Accounts? No. (Sorta).
If you assume you need to post frequently and at regular intervals because some other social media guru told you that was best practice, well, the idea that you need to arrange some sort of funny but relevant personal and quasi-behind-the-scenes peek at the organization sounds like a bit of a nightmare to do, especially when you are the one person on marketing/advertising/public relations/social media in your small nonprofit.
But do you really need to post so often?
No.
I know, you can throw a zillion blog posts at me by people who will tell you otherwise. You know . . . . I think the likelihood that someone who is telling you that's how to do social media ALSO happens to have a financial stake in your doing just that is high. They have a subscription/consulting service or an app or a template or a book or a keynote to sell. Look, I get it, so do I. There is no way in an all-marketing world to avoid this kind of problem, but I am skeptical of an approach that tells us best practice is beyond our reach to do it alone. I think we should use extra scrutiny in such cases.
What if, instead, we built social media content for a different, and more durable and strategic purpose? If you weren't desperately trying to push quantity out of a vaguely understood sense of how this puts you into people's feeds, how much would you post? A lot less. If you only posted when you had something good to share, well, how often does that happen?
Relax.
Below I will explain why that is okay. For now, if you ultimately believe it is okay to do less to get better results, isn't that relaxing on its face? And doesn't that take some of the sting out of time stresses of getting more personal?
I am telling you you can do it alone. In much less time. I know lots of social media professionals that do. Here's what that looks like:
Let's take these in order.
Notice? Look, you can't beat the algorithms. I always get versions of this meme looking back at me when I say this to a room of professionals who have been trying to do just that for years (and who may have had some moments of viral success):
Everyone I know whose job it has been these past 8 or 9 years to beat the system has crashed into diminishing returns. Facebook is a rich FAANG company because it hires lots of people to beat you at this game. They profit when you fail to program your content to show up on a lot of screens without resorting to paid ads. So they will beat you. Now they own Instagram. They will beat you there, too. Too many hashtags to do what used to work to get views? Shadow banned. Twitter is a bit more wild west, but in the firehose of Twitter data, do you really think the old Guy Kawasaki tip to repeat each of your posts 9 times is really going to make that big a difference in getting new eyeballs before it alienates your quality followers, your "whales" (thanks, Seth Godin), who put you into a curated list and will lose patience with your shenanigans? I just dropped 8 people from my own main Twitter list for just such behavior, even though I usually liked their stuff. It crowded everything else out.
I had this conversation with about a dozen people at the IAEE Expo and asked how long their patience extends for such things. The first thing to know is that they needed no explanation here. They knew EXACTLY what this kind of posting looked like. The second thing to know is that, although I suggested my patience wore thin and I unfollowed in a few weeks, not one of these dozen gave me more than an hour or two before they unfollowed.
Back to basics, eh? Do no harm! Don't alienate people.
Thus, make "destination pages." Make your social media stream "sticky," to use the words of web designers. When I finally get to it, I will stick with it. Don't make social media posts which advertise themselves. Make posts which drive people back to your collected stream of posts. A scroll-by liker doesn't do you much good. You want someone to go and find your Instagram and look at all of your stuff. And stay. And comment. And tell someone else. Quality of interaction over quantity of epiphenomenal browsing is what we want. Here are some examples of social media feeds that are destinations: this and this and this and check out the San Antonio Museum of Art on Snapchat.
Engage? That's what we've already done. We've done some simple analytics designed to show us patterns of what people engage with and what people don't. The other thing I spend time on in these keynotes is that there is no such thing as quality content. We have shifting needs and are generally bad at processing information. We want to build a relationship on social media so that we don't feel like total waste-of-time failures when we are scrolling through our feeds on our phones. So there is no way to know whether you've made "good stuff" until you find out, tactically, and then build a strategy for deepening those connections with your followers.
Of course, maybe you always had a vision, a strategy, and you just kept working away at building an organic following. So now, when you do analytics, you are not surprised. You know them well enough to know what will work. There's a good metric for you: The quality of social media strategy is inversely related to the number of surprises in your analytics data.
Share? Here's the key to this whole section.
Make every post shareable and ad-worthy. In other words, make each post awesome enough (based on what you found out above) that people will want to share it. Maybe you strike viral gold and you feel good about your views and connections. Maybe not. But then, once you have some social proof (likes and shares are not sad 0s staring at us, telling us the content is not worth our attention), you can then make that content into a paid ad. Most platforms, including LinkedIn, push you to build ads based on posts you promote. You don't want to promote content with a small number of likes and hope people will add likes via their exposure to the ad. Yeah, right.
If you can't beat the algorithms you will need to turn to ads. Your ad strategy should be built out of what works organically with your fans in this way.
In case your eyes are blurring with this longish blog post, here's the general set of of things years of analysis in various audience and industry contexts has shown tend to work. You can hold onto this at the end of this post and feel like a winner even if you spaced out a bit at the end there. In my next post I will detail this list with examples of what this looks like and by connecting it to a larger strategy discussion.
Good luck out there!
