I’ve recently been required to focus more attention on social media from a federal agency standpoint and this has directly led to a greater consideration of content. One of my first steps was to begin sharing various forms of content and gauge the success of each type. In today’s post, I’ll share what I have learned and hope it opens your eyes to how we measure success and whether our metrics are right or completely meaningless.
The Fading Relevance of the Click
While certain forms of content still use clicks as a measure of success (mainly advertisements), the days of measuring the effectiveness of most content by clicks alone is long past. As the Web grew more sophisticated (and commercialized) more detail was needed. Time spent on a piece of content was one factor that was used to try and determine its effectiveness. And since I have used that word twice, it is worth mentioning that how effective your content is depends on your goals. How effective at doing what? Converting people to click a link (ok, they’re still sort of important), watch a video or just read a piece of content?
Many times in the public sector our goals are much less quantitative. We aren’t measuring sales using a customer’s actions on our site with monetary goals. We aren’t looking for conversions in the common sense of the word. We may, however, be interested in Click Through Rates (CTRs) when purchasing services from a vendor such as sponsored content or job postings. But we still have a slightly more nebulous task when asked to measure the effectiveness of the content we are producing, and the advent of social media hasn’t made our lives any easier.
More Complex Interactions
As social began to be the dominant method of exposing users to your content, so grew the interactions. It began to be popular to measure how many “likes” a piece of content garnered, or how many times it got shared. LinkedIn measures content engagement as a percentage of impressions and shares or clicks. It also shows how many new followers resulted from a piece of content. If one of your main goals is to attract new followers (and while not a given, it is a common goal) then this allows you to directly measure the effectiveness of your content towards that goal.
But what about the other more popular actions users can take in response to your content (and the new Facebook emotions are too new to really discuss)? Last September Twitter stopped disclosing share data and basically admitted the data was meaningless. Twitter product manager Michael Ducker stated that shares didn’t accurately determine the impact the content was having to a conversation. Does that mean all share data is meaningless? It does give you a sense that someone values your content enough to share it, but did they read it first? Not necessarily.
Many people continue to simply scan Web content, and social media hasn’t changed this, but it has raised the possibility that people aren’t reading what they’re sharing. Let’s be honest: how many times have you seen an interesting headline and either read a little of the article before sharing or just shared the article without reading it at all? And how many times did you never go back to actually read the article you shared? Come on, I’ve done it. No judgement in The Content Corner.
Social media videos aren’t much better. Just as how pageviews and time on page don’t actually track whether or how a user is engaging with your content, Facebook videos, for example, auto-play now by default. So how do you determine if a user actually watched the video or not? Or if they started to watch and then lost interest (pointing to possible room for improvement to your video production or subject matter)? YouTube does delay the counting of views for 30 seconds, but that still can be a less than accurate count.
Can We Effectively Measure Anything?
My first reaction is to actually say no, but that’s not completely true despite the discouraging points noted so far. Relatively new publishers such as Upworthy and Medium are developing new metrics to better measure how much people are truly engaging with their content. Medium for example is developing Total Time Reading (TTR) by using a variety of metrics including scrolls. Eye tracking can also provide a very accurate picture of how a user is digesting your content, but in many instances the software just isn’t practical. And eye tracking remains much more useful for larger design and UX considerations.
The most important thing we can do is be clear about the ambiguity that may exist in certain numbers and remember to be clear about the limitations of certain data. Do shares mean that each person that shared your content read it and would read or share similar content? Eh, not necessarily, but higher share numbers for certain types of content do help develop a sense of topics to explore further.
Comments can also be a mixed bag. I have found them pretty useful on LinkedIn, for example, but on other platforms such as blogs they are so ineffective that companies (and agencies turn them off. It saddens me to say, but the instances of a really valuable and value-add conversation popping up in a piece of content’s comment section are pretty rare (and most comments today are generally downright horrifying).
By understanding the limitations of my metrics, I am trying to take a measured approach to conclusions drawn regarding various content types. Engagement numbers, for instance, do have value to me as they help me determine at this very early stage what types of content I should try to post or share more of. And you need to keep in mind what you are hoping to achieve. We won’t be looking for conversions or sales, but we do want to measure the effectiveness and usefulness of our content somehow. Be clear about your goals and how metrics help you determine how well you are reaching them.You’ve just finished reading the latest article from our Monday column, The Content Corner. This column focuses on helping solve the main content issues facing federal digital professionals, including producing enough content and making that content engaging.