Data-driven points of view: Tripadvisor, Spotify, Uber and DigitasLBi on data

The Drum has selected highlights from the points of view across the day from the speakers.

Chris Clarke, chief creative officer, international at DigitasLBi on getting it wrong with The Cluetrain Manifesto

In discussing why his generation of creatives went into the business, all guns blazing in the fight for digital, Clarke said hindsight meant that perhaps caution should have been taken earlier.

“For us it was about, can you collaborate with a wider network of people? There was a messianic zeal to disrupt the nasty bad old world of business. The Cluetrain Manifesto was our book for disruption; it is where the idea that information wants to be free came from. Or free or funded by ads, and it’s one of the most dangerous things to happen,” he said.

Joel Fisher, director of display advertising sales, APAC at TripAdvisor on using data to solve problems not margins

In talking specifically about data and advertising, Fisher discussed recent issues that have emerged, such as transparency between agencies and brands.

“Budget would once return the best for clients in terms of sales or ROI. We now have ads through trading desks and some place budgets where margin is the highest. There is an inherent conflict in that and agencies should always put budgets where it suits the client, not where it will benefit from the margin. As an industry we really need to look at the role of the agency in programmatic and make sure we best service the client from a publisher side and agency side. Budgets should go where it makes most sense, instead of margins,” he said.

Sunita Kaur, managing director of Spotify Asia on people wanting algorithms

Spotify has invested heavily in data, both within its platform and also in using it to forge recent advertising campaigns, personalising to tastes and quirks in different geographies. Kaur says this means people want companies to understand them.

“I remember in 2009ish, when Facebook introduced targeting, the whole world was up in arms about ads knowing what they liked. Then three years later, feedback was ‘how could you not know what I like?’ The world we live in now, and it’s shown in countless privacy studies, is that people are sharing more and more online because they want algorithms. They understand them and they want companies to understand them. They are entrusting them to use it for good and not evil, so as long as we stay within the good side of consumers, it’s ok, but the second you screw it up they are pretty unforgiving,” she said.

Grace Tang, data scientist at Uber on the marketing industry rushing into data science too quickly

While Tang admitted she was more on the developer (instead of marketing) side herself, she also thought that many businesses less directly associated with technology were rushing into hiring data scientists.

“Data science in the wrong hands could be dangerous. Data is a buzz word and a lot of agencies and companies are rushing too quickly into hiring a data scientist They don’t know what they want and it is such a generic term. It can refer someone who can analyse data or someone who collects and cleans it. Brands need to establish a pipeline before they use it. I do think that companies need to understand data science and what they want from it first,” she said.