In the digital world, data is the new oil. Segmentation and acquiring the right datasets are critical to the success of omnichannel marketing campaigns. We spoke to Alex Wright, Head of Insights, Blis, to understand how data is verified and the criticality of location data in marketing attribution.

Tell us about your role at Blis and the team/technology you handle

My role as Head of Insight is to take the huge volumes of data we collect, understand its origins, use it to test hypotheses or challenge preconceptions, and turn it into a tangible story that answers a brand’s questions and contributes to how they solve their business problems.

This role didn’t formally exist at Blis before I joined the company, so firstly I needed to understand what the business is about, where it fits into the location mobile ecosystem, where mobile location fits within the mobile ecosystem, and where mobile fits into the wider media context. I want to find the best way to portray our qualities and the unique work that Blis does.

Prior to Blis, I worked on the Google account at a media agency, which was my first step into digital media. I learned there that the real challenge is to keep up rather than catch up (which seemed like a convenient viewpoint!), but more importantly that the digital ‘exhaust fumes’ of people’s natural behavior create a lot of passive information we can use to better understand them.

At Blis, how do you verify the authenticity of data? How does it impact your ability to build better products for marketing technology companies?

We capture data via GPS, wifi, and beacons and pass it through our quality control technology to filter out inaccurate and fraudulent points, meaning that we’re only working with sources we can trust. With 50-70% of GPS data being inaccurate, these publishers need to be removed which leaves the data sets at a fraction of the size with the need to be scaled up again in order to identify actionable insights. Want we do is use verified GPS data and then scale it out to public wifi. With our level of accuracy and CPV model, brands are getting the best of location-based services by using Blis.

What is the ‘State of Location-based Data’ technology in 2018?

Location-based services are expected to be worth around 34.8 billion in 2020. Compared to two years ago, marketers now better understand the magnitude of the data-quality problem facing the industry but are having issues with understanding what is inaccurate or how to make up for the data lost.

The increased scrutiny on mobile location data is a positive thing – the industry’s willingness to unpick and understand the data and its origins not only reflects its level of potential, but will also hopefully lead to an increase in data quality as inaccurate and fraudulent players get found out and squeezed out of the market.

Consumers have become more comfortable sharing location data, but they are also selective about when they want to share that information. According to our research as well as eMarketer’s report from earlier this year, consumers have also become more confident in demanding value in exchange for providing their data, prompting many marketers to have to explain how data will be used.

What are your predictions for location-based behavioral marketing? How does Blis help customers to build better behavioral marketing campaigns?

We believe the cost-per-visit is the future of location-based advertising. Leading the development of this technology, Blis is pioneering campaigns where brands will only pay for actual site visits. This level of transparency and performance-based compensation model keeps agencies and technology companies honest, while potentially rewarding them for effective campaigns.

Location-based behavioral profiling could be a great complement to the existing audience profiling tools currently utilized by client-side marketers and their agency partners. If behavior – and decision-making – are the actions we as advertisers want to influence, then profiling based on behavior, and attribution using behavior as the key metric makes a lot of sense.

How do you define ‘Customer Experience’ at Blis?

We define Customer Experience as an engagement with a brand that is messaged at a relevant moment in time. This engagement provides value to a customer based on where they are and is amplified cross-channel in a meaningful way, ensuring alignment on all channels from OOH to digital.

Why do marketers find it hard to attribute customer success to experience management? What are your recommendations to overcome these challenges?

It will often boil down the disconnects in the media / behavioral transfer – essentially, we apply labels and models to behavior in order to help us understand and explain them, but consumers don’t apply these same labels to themselves.

This ends up meaning that the profiles we create (whether through 1st party data, proprietary research, audience segmentation, syndicated sources etc.) are never truly representative of the audiences we hope to reach, which in turn means the media we buy becomes a loose mix of over-indexing channels. We then pick whichever attribution model (first-click, last-click etc.) seems to work best – which will often be dictated by the channel that delivers it! – which results in a very restricted view of a consumer’s brand touchpoints (whether they’re advertising, prior experience, personal recommendations, reviews etc.).

Being able to tie physical attribution (through store footfall) to digital exposure (across screens, and across publishers) is a real leap forward.

How would you define Personalization- Segmentation- Optimization, and Automation at various stages of your digital campaigns?

Segmentation first – define, then identify an audience based on passively observed location behavior.
Personalization next – through creative. Whether this is message, format, or time/day/location for delivery.
Optimization and Automation – these work together, depending on the buying route.

How do you leverage AI/ML and data science at Blis? What AI companies are you particularly interested in?

Last year we released Blis Futures, an AI-powered, predictive location modeling solution. It identifies consumers most likely to visit specific locations, then provides advertisers a guarantee they will only be charged per consumer visit. Blis Futures uses machine learning and predictive analytics to group like data sets and identify patterns. The solution builds audience segments based on their predicted conversion rates and indicates when and where to target this audience group to achieve the advertisers’ business objectives. This ultimately eliminates waste in spend and increases campaign performance.

From my point of view, I think those more commercially driven businesses have a far harder task trying to predict consumer behavior – which is not to say they won’t do a good job influencing it! I think using AI in medical/pharma – where cause and effect are well understood – is where real progress will be obvious sooner. Decision-making is too often impulsive and irrational, which is partly what makes it so interesting and intriguing – the allure of the ‘impossible’ task!

Thanks for chatting with us, Alex.