How to utilise Quantitative Analytics and Qualitative insights

It’s easy to forget just how accustomed we’ve become to dealing with data in our everyday lives. Millions of us rely on our phones and devices every day to tell us how many steps we’ve walked, whether we’ve hit our dietary ‘macros’ or let us know how good a night’s sleep we’ve had.

Behind our relatively new-found obsession with number-crunching and data, the implied mantra seems to be “the more data we have on X, the better we understand X - the better our lives will be”.

While I’m definitely of the opinion that having access to data about what’s important to us is generally a good thing, what we do with that data is far more important. For instance, it’s relatively easy to draw correlations between unrelated events and use statistics to tell nonsensical, yet seemingly ‘true’ stories. Just ask Tyler Vigen for proof of this - a man who has devoted much of his time to showing how easy (and often comical) a process this can be.

But the aim of this post isn’t to bash quantitative data analysis, indeed far from it – it’s the bedrock of digital analytics and certainly essential. Instead, I want to outline some of the benefits that companies and marketers alike can reap by adopting qualitative analytics into their analytics process.

We’re all statisticians now

The exponential growth of data in everyday life is an exciting and brilliant thing. Outside of our personal lives, perhaps nowhere is our growing familiarity with data and data analysis more obvious than the sphere of digital marketing – an economy virtually built on data. Looking back over the last few years, businesses of all shapes and sizes have embraced digital analytics to better understand their customers, website visitors, their behaviours and habits to largely excellent effect.

The type of digital analytics many of us are accustomed to (the quantifiable things) can be hugely beneficial in helping us understand who our website visitors are and how they behave on our websites. However, where they tend to fall short is in telling us why a user did or didn’t carry out the action we hoped they would.

To take a simple example, Google Analytics can tell us in an instant that a newly redesigned homepage is causing visitors to bounce 12% more often than an old design, but that statistic alone does very little to tell us why that change might have taken place, and importantly, what we need to do to improve the user experience.

Why is qualitative data analysis important?

Qualitative user research methods like user testing, heat map analysis and surveying (to name but a few) have the ability to give us deep insights into how users think, feel and respond to aspects of our websites that traditional analytics fall short of explaining.

In the case of user testing, the process of watching (and listening) to users as they complete specific tasks gives us an entirely different window into user behaviour. As the Nielsen, Normal Group (2014) suggest, “the most effective way of understanding what works and what doesn’t in an interface is to watch people use it”. Although this kind of research yields a markedly different type of data, it’s really just the other side of ‘traditional’ quantitative analytics coin and equally as valuable.

Combining qualitative and quantitative data analysis

Where we can often drive the most value as analysts is in combining these approaches. For instance, we can use quantitative click-stream analysis to dig deeper into ‘hunches’ we have about a set of behaviours on a website, or alternatively we might build on user testing findings about a website feature by developing A/B tests to drive website improvements.

In both cases, the data alone can’t tell us what to do - we as analysts and marketers must interpret the most appropriate response and act according to the needs of our business and our audience.

As has long been espoused by analytics ‘evangelist’ Avinash Kaushik, the quest for a ‘single source of the truth’ to answer all our website questions is often a futile one which can lead us to arrive at false conclusions.

Instead, we need to be prepared to embrace a mixture of methods, data points and views to create a richer tapestry of insights than any single source might afford. Armed with these qualitative and quantitative data sets, we can be better be prepared to drive the improvements our customers and businesses both need and want.