In the digital landscape, where search engine optimization and email marketing can be considered members of the old guard, social networks are still a fairly new channel for many brands and marketers. Given the huge potential reach of these platforms, most organizations know that they need a strategy for the social space, but often have difficulty measuring performance and judging success for these campaigns. For ad campaigns on social media, we can typically use some more established & familiar metrics, but when it comes to tracking “organic” social efforts, everything becomes much less defined. How many retweets should I expect to see? Are Facebook Shares of my posts actually generating anything? What is the value of one Pin?
Like any new marketing channel, there is a tendency to rely on big, high-level metrics that are easily understood – for example, Facebook Fans, Twitter followers, etc. While follower counts are important to monitor, I would urge you to look beyond these numbers and dig a bit deeper in order to identify your true key performance indicators (KPIs). While these data points are not in short supply, selecting your KPIs should be a well-considered and focused process. Based on your organization’s goals in social, you should try to identify one or two KPIs that you religiously attempt to improve with every action.
Below are a few examples of common social media goals, along with some common metrics that you can use to measure your efforts towards that goal. Remember that starting with a solid strategic foundation should be your first step; you’ll want to make sure that whatever KPI you tie yourself to, it directly impacts your marketing goals (which should, in turn, directly impact revenue)! I’d also urge you to read Avinash Kaushik’s typically-excellent post on the best social media metrics – while the numbers below are platform-specific, all of them can be grouped into his 4 major buckets of measuring social performance: conversation, amplification, applause, and economic value. A very informative read, and highly recommended.
Now, onto the numbers! Note that in this post, we’re focusing on Twitter and Facebook metrics, specifically, but most social platforms provide very similar metrics.
Goal #1: Increase # of Followers on Twitter
This should never be a goal. NEXT.
Goal #2: Increase reach of content posted on Twitter/Facebook/etc.
This is much better! Follower count, in and of itself, does not directly contribute anything to the brand or goals of the organization. What use is 200K followers if they never bother reading your wonderful posts? A goal like this one can be considered a branding effort, similar to many display campaigns, or traditional mass-media ad buys. In this case, the key is simply raising awareness and getting in front of as many eyeballs as you can.
Common KPIs: Impression-based
Facebook – Post Reach is an easy way to judge your organic visibility, as it simply measures the number of people that your posts were served to. This number here is heavily affected by how much engagement your content drives – we’ll talk about that more in goal #3, but keep these in mind for corollary metrics. You should also monitor Shares of your posts – a user sharing your content to their Facebook friends can be a great way to get in front of a new audience. Also monitor your negative actions – things like Post Hides, Spam Reports, and Unlikes can indicate a disconnect between your content and what the user expected, or potentially an issue with the frequency of posts.
Twitter – tracking raw impressions of your Tweets can be difficult. Many tools define this as “the number of user feeds this tweet appeared in”, but that doesn’t mean that the tweet was actually viewed. Instead, more accurate impression metrics can come from using a third-party link shortener tool such as bit.ly to track clicks on shared links, or an image service such as imgur to see how many times an image was viewed. And, of course, retweets should be closely watched as well – similar to the Facebook Share, increasing the number of retweets is key to building organic visibility on Twitter.
Goal #3: Increase engagement on content posted to social media.
This goal is even better than the previous one! While extending the reach of content shared via social media is ideal for raising visibility and brand awareness, the real power of social networks is the community building element. By driving this interaction with your followers, the visibility of your content can be dramatically extended, you can cross-promote other marketing efforts, and ultimately foster a better connection with your customers.
Common KPIs: Click- and comment-based
Facebook – you have a big incentive to drive engagement on Facebook posts, since highly active status updates are rewarded with better visibility in users’ News Feeds. As such, you’ll want to closely monitor the number of comments, Likes, and Shares that your posts generate. You may find that a calculated metrics, such as “Average Engagement Points” ( [total number of Likes, Shares, comments] / [number of posts] ) is a good way to summarize this data. Don’t forget the power of segmentation here, as well – do your Fans prefer photos, text posts, or videos? What time of day is ideal for driving interaction? Am I doing everything I can do encourage Likes, Shares, and comments?
Denny’s is surprisingly very very good at social media. And at food.
Twitter – retweets remain important for measuring engagement on Twitter, but you’ll want to measure post-level replies, and retweets which specifically add a comment or respond in some way. Measuring the number of tweets that contain your brand (either via #hashtag or @username-mentioned) is similarly important, as is your follow up action (you have prepped your customer service team for issues in social, right?). Tweets heavily marked as “Favorites” should be tracked and monitored – similar content should be planned in the future.
Goal #4: Increase revenue/leads generated by social media traffic.
Now we’re getting into the big boy – I want to see what kind of revenue my social media traffic is generating. A great question, but one that is a little bit harder to answer, and one that requires a bit of upfront work. Since the vast majority of revenue or sign-ups or new subscribers or whatever will actually happen outside of your social media networks, you’ll need to plan accordingly – for many businesses, this means setting up an analytics tool on your website, in order to capture visitor information. This would include traffic source data (for example, how many visitors arrived on my website from a link on Pinterest?) but also conversion data – an action that a website visitor takes that indicates a “successful” visit.
Common KPIs: The ABCs of Analytics – Acquisition, Behavior, Conversions
Acquisition – the first stage of the visitor’s experience on your website/mobile app/service platform, acquisition metrics help determine not only the quantity of traffic you’re receiving from social platforms, but also some important data about visitor frequency & recency. A common question is, “I have X% first time visitors to my website, is that good?” This really depends on your business model and product or service – for content-driven websites, or for products/services with a long buying cycle and lots of research time, you may want to see this somewhat lower, as it indicates that your website/content/product is compelling enough to pull users back in. For simpler, transactional products, you’ll likely want to see a higher percentage of new visitors – knowing that if users don’t convert on the first visit, they probably won’t later, you can focus your energy on driving new prospects instead.
Behavior – in this group of metrics, you’re really focused on how visitors react to your content. A great number here is bounce rate (the percentage of users that viewed only one page and then immediately left), since it’s a great indicator that whatever the person saw on your website, it wasn’t what they expected. I usually like to see this number around 35% – 40%, with some leeway for certain kinds of content (e.g., blogs and customer service articles may have a higher bounce rate, but hopefully it’s because the visitor immediately found what they were looking for!). Average pageviews and session times are helpful as well, but try to think relative here – for example, is a session time of 6 minutes good? Is that better than 4 minutes? I certainly don’t know, but if I notice that users from Reddit spend 3x as long on my blog as the average social media visitor does, I may start looking at that particular channel a little more closely
Conversions – If you ever find yourself looking at a report that does not directly measure outcomes, please stop and lay down for a minute. Conversions, revenue, transactions, outcomes – these are all ways to judge the quality of any given visit to your website, and ultimately boils down to, “was the person able/willing to do what I wanted them to do?” While of course you’ll want to monitor the major conversion points on your site (purchasing a product, signing up for your newsletter, filling out a demo request form), don’t forget that other actions can be great signs of intent as well. These micro-conversion points might be downloading a PDF, adding a product to a wishlist, sharing a product to their Facebook feed, or Pinning a photo. Remember that whenever possible, you’ll want to assign a value to these actions. While this is easy for things like transactions, others require a little more research – for example, if your business typically signs clients to $10,000 deals, and they close 50% of the leads, the value of one lead is $5,000. Don’t worry about getting this absolutely right! Ballpark metrics are better than nothing.
Example Insights and Wrap-up
It bears repeating that whatever metrics and KPIs you ultimately end up selecting as your primary focus, you should be able to directly relate these to organizational success and marketing goals. Tying digital metrics back to broader, C-level numbers is a great way to prove your own value as a marketer, and ultimately helps convince people that analytics is more than just reams of data. To close, let’s take a look at a report straight out of an analytics tool, and see what data we can extract.
(Click to embiggen) The social media channel report from Google Analytics, divided into acquisition, behavior, and conversion metrics.
The Google Analytics report above is taken from the website of a high-end grocer that has a fairly small ecommerce section. The date range is the first half of 2014, and even at this high level there is some interesting data here!
- Users coming from Urbanspoon view almost twice the number of pages as other people, but don’t convert – why? What can we do to capture these people? Are our landing pages ugly?
- Yelp users love you! It is extremely hard to get a bounce rate that low, plus that conversion rate is nicely trouncing the site average. You should make sure you’re doing as much as you can to increase volume here
- And holy cow, look at that bounce rate on Twitter users. What is our social media team doing here? Clearly there is an issue with whatever website content we’re promoting on Twitter, since it’s not effectively driving people further into the site – our conversion rate for Twitter users is nil!
- And finally, the white elephant in the room – that ecommerce performance. The average order value isn’t great (about $50 for social media traffic), but an even bigger factor is the ecommerce conversion rate. We like to see this between 2% and 3% for ecommerce websites.
So! Imagine your instant promotion if you said to your boss, “Ms. Boss, our goal for the first half of 2015 is to monetize social traffic. We propose two primary objectives: raising the average order value 50% to $75, and increasing the ecommerce conversion rate to 1%. Assuming 200,000 visits, by achieving these goals we stand to generate $144,700 in revenue – a 733% increase over last year.”
At this point you may drop the mic and walk off stage.