Many social media marketers are moving away from metrics that purport to measure ROI. They've realized that social media isn't a transactional engine or sales machine.
Social media is mainly good at building brand presence and deepening relationships with customers. It makes sense to benchmark social media strategies according to what social media is good at doing, rather than using weak metrics to try and get at the dollar impact of a campaign.
The metrics to watch are audience reach, engagement, and sentiment. It's always important to remember that due to Facebook's algorithmic filtering, brand or business posts will only be seen by an average 16% of their fans.
Most basic metrics like number of fans or likes are known as "vanity metrics," because they don't offer much beyond making social media managers feel good about themselves.
Insights, Facebook's built-in analytics tool, offers great basic data for measuring reach and engagement. We show you how to transform those numbers into richer and more valuable metrics.
Shares are particularly valuable, because normal users' posts are seen in a relatively high percentage of friends' news feeds (compared to posts by brand pages); between 29 and 35% according to one study.
Surprisingly few brands and businesses use metrics to perform a competitive analysis of how they stack up against other brands on social media platforms.
Executive Summary: Why Metrics Are Misused And How To Use Them Properly
Growth in social media budgets means there is more interest than ever among marketers on how to track and improve their social media activity, particularly their Facebook campaigns. Survey results indicate that marketers are starting to move away from tracking dollar-value ROI in favor metrics that measure what social media actually does (i.e., building brand awareness, and cultivating an engaged fan base).
Unfortunately, many of the metrics offered by Facebook Insights and other social media analytics platforms are misused, misunderstood, or both. They're often referred to as "vanity metrics" because they don't do much beyond helping social media managers pad out their PowerPoint presentations.
It's important to remember that metrics are indicators. Even the best ones can only tell part of the story, not the whole story. Using complementary indicators to cover weak spots in individual metrics makes it possible to extract actionable insights from a number. In this report, we offer a handful of examples on how to do this, including how to get a better idea of brand awareness from reach indicators.
The best package of social media metrics will analyze activity, impact, and competitive context.
Activity metrics measure what the social media manager is doing. Impact indicators measure results. Competition indicators put activity and impact indicators in context by comparing them to numbers attained by competing brands in the same industry.
In this report, we'll focus on Facebook, but our advice and guidelines can be easily adapted for other social media platforms.
Click here for the PDF version of this report »
Click here for the charts and data associated with this report in Excel »
The End Of Social ROI
Most social media professionals are awash in data. Their high-tech tools produce a fire hose-stream of analytics. But there's a sense in the social media industry that most of the data is dispensable, at least in its raw form.
The frustration is palpable. But trying to measure and single out ROI as the holy grail of social media analytics is not the answer.
The obsession with ROI is still out there. In a survey earlier this year, marketers said that figuring out how to measure ROI ranked third on the list of the questions about social media they needed answered.
But many social media managers are moving to metrics that benchmark social media's primary value — as a powerful audience-building and customer relations platform. They're getting away from half-baked indicators that gauge secondary effects, such as financial return.
Industry participants have begun to realize that it's often wrongheaded to view social media as a direct revenue-driver. Of course there are exceptions. Some social commerce applications and certain types of direct response campaigns will achieve measurable results on Facebook, or other social networks. Facebook chief operating officer, Sheryl Sandberg, mentioned in the last earnings call in July that direct-response spend on Facebook has doubled.
And placing social media in the attribution chain can offer valuable insights. We've discussed how 1-800 Flowers used Facebook mobile app-install ads to drive holiday customer acquisition, and how Caesars, the hotel and casino operator, discovered that its Facebook apps were referring customers to hotel and travel reservations pages.
However, in general, clients are less interested in ROI, says Jessica Jacobs, marketing manager at Unmetric, a social media analytics firm. She adds, "Social serves a lot of purposes beyond just driving transactions, so it's difficult to calculate true ROI."
The 2013 CMO Survey found that ROI-flavored metrics are falling out of favor.
Between 2010 and 2013, the percentage of marketers using a "revenue per customer" metric on social media went from 17% to 9%.
The percentage using a "conversion rate" metric that tracked how many social media visitors became buyers also dropped, from 25% to 21%.
Use of "sales levels" metrics fell from 18% to 9%.
A big argument in favor of ROI metrics was that social media would need to prove with data that it was a revenue-driver if it was ever going to grab a larger share of marketing budgets.
But it turns out that fear was unfounded (and a bit mystifying, since TV marketing budgets never had that burden placed on them).
How do we know that fear was unfounded? Even as the vogue for ROI indicators fades, social media budgets are ballooning. The 2013 CMO Survey found that, on average, top marketers expect to devote 9% of their budgets to social media spend in 2014, and 16% by 2018.
How To Avoid Vanity And Junk Metrics
The end of the ROI obsession doesn't mean that metrics can be thrown out the window. As social media budgets grow, there will actually be more scrutiny on social media campaigns and a need for numbers that prove they're having a real impact. These metrics won't be fixated on ROI, but they will be concerned with audience and engagement, and the quality of fan-brand interactions.
There are an endless number or metrics offered in the crowded world of social media analytics. How to separate the wheat from the chaff?
First, let's look at what to avoid.
Many out-of-the-box metrics provided by social media sites and social analytics platforms are vanity metrics. They're deceptive in that they appear straightforward, easy to understand, and indispensable. But they don't tell the story that they appear to tell.
A great example of a worthless vanity metric is Facebook fans. At first glance, the number of fans a brand page has appears to indicate the brand's audience, but this is misleading for at least two reasons:
People are fans of the ad, but not the brand. Facebook sells paid ads that can drive massive increases in a page’s number of fans, but that doesn't mean that those fans know what they are becoming a fan of. They're clicking to like a particular photo or post, and they may not ever give a second thought to the brand.
More concretely, posts don't show up in every fan's News Feed. Social media managers know this, but it bears repeating because they seem to conveniently forget this fact every time they use fans as a rough indicator of audience. Facebook filters posts based on an algorithm that decides what a fan will find interesting, based on their past social media behavior.
Recently, Facebook announced that, on average, a brand's organic posts (i.e., without paid promotion) only reach about 16% of fans.
In other words, a brand that does a great job of building a modest audience that's highly interested in its content may actually achieve greater effective reach for their posts than a brand with a massive fan base, but no real connection with its audience.
The upshot is that just because someone likes your page does not mean that they see your posts. So Facebook fans doesn't say much by itself.
Realistic Metrics
These days, activity, engagement, reach and competitive analysis are the most important concepts to keep top-of-mind.
Concepts like engagement and reach are too complex to be explained with one metric. Too many social media managers reach for one "killer stat," and then leave it at that. A combination of metrics chasing similar ends, and evaluated alongside one another, are usually what's needed to convey actionable information.
Also, the efficacy of metrics as sources of insights is improved if they're analyzed with a concrete and fairly specific question in mind, rather than expecting the metrics themselves to guide the analysis.
Activity Metrics
Activity Indicators are straightforward and useful. In tandem with other indicators, they can be used to gauge how well a social media manager is performing and what actions are working.
Indicators to track include:
Post rate: This tells you how often you post. You just divide your number of posts by the unit of time you're interested in. So you can track posts-per-week, posts-per-month, etc.
Post rate by type: Same as above, except you only look at certain types of posts. This lets you know how often you use particular types of content, like photos, links, quizzes, etc.
Post rate by topic: This helps you understand the message your brand is sending. You need to classify your posts by topics, and then use the same technique to see the frequency with which you hit certain themes (broad categories are helpful unless you want to lose your mind creating detailed classification systems).
Response rate: This one's different, because your denominator is not a unit of time. You divide the number of audience comments by the number of your responses (over the course of a day, month, or week). This metric tells you how often you are getting back to your fans. You can track your responses to Facebook comments, @tweets, etc.
Impact Metrics — Reach And Engagement
While reach and engagement don't give a dollar value for return on investment, they do paint a picture of whether the right audience is being reached and cultivated, and whether brand awareness is being boosted.
We offer simple ways to enhance the numbers that Facebook Insights generates. The metrics and screenshots we provide are from the opt-in beta version of the Insights interface, which has recently been made available by Facebook.
The beta version of Insights is built to give simple metrics that are easy to understand outside of the context of Facebook, in order to help marketing professionals compare apples to apples when they try to evaluate the effectiveness of their campaigns on Facebook to those they may be conducting outside the network.
For example, "engagement rate" has now replaced "virality" as the main measure of engagement. Virality tracked comments, likes, and shares on posts. Engagement rate, meanwhile, includes all clicks on a post as engagement. If someone clicks on a photo to enlarge it, it's weighed the same as a like.
"The feedback we got from advertisers on the old dashboard was that many of the metrics offered were too Facebook-specific" says Galyn Burke of Facebook, "So the new dashboard uses broader metrics."
Take Post Reach Further
Tracking post reach can help determine whether or not a campaign is worth investing in. For example, producing a $10,000 video may not make sense for a brand if its average reach on a post is only 500 Facebook users.
Found on the Facebook Insights interface, post reach indicates how many users have seen your posts in their news feeds, on the desktop, and on mobile. Total reach is less relevant since it also adds in direct visits to your Facebook page, which is not where audiences typically see your content (they see it in their news feeds).
To get more mileage out of the post reach number, divide it by fans in order to see what percentage of your fans are seeing at least one of your posts.
First, take average post reach for your desired time period. You can do this from the "Post Reach" tab on the "Page" menu. The Insights dashboard allows you to break post reach time-series down to months, weeks, days, and quarters. You then divide post reach for that period by average number of fans for that same time interval.
This will tell you what percentage of your fans, on average, are seeing your content. It serves as an indicator of how effective your messages are.
You can track the reach of individual posts in the same way by clicking on the "Posts" menu on the dashboard.
Remember that the post reach of any single post is typically only 16% of your fan base, though this number can be improved with a good content strategy.
In the perfect scenario, the reach of all your posts as a percentage of fans should be increasing even as you add fans. This would indicate you are recruiting fans to your page who are actually interested in your message, and that you are becoming more effective in sustaining relevant communication with them.
How To Understand Engagement With Facebook Insights
Facebook Insights offers many tools to understand how engaged your audience is. Similar to Post Reach, Facebook also provides its users with an engagement metric.
Engagement is the number of people who clicked on a post. An important note: Facebook uses all clicks on a post to quantify engagement — including shares, likes, comments, and user actions such as enlarging a photo. Engagement is available for all posts, on the "Overview" page, or on a per-post basis.
Engagement doesn't say much without context. To take it further Facebook also provides an engagement rate metric, which is engagement (the number of people who clicked anywhere on one of your posts) as a percentage of reach. Engagement rate is not aggregated as a metric for all your posts, but shown on a post-by-post basis. (See screenshot, below.)
This metric tells you how well your posts are engaging the audience that sees your content.
Engagement rate is useful, but it weights all clicks equally. This is counterintuitive because some users' actions, like comments, seem like they require more engagement from a user than likes.
To drill down further you can look at some complementary indicators. Luckily, Facebook breaks user actions down according to type. Dividing the number of times a certain action was performed on a post by the reach of that post can provide useful indicators for fine-tuning a social campaign.
If brand awareness is a priority, you'll be interested in boosting shares and likes as a percentage of post reach since these actions are more likely to be seen by friends of fans in their news feeds. Shares can be huge, because normal users' posts are seen in a relatively high percentage of friends' news feeds (compared to posts by brand pages), between 29 and 35% according to one study.
If you’re after depth of engagement, look for increases in comments and shares as a percentage of reach because these take the most effort of any action, and shares only result from the most relevant or compelling content.
Competitive Analysis, Tools And Tricks
Jacobs of Unmetric says, "One thing that brands are really interested in is industry comparisons and reports. You can’t compare the performance of Home Depot to McDonald’s because that is comparing apples to oranges and you want to be comparing apples to apples. [Brands] want to be able to figure out a competitor's or peer's content strategy without having to go through post-by-post."
Unfortunately, it is difficult if not impossible to get reach statistics on competitors. And other pages' engagement rate, as calculated by Insights, can't be pulled for competitive analysis either.
Burke says that although they get lots of requests for this metric, Facebook has no plans to provide it.
This makes it difficult to benchmark a strategy against industry rivals.
A quick-and-ready way to sidestep this problem — without an analytics partner — is to check engagement by taking Facebook's public page metric, "People talking about this," and dividing it by the number of your competitors' page fans.
Using this back-of-the-envelope technique, many brands will have engagement rates between 0.5 and 3%, and you can compare your results to theirs, or average results across competitors to get an industry benchmark.
People talking about this is calculated by Facebook insights using likes, comments, and shares as well as less common actions (i.e., confirming attendance at an event). People talking about this is calculated as a seven-day average. So, it's important to remember that the engagement rate is only relevant for the last week.
But there are two metrics that can be used to evaluate performance on a broader scale, going beyond Facebook to encompass other social networks. The best way to get both of these indicators is to use text analysis tools found in many analytics platforms (i.e., Radian6).
The first indicator is share of voice. Share of voice allows you to see what percentage of people talking about brands in your industry are talking about your brand. This is helpful for understanding both how big the chatter about your industry is, and how effective your strategy has been at carving out a share of that buzz.
The second metric, share of conversation, indicates whether your brand is being mentioned in the "right" conversations.
For example, a brand might look at what percentage of "Super Bowl"-related comments it was mentioned in versus the percentage of mentions that other brands got, after buying a TV ad spot during the Super Bowl.
With social media marketing, competition isn't limited to companies producing similar products. Competition can also be for a consumer's attention, in which case all brands focusing on a particular audience could be considered competitors.
A Note On Soft, Sentiment Metrics
The softest metrics out there are those that are used to gauge how an audience feels about a brand.
Many technologies have been developed to gauge how an audience feels. These sentiment analysis tools can be used as alarm bells for PR crises, but because they overlook important conversational elements like sarcasm they shouldn't be used as a replacement for human analysts.
Actually reading and manually recording the mood of comments is the most accurate, albeit resource-intensive way to gauge sentiment.
THE BOTTOM LINE
Social media marketers are moving quickly away from metrics that purport to measure ROI. They've realized that social media isn't a transactional engine or sales machine.
Social media is mainly good at building brand presence and deepening relationships with customers. It makes sense to benchmark social media strategies according to what social media is good at doing, rather than using weak metrics to try and get at the dollar impact of a campaign.
The metrics to watch are audience reach, engagement, and sentiment.
Most basic metrics like number of fans or likes are known as "vanity metrics," because they don't offer much beyond making social media managers feel good about themselves.
Insights, Facebook's built-in analytics tool, offers great basic data for measuring reach and engagement, we show you how to transform those numbers into richer and more valuable metrics.
Surprisingly few brands use metrics to perform a competitive analysis of how they stack up against other brands on social media platforms.
Join the conversation about this story »