Often we find that clients know exactly what they want to achieve on social media, and that’s the way it should be. But what advertisers so often struggle to grasp is how to measure social media performance. They receive regular reports filled with numbers, charts and graphs. But what good is gluggy data if you have no benchmark and no guidelines as to what’s important and what’s not? In this article, we’ve listed the most common social media objectives that clients come to us with, and we’ve broken down what metrics you should focus on in order to measure performance.
These metrics will allow you to strategically measure the success of social media activity, and use data to form insights to optimise campaigns, ongoing. The success of a campaign can’t always be measured with one metric alone – you need to look at the big picture. But these guidelines will help you prioritise data, and make your own comparisons between time periods, channels, audiences and content.
Set up a social media measurement model
Unfortunately, there’s no one button you can press to magically set up the infrastructure required for such measurement (one day, perhaps). Reports will be unique to each advertiser based on your objectives. With that in mind, here are a few factors to consider based on the objectives outlined later in this article.
A UTM code is a simple code that you can attach to your campaign’s landing page URL in order to track a source, medium, campaign name, ad set name and even creative label. This enables Google Analytics to tell you where website visitors came from and how they acted once they got there. To generate UTM codes, you’ll need a UTM Code builder. Google provides a good option here. If you live in Facebook land, there’s a UTM builder built into Ads Manager too, which you can find here.
Google Analytics Goals and Events
For any campaign objectives that drive traffic to your website, you’ll need to set up Google Analytics. That doesn’t just mean installing it and walking away. You’ll also need to define events and set up goals based on the metrics you’re going to measure. For example, to measure Scroll Rate, Add to Cart, or Purchase, this will require you to place a piece of code on certain pages of your website. Although using Google Tag Manager simplifies this process so you don’t need to be a developer to set up events, it can still be a little difficult to navigate to the layman. We recommend talking to a professional about setting up your Google Analytics so that you can be sure it’s robust and so you don’t break the internet.
Here’s a myth-buster: clients and advertisers often assume that a pixel is placed on a website purely for measurement purposes. But all marketing platform pixels – Facebook, Twitter, LinkedIn and Google – serve another, all-important function. In order to optimise for conversions in your campaigns or build remarketing audiences, you must have the pixel placed on your website, with Event Tracking for your conversion goals. Please, just go ahead and place the pixels on your website, trust us on this one.
App Install Tracking
If your objective involves app actions, and you’re using Facebook, you’ll need to set up tracking via Facebook SDKs for iOS and Android. Here, you’ll find predefined events such as ‘install’ or ‘add to cart’. For more, talk to your developer or check out information about Facebook app tracking here and Google’s equivalent here.
Foot Traffic and Retail Sales Tracking
Although there’s no digital set up you can implement to track foot traffic or retail sales (yet), it’s good to define your measurement model prior to campaign set up. Know what sales or foot traffic numbers you had in the previous period, and list all external variables that could affect an incline or decline.
Build your reporting template
And finally, for best results, launch your campaign knowing what metrics you are measuring before you start. This will allow you to monitor performance and adjust your strategy in real-time, rather than waiting until it’s over, and it’s too late.
Okay, without further ado, let’s get into the good stuff…
Primary metric: Cost per Lead
A database of email addresses and audience information is an invaluable tool for a marketer to use, both for email marketing, and for social media targeting, for more on that see here. EDM (Electronic Direct Mail) marketing is an economical channel, particularly effective towards the end of the funnel, and particularly for Ecommerce.
To measure the success of a database or lead generation campaign, Cost per Lead is the key metric. When optimising campaigns and adjusting strategy, you should also consider Lead Conversion Rate. On Facebook, that’s leads divided by reach. On other platforms, it will be leads divided by impressions. This will give you an indication of whether the people you’re reaching are converting to leads. And finally, check the quality of leads acquired by running an audience demographic report: are they part of your target audience?
Primary metric: Cost per Eyeball or Cost per Estimated Ad Recall Lift
Put simply, brand awareness is about familiarising your audience with your brand – that’s straightforward. What’s not always straightforward is measuring your audience’s familiarisation; how are we supposed to know who remembers your ad?
Social media’s unique value when it comes to brand awareness is in its measurement. A billboard may provide great awareness, but measurement is left to assumptions.
Facebook’s unique Brand Recall Lift metric allows advertisers to measure how different audiences are likely to recall your ads. By applying your spend to this, you’ll get a Cost per Estimated Ad Recall. However, this metric is only available by using the platform’s Brand Awareness Objective, which is limiting in many ways, particularly if you don’t have tens of thousands of dollars to spend.
A more traditional means of brand awareness measurement is the old-fashioned Eyeball. It’s not a particularly technical term, but ‘Eyeball’ is our all-encompassing word for the human gaze on your brand’s content: Impressions (the number of times your ad is shown) or Video Views to 25%, dependent on the format of your campaign. The metric for tracking Impressions is CPM – this refers to Cost per 1000 Impressions (if you were to put a cost to one single impression, it would look something like $0.00001, so we make it per thousand to avoid those clumsy-looking numbers). Cost per Video View refers to the number of times your audience started your video (not necessarily completed the whole thing).
And finally, as with any most objectives, check the quality of those reached by running an audience demographic report.
Primary metric: Cost per Engagement or Engagement Rate
Not to be confused with Brand Awareness, Brand Preference is – as the name suggests – about driving your audience to prefer your brand over competitors’. In other words, the importance of brand awareness is its measurement of how well you’re known, whereas brand preference is the impact of that awareness.
To evaluate Brand Preference, measure Cost per Engagement (Mosh’s measurement of engagement includes likes, comments, shares and video views – not clicks, that’s cheating). And at what rate are people engaging? To calculate Engagement Rate, take Total Engagements and divide by Impressions (or Reach on Facebook).
Over time, is your Cost per Engagement decreasing? Is your Engagement Rate increasing? Although they’re not completely watertight, these measurements are good indicators of the impact of your brand.
Product or Service Education
Primary metric: Retention
If you’re looking to educate your audience about your product or service, you’ll either do it via video on a social platform, or you’ll host this information on your website. Either way, the objective here isn’t to sell a product, it’s to educate your audience in order to better understand the product. And so, you’re aiming to retain the users’ attention for a period of time. Our umbrella term for the various applicable metrics is ‘Retention’.
If the educational resource is on your website, consider setting up a Google Analytics goal that’s triggered when a user spends two minutes or more on the page, or scrolled 75% of the page. Or if the educational piece is a video on your page, consider the primary metric 90% video views (users often leave before the video is 100% complete, so 90% tends to be a good indication of Retention).
Foot Traffic and Retail Sales
Primary metric: Lift in Foot Traffic/ In-store Redemptions
It seems an impossible task: measuring how a social media campaign affects in-store sales or foot traffic. But here are a few ideas.
Perhaps most impressively, Facebook has released a Store Visits metric. According to Facebook for Business, it’s calculated using a range of measurement signals, including “Information shared by people who have chosen to turn on Location Services on their mobile devices; Satellite imagery and mapping data from third parties that show the location boundaries of your stores; and filtering out people who may be employees or appear to be moving past your stores rather than visiting.”
But this metric is only available to a handful of advertisers, most likely big spenders in the US, but if it’s successful, it’s likely to become available more widely, so watch this space.
In the meantime, there are other means of measurement to try. The first, measure Foot Traffic and/or Retail Sales over the campaign period, and compare this data to the previous period. Of course, other variables will come in to play here, hence our caveat: this metric isn’t water-tight.
Another means of tracking Retail Sales is to use In-store Redemptions. For example: in your social media ads, use the sentiment: “Show this in-store and receive a discount”. Again, it’s not watertight and there’s a bit of a barrier to entry, but it’s better than no measurement at all. A note on this method: use this sentiment alongside other creative so that you can see, objectively, the effectiveness of this specific avenue.
Primary metric: Cost per Page Like/ Follower
Although these days organic reach on social platforms is harder to achieve, we do still find value in building Followers and Page Likes.
Organic reach started declining in 2014, as Facebook shifted its focus to paying advertisers. With the decline in value of organic reach came the decline in value of Page Likes.
However, an online community isn’t just about organic reach. Perhaps even more valuable, these days, is the credibility associated with a strong social media presence. It’s true what the questionable 90s rock band Soul Asylum says: “nothing attracts a crowd like a crowd”.
The best way to measure community growth is using Cost per Follower/ Page Like and audience demographics to gauge the quality of those followers. Another useful metric is to analyse the change in Engagement Rate over time. How is this changing month-on-month or week-on-week? Has the increase in new followers resulted in a decrease in engagement? If so, all those new people just aren’t that into you.
If your marketing objective is Engagement, there’s something wrong. Chances are that your overall objective is actually to generate sales, build a database or grow your online community. In this sense, Engagement is a metric used to indicate the success of your true objective. Without a more significant goal, engagement doesn’t mean much at all. Yes, this was a trick, if you passed, well done.
Primary metric: ROAS (Return on Ad Spend) or ROMI (Return on Marketing Investment)
Perhaps the most direct line of measurement available online is measuring online sales. What’s unique about Ecommerce or conversion campaigns is there’s no guesswork: every transaction and even the value of that transaction is traceable if tracking is set up correctly.
Although different brands will have different models of sales measurement, the most effective metric to use to gauge performance is Return on Marketing Investment (ROMI). However, this metric takes a little time to calculate as it requires you to know the total Marketing Investment for that given campaign. Consider ad spend, agency management fees and creation of assets. Next, take total transaction value generated, and divide it by marketing investment and times it by 100.
For example, if your total Transaction Value was $10,000 and your total Marketing Investment was $2,000, your ROMI would be 500%.
If you don’t have access to Marketing Investment data, the next best metric is Return on Ad Spend. The same calculation can be applied, which will give you a good indication of the success of the campaign.
For the advanced class who want to attribute conversions across their funnels, read up on attribution models, We might need to do another blog post on them, let us know in the comments if that’s something you’re interested in.
Primary metric: Cost per Lead
For big-ticket purchases, you can’t expect a user to buy online (think cars, electronics, furniture, houses – who buys a house sight unseen?). Instead, you’ll want to open up a line of communication that allows you to slowly but surely build a rapport with a potential customer. Remember there’s a difference between sales leads and database leads: sales leads is about starting a person-to-person conversation.
Cost per Lead is a good measure of Sales Leads campaign performance, and Lead Conversion Rate will also give you an indication of how various creative and targeting variations are performing: to calculate Lead Conversion Rate, take total Leads and divide by total Impressions. This will give you a good indication of performance both during and after the campaign period.
Finally, monitor the quality of leads acquired by using audience demographics, but also by manually monitoring the value of the leads – are they converting into sales?
Primary metric: Cost per Response
Market research comes in so many forms. And there are few more effective means of distribution than social media.
The most common form of qualitative market research conducted on social media comes in the form of a survey. If hosted on a website, the Cost per Response can be calculated in the same way as a Cost per Sales Lead. If you intend to run an informal survey via Comments on a social media post, the metric you’ll want to gauge is Cost per Comment.
Aside from qualitative research, social media is effective quantitatively, too. It allows us to A/B test various messages to various audiences and track the performance of such tests. For example, you might test an ad with a green button and an ad with a red button amongst the same audience. Or you might test the same content to different audiences and track performance. These tests, if run well, have the potential to provide valuable market insights for product development and broader marketing strategy. Of course, there’s no one metric that will give you the answer to every A/B test, but feel free to get in touch if you need some more guidance on this model.
Primary metric: Cost per Response or Cost per Attendee
We all know, possibly from personal experience, that when someone clicks ‘attending’ to a Facebook event it doesn’t mean they’re going to attend. But no matter how unreliable your friends are, there’s no denying Facebook’s event functionality has served as an exceptional tool for event planning, and it’s likely one of Facebook’s features that will take it into the future.
When measuring Event campaign performance, the obvious metric to use is Cost per Attendee. In order to get more reliable, marketers could introduce a ‘how did you find out about this event’ survey upon check-in (if it’s not weird and you do actually have a check-in).
Primary metric: Cost per App Install
App Install campaigns have a very clear objective: to generate App Installs. The most obvious metric to use for such campaigns is Cost per App Download. Note that setting up tracking for App Installs requires some mobile-specific infrastructure. You can read about Google Analytics SDK here and Facebook SDK here. To track the quality of App Installs, use retention metrics – uninstalls and average time in-app – to gauge the user’s interaction.
Conclusion: social media measurement matters
If you’ve made it this far, well done! We hope this list of objectives and metrics has given you what you’re looking for. All-in-all, we’re living in a data-driven world, and we’re working in a data-saturated sphere. With the right knowledge and just a little know-how, you can draw your own meaningful insights from the often-gluggy reports you receive. Oh, and if you didn’t find what you’re looking for here, just get in touch.