Social Marketing ROI through Analytics
You’ve put your social media ad strategy into action, you are seeing activity and the interns are typing like mad and people are talking about your brand. At least you think they are. Return on Investment (ROI) measurement on social media advertising is still a young science and you believe that you’re doing everything right to track influence, engagement, actions and [insert newest industry buzzword here]. But, so much has changed in the last decade of marketing that even the definition of ROI now means, “Return on Influence”.
Your marketing budget depends on both definitions of ROI, so crack those knuckles and let’s talk analytics.
Tracking social media results is no walk in the park, unless your park is Fenway and you’re still celebrating the World Series. Benchmarks and metrics will vary depending on your business model; however there are some high-level standard measurements you can look at, regardless of the type of business. When it comes to social media advertising, the biggest factors for any brand are return on brand perception (awareness) and lead generation (sales conversions).
When it comes to tracking leads and conversions related to your social media campaign, the tools are in place to make this process fairly pain-free. Whether your outreach is on Facebook, Twitter or email marketing, the trick is to set up everything correctly in the planning stages. Tracking URLs and dashboards are your best friends in this situation. They’ll be driving your reportable numbers, down to every single click. Organize your campaigns, code the links and your analytics team should have little problem correlating your social media efforts to direct sales or leads.
Tracking lead generation through clicks can pinpoint your audience. You can see where the most sales originate and target future campaigns there. But before you get that click, how do you track how your brand is being perceived in the marketplace? How can you track the people not clicking through to convert to a sale, but still talking about your brand? That’s where things get a whole lot more interesting.
Quantifying Likes, Shares and Retweets
Leads are just one piece of the data you need to put together a comprehensive reporting package. While you can easily use analytic platforms to report on URLs included in your campaigns, you’ll need to use a different tool to gather things like shares, likes and other social media actions. There are many tools on the market to help with this, or you can gather them yourself each social network at a time. Either way, once you have this data, how can you possibly know how your brand is doing in the marketplace through shares and likes? Are these things really indicators of brand influence and perception? Of course they are, you just need to know where to look. Consider the following metrics:
- Churn rate: followers lost divided by total followers at beginning of any given measurement period.
- Change rate (posts): churn rate bounced against the number of posts you are creating in a given measurement period.
- Change rate (audience): change in audience size based on number and/or content of posts in a given measurement period.
- Growth rate: tracking increase/decrease in followers at a particular starting point.
Math Meets Marketing
To determine the value of your social media activity, let’s apply some algebra. Basically, you want to find the value of every social media follower (for the sake of brevity, let’s just call them all “followers” instead of network-specific designations). There are current formulas out there to calculate the value of a follow or like, but they’re missing the value of a share calculated into that value. Often you’ll see more click-throughs on a social media post than your average percentage of follower click-through. This is because of sharing, which has to be factored into the calculation.
So a modified formula to calculate the daily value of a follower looks like this:
- X = number of total followers / unfollowers
- P = Posts per day
- Y = average click-throughs per post / followers
- Z = conversion rate * average conversion value
- D = Dunbar’s effect
- (X*D)*(P)*(Y*Z) = value of a follower
The Dunbar Effect
You might notice something in there called Dunbar’s effect.
Dunbar’s number is a cognitive theory established by Robin Dunbar, a British anthropologist and University of Oxford professor. He suggests that we can only maintain roughly 150 “friendships” in social relationships. That is, the number of people we keep some sort of social contact with but does not include, for example, childhood friends we no longer have any contact with or people we might know casually but don’t maintain a social relationship with. These caveats might produce a number much bigger than 150, and depends on our long-term memory.
Dunbar’s effect is based on Dunbar’s number.
Most studies have found that we really only hold quality relationships with about 3% of that number. So when we share on social networks, the psychological theory is that 3% of our network at any given time truly cares about what we have to say. This is Dunbar’s Effect. Consider your social media activity and compare your average likes on a post to the number of followers you have. Simple math can prove that theory. This is a constantly shifting 3%, meaning it is not always the same group of followers at any time, which makes it interesting to campaign design, but virtually impossible to actually measure.
Frankly, it’s odd to assume that when advertising on a social network, no one’s sharing. Of course they’re sharing, that’s why they’re there. So the industry-standard formulas many marketing firms use don’t appear to take such sharing into account. While they capture follows and unfollows, that’s not the only thing happening on social networks and it’s highly possible that most estimates of social media ROI have been grossly understated. Facebook, to its credit, does have campaign management tools that take indirect engagement into account (or that you can use to calculate such).
Applying the Formula
The assumption that psychology shouldn’t play a part in determining social media advertising ROI should fall into one of those assumption clichés. Social media is not a radio ad. The human element should always be part of the equation.
So to wit; let’s pretend you have 100 followers (you probably have a thousand times that, you superstar). You’ve had 10 people unfollow you today. You posted five times today, with an average of 15 click-throughs to your site per post. You tracked the URLs locked to this particular social media campaign and 25% of those converted into sales worth $10 each (either value of product, service or lead). Dunbar’s effect assumes that each follower shared that post with their network and their current five best buddies paid attention (out of the aforementioned 150 best buddies).
So the formula looks like this:
- ((100/10)*5)*5*(15/100)*(0.25*10) = 93.75
So now that you’ve determined the value of your follower and the presumption of their sharing activity and how that translates to sales, it’s time to take a look at what they’re actually saying about you. You can instantly chalk up converted sales to the “positive” category when it comes to your brand. You can chalk up the unfollows to the “negative” category, as well. That’s two markers out of the way. So let’s do some more math, because math is fun: you had a 25% conversion rate on the campaign today (all these calculations are based on one day’s activity).
So the formula for a base positive approval rating looks like this:
- X = approval rating
- Y = conversion rate
- Z = followers
- U = unfollowers
- P = total clickthroughs
- X = (((Z-U)+P)*Y)/Z
- or 41.25% = (((100-10)+75)*.25)/100
Now you know that 41.25% liked you enough to click through. Some of them even bought something. You know 10% can’t stand you, they jumped ship earlier. So what about the remaining 48.75%? Where do they stand? Today, on this campaign, they stand indifferent. There are (tedious) methods to track influence through targeted searches of your follower base, and tools to categorize those searches to actually read what people are saying, but tomorrow that might all change. Social media is a fickle beast, what people are saying today is not always what they are saying tomorrow.
These formulas will give you a starting point for evaluating what’s working and what’s not. You can see the effect of your campaigns, what followers react positively to, and what they don’t like. This will help your brand tailor social media campaigns to audience reaction and their propensity for sharing, instead of relying on a system of randomly seeing what sticks to the wall. It all sticks to the wall; you just need to know where and when to throw it.