Rapid prototyping to integrate TV and web data

A leading Swiss online retailer, which relies on both online marketing like display ads as well as TV to advertise its products, wanted to measure the impact of its marketing spending. TV remains the most effective channel for reaching a mass audience – a German study shows that adults spend an average of 208 minutes per day watching TV. Therefore, the retailer was looking for a neutral expert partner to create a solution that would enable optimum allocation of its high TV advertising expenditures.  

For this purpose, the client wanted to reliably integrate TV data into their web data analytics tool that would measure TV-driven online conversions – something very few service providers could deliver. To further complicate the challenge, the project was spearheaded by a small internal team, which needed to win the backing of other decision-makers. Therefore, based on our analytical and data-driven approach, we were tasked with ideation and product development with the aim of producing a software prototype. 

The FELD M team worked with the client’s internal team of technical experts and budget decision-makers to ideate and advance to the prototyping phase as efficiently as possible. We used workshop methods like data thinking to explore the concept of TV integration and give all stakeholders a clear picture of the problem posed by missing TV information. We drew on experience in data architecture to design the infrastructure for TV and web integration. As a result, we built a quick prototype in cooperation with the client for analysing TV-induced web data in near real time. 

From prototype to software solution  

In this project FELD M developed the “Push2Web”- Service, which automatically detects TV spots in real time and registers website visits within a significant time period of airing via an application programming interface (API). To achieve this, FELD M wrote a customized JavaScript tag, implemented in Tealium iQ, to obtain TV spot information like channel, genre and block position for each website visitor. This information is channeled through the data layer and on to the Google Analytics or Adobe Analytics platform.