Simo Ahava is a recognized expert on customizing web analytics and tag management solutions to improve the entire “life cycle” of data collection, processing, and reporting. In this interview Simo is talking about the difficulties to measure engagement – but also sharing advices how to overcome them.
Simo Ahava is a well known guru in Web Analytics. His main areas of expertise lie with Google Analytics and Google Tag Manager, and Google has appointed him as a Google Developer Expert in these fields. Simos main focus is on increasing awareness, skills, and critical thinking around data. During Analyticsdagarna, Simo will show some concrete examples from the world of Google Analytics and Google Tag Manager to illustrate the complexity (and the simplicity) of tracking various engagement interactions on any website. In this interview, Simo shares his thoughts about the difficulties to measure engagement – but also advices how to overcome them.
Simo, can you first of all mention some common difficulties for organisations when measuring engagement?
1. Organisations are immature with data and analytics. It’s very typical to invest heavily in an analytics platform to chase the elusive “360 degree custom-centric holistic view” paradigm in measurement, but because the organisation lacks the maturity to properly implement a data-fuelled service development process, the end result is an expensive toy which builds beautiful but useless dashboards.
Measuring engagement is a great proxy for detecting what the organisation’s maturity is, because engagement requires a very broad understanding of the business and how users interact with it. By improving the organisation’s capabilities of understanding what “engagement” means for them and how it should be measured, the overall maturity of the organisation will improve as a consequence.
2. Not understanding or agreeing on what “engagement” actually is. This is very typical. Analytics is deployed to measure “engagement” and “interaction”, but because there is no consensus (or no forethought at all) to what these words mean, the organisation ends up using vacuous metrics like “Bounce Rate” or “Time on Page” to indicate how involved users are that visit the site.
Engagement is extremely sensitive to what the business does and why the website or app exists. There are no universal, generic metrics that can be applied out-of-the-box – engagement measurement requires, without exception, a very robust understanding of what users do on the site or app, and what the organisation WANTS the users to do on the site or app. There are no short cuts.
3. Lacking the skills to implement the measurement. This is perhaps the easiest problem to fix (via proper hiring process), but it’s also the most common one to witness in the wild. Implementing measurement is more than just adding page views and events to the site. It’s about developing a common structure for the semantic metadata that each hit collects, and understanding WHEN and WHY to collect data are just as important as WHAT data to collect.
How can you overcome these difficulties?
One way to overcome ANY difficulties in any organisation is to work on the communication structures within the organisation. Failure can almost always be attributed to a broken channel within the organisation. If there is a bug in implementation, it’s typically because there wasn’t enough communication on how to deploy some measurement point. If the analysis is flawed, it’s typically because there wasn’t consensus on what insight stakeholders actually want to derive from the data. But with engagement, specifically, it’s important to build a strong understanding of the technical aspects of the implementation. Understanding how the website and app react to user interactions, and adding measurement points to these interactions without disrupting the functionality of the site or app is paramount in developing competencies in any analytics implementation project.
Google Analytics, for example, is spectacular in how many different ways it can be used to track one single thing, but to be able to use it at a level required for robust engagement tracking requires quite a bit of understanding about ALL the aspects of GA’s data processing pipeline. What is collected on the web or in the app is just the beginning – the data has to pass through filters, sampling thresholds, and collection quotas, and you need to prepare custom dimensions, calculated metrics, dashboards, custom reports, and so forth to collect the data.
And if you want to use the analytics data collected from the site or the app to join together with other data that your organisation is collected, a basic understanding of data engineering principles is necessary.
What are you most looking forward to at Analyticsdagarna?
Three things 🙂
First, I want to experience first hand this event that so many people talk about and that has seen so many illustrious speakers on the stage.
Second, I want to learn new things! I’ve been to too many conferences which just rehash things I already know just for inspiration – I want to learn things that are outside my comfort zone or that I’ve never though of before.
Third, I want to share my own experiences and approaches in analytics data collection and engagement measurement with the audience.
I’m sure all three expectations will be satisfied!