Marshall Sponder is a specialist in social media measurement, as well as a prolific speaker, blogger and consultant. He has also written two books on the topic, Social Media Analytics: Effective Tools for Building, Interpreting and Using Metrics, and the forthcoming Digital Analytics for Marketing.
We asked Marshall for his view on social media measurement in 2017, and here are his tips for social media marketers keen to expand their understanding of a challenging topic.
1. Be selective with data
There is too much data for anyone to make sense of today. Instead, marketers should try to figure out what information is needed at any given moment and then gather it. The image below demonstrates how collecting and understanding web analytics data from a single website can quickly become a complex affair. For example, specific events that are taking place on a site (such as filling out a form, viewing and interacting with a video, interacting with an AJAX widget or an element in an iframe,) turns out to require custom programming and analysis with the website code and the web analytics software.
Consequently, organisations are now able to store all their data, gathered in real time, in a data lake or data warehouse and try to look for patterns, either immediately or later. I think that approach makes a lot of sense given the low cost of data storage combined with highly tuned data algorithms that can run on distributed computing and cloud-based platforms. In that way, organisations can have their cake and eat it too.
2. Don’t get bogged down in ROI
Each social media platform has its own metrics and analytics around the actions that take place there; I call these Intermediate Metrics. These are useful as signs of user engagement, but they are not meant to answer ROI questions, and it becomes frustrating when stakeholders try to get these kinds of answers from Facebook. Social media platforms have immense value, but they are best seen as an ancillary channel. No-one argues that the right vitamin and diet routine promotes a longer life, but we are not currently able to measure the ROI of our health program. However, when our health program works, we’re around much longer to ask those questions!
READ MORE: Can we really measure social media ROI?
3. Bigger isn’t necessarily better
When I worked at IBM, there were a variety of metrics and data that I needed to assemble from many different groups within the organisation, and some of that information was hard to get a hold of. Often, the data was contained in various hybrid and legacy systems and required specific access to people and resources to get it. Smaller organisations may have their own roadblocks, but they are unlikely to be as complex or difficult to navigate than pulling together the kind of data that most medium and large corporations generate and have at their disposal.
4. Snapchat is a law unto itself
Leveraging Snapchat is extremely frustrating and hard to do. As I stated in this article: “With Snapchat, the medium almost rejects analytics in its current form by definition. The app can tell its users how many people viewed one of its stories, how many took screenshots, and how many opened direct snaps and when they did so. Beyond this, the ephemerality of Snapchat causes difficulties.”
The platform’s architectural design and intention are intended to be antagonistic to real user metrics, though Snapchat provides decent analytics to advertisers who target audiences with customisable ads. Another way of putting it: every social platform provides its own challenges, but the analytics provided are adequate for users or advertiser needs, with the exception of Snapchat, which is skewed towards metrics for advertisers only. Even then there are workarounds, as location filters now provide analytics and anyone can run ads.
5. Best measurement tools
When it comes to finding the right tool for your business needs, you should carry out a rigorous needs assessment first. If you can, try out as many different platforms as you can to make the best choice.
However, a full evaluation of competing platforms is almost impossible to perform, as the only way to know if a platform or set of platforms will work in your environment is to set them all up, one by one, and test each one with a small subset of data. This is impractical, as no-one has the time or energy to do that kind of testing. The best alternative is to read around about all the tools and what various experts say, then the decision on what platform tools to purchase will hopefully become easier.
Sign up for our free webinar, How to calculate the ROI of social media, featuring a team of leading social media experts.