Reach, Audience and Impressions - on Twitter and Beyond
I've already written about how I feel about engagement rate and how I think it should be measured. Thankfully there are tools out there that already calculate it correctly, such as Simply Measured, while other tools have updated their platforms to adapt to the change, like SocialBakers.
However, things are not that simple when you look at Twitter. When it comes to analytics, Twitter isn't that friendly. In fact, the official Twitter Analytics dashboard leaves a lot to be desired, and it's a shame considering that social analysts and marketers had been waiting for years for a good dashnoard from the social media giant.
This leaves us with a few questions, one of which is: how do we calculate reach, audience and impressions on Twitter?
While worded differently, reach is the same as audience. However, there is a difference between reach and impressions.
- Reach: number of people who may have seen your content
- Impressions: number of times the reached people may have seen your content
This doesn't only apply to Twitter, but other social networks, as well as websites (those same terms are used in web analytics).
A Quick Look at Web Analytics
The same concept applies to web analytics as well, where your reach/audience is the number of visitors, while the number of impressions is the number of pageviews. By the same logic, the number of impressions is always equal to or more than the number of visitors.
A Quick Note on "may"
You may have noticed the may as part of the definitions for reach and impressions. This is because the accuracy of how certain platforms detect the people reached and their relative impressions isn't 100% accurate.
Let's take Facebook for example: a user is marked as reached when they scroll to (or past) a post, as well as when they open a post (e.g. clicking on a post URL). However, scrolling past a piece of content doesn't necessarily mean that I've actually seen it. Hence, reach is an accurate way of saying how many screens the post was displayed on (and impressions tells how many times they've made such an appearance), while it doesn't always mean that a user has viewed that post.
The Problem with Reach
Calculating reach is easier on some platforms than others. Facebook easily puts it in its Facebook Insights, while Twitter doesn't provide that metric at all. Getting the reach on Twitter is apparently quite difficult, hence why there are only a handful of tools out there that provide some sort of calculation.
When reach is not easily available, marketers resort to 3rd party tools to calculate that for them. Unfortunately, not all of these tools calculate it in the same way.
There are currently three main ways in which the reach of a tweet is calculated:
- Overcalculation: this is the method used by most tools out there, especially free ones that decide to take "reach" on board (e.g. one of marketers' favourite tools, Tweetreach). This method calculates reach as the sum of all your followers, the people who have interacted with your post and their followers. This revolves around the assumption that all your followers and people who interacted with you (along with their followers) see your posts. While we know that to be statistically wrong, it's the most common calculation used by a lot of tools out there. This is then taken further:
- Absolute overcalculation: this method calculates reach as the sum of all your followers, the people who have interacted with you in any way, and their followers. This is regardless of interaction - be it a reply, a retweet or a favourite.
- Selective over-calculation: this method calculates reach as the sum of all your followers, the people who have amplified the reach of your tweet by retweeting it, and their followers. Hence, if someone replied to your tweet, or favourites it, it wouldn't be counted in, but only Retweets would. This is used because a mention or favourite by a user doesn't put your tweet in his/her followers' timelines.
- Approximate calculation: this method calculates reach in a way that I still haven't seen a lot of brands adopting, but it's still worth knowing. This method lets you set a percentage of reach, so if you believe that 15% of an average Tweeter's followerbase sees your tweets in their timeline, you can apply 15% to the number of followers who have interacted with you, as well as your own followers.
- Adaptive approximate calculation: This can be further tweaked to accommodate paid campaigns, which have a greater reach than organic posts. Hence, you could have a fixed percentage for organic posts (e.g. 20%), and another fixed percentage for paid posts (40%).
- Under-calculation: this is a drastic calculation, in which you calculate the number of unique users who have interacted with your tweets (so if User A replies to and favourites your tweet, by this method you will have reached one person).
Which is the best method?
Which one is the best method out of the 3? They're all an approximate calculation for reach, not an exact one. Hence, none of those methods can give you a perfectly accurate figure for your reach. Here's why:
- Over-calculation allows for a big chunk of people to go through the "reach funnel", making a big assumption that all of your followers and people who have interacted with you as well as their followers make up your reach.
- Approximate calculation is what I'd call a "static calculation" - it doesn't take into consideration the unpredictable virality of social media posts. For instance, this approximate calculation doesn't take into consideration the fact that hashtags can be viewed by anyone and as such tweets with a trending hashtag have a much bigger reach than tweets with no hashtags.
- Under-calculation is a very strict way of calculating reach, it's the "safe way" that makes no assumptions. This says that reach is the number of people who have interacted with your post. While it's true that people who have interacted with your post are part of your reach, reach isn't limited to just them.
So, Which Method Should I Use?
Whether you rely on free tools or premium tools, you need to be aware of how your reach is calculated, as this can dramatically affect the way you calculate reach-dependent metrics, such as engagement rate. For instance, if you choose to over-calculate your reach, you'll end up with a low engagement rate, while under-calculating your reach means you'll get a higher engagement rate yet a lower volume of reached users.
So, which method should you choose? Before we can answer this question, we need to change how we refer to reach and impressions for these platforms. Some social platforms do a very good job at providing some sort of reach (Facebook), others do a stellar job at it (YouTube), while others don't (Twitter). So, to answer your question, we first need to change how we refer to such metrics for certain platforms:
- The "potential reach" is the number of people who may have seen your post.
- The "potential impressions" are the number of times the post was shown to the potential reach.
- The "true reach" is the exact number of people who have seen your post, not based on any assumptions, or a predefined percentage.
- The "true impressions" is the exact number of times your true reach has seen your content.
Defining the metrics in this way regulates the expectations we can have towards a social Analytics tool or a social platform.
Now, for platforms that do not provide you with their true reach, you can choose any method you're comfortable with out of the three we've listed. However, what you get from the three method is only the "potential reach" and "potential impressions".
For that, more often than not you'll need a tool. However, before signing up or buying a tool, ask how the reach is calculated, make it one of your specifications for the ideal Social Analytics tool for you.
Choose a tool that gives you flexibility in calculating reach, one that calculates reach in a way you're comfortable reporting on. A great tool that does just that is Simply Measured. Simply Measured supports all three main methods of calculating reach, giving you total flexibility in how to calculate your reach on Twitter, as well as giving you accurate figures for reach for other social networks that openly provide reach figures (e.g. Facebook, YouTube).
If you're still unsure and want to know how the tool you have in mind (or the tool you're already using) stacks up against Simply Measured or some other tool, check out G2Crowd - it's a great comparison site for corporate tools (be it for marketing, analytics, management etc.). Click here to register for free (all you'll need is just a LinkedIn account et voilà, you're in!)
True Reach for Everyone?
So, will we ever see "true reach" available to everyone for every social platform? In a perfect, utopic world, yes. This depends on the actual platforms and what they make available. Due to the complexity of some platforms, some of them will find a hard time calculating reach.
Until the platforms allow for this to be calculated accurately, True Reach remains one of the Holy Grails of Social Analytics.
Update: 8/04/15 - Twitter Analytics 2.0 now available, and so are (True) Impressions: brnrd.me/twitter-analytics-2-0