CES 2017 Recap: Amazon Alexa, Machine Learning, And The Future Of Data

Connected devices are already producing huge swaths of under-utilized data. Kinetic UK's managing director, Rosh Singh, explains why we need to think about it now.

At this month’s CES, technology like VR and unexpected connected devices predictably got most of the glory. But Rosh Singh, managing director at Kinetic UK, was considerably more excited about data and machine learning.

“We all just want to show the right message to the right person at the right time — and data is the way we do that,” Singh said. “What’s exciting… is that AI and machine learning [are where] we can finally start to evolve that data set.”

In your mind, what are the main takeaways from this years CES? What trends will be the most impactful?

There are a few, but for me, there are two main elements: This idea of AI moving into automation, as well as the Amazon Alexa trend — personal assistants and voice search.

We have a cornucopia of data — whether it’s our data, planning data, or our client’s data — and we’re using AI and custom algorithms to understand that data. I think it’s hugely interesting for what we do. I wouldn’t say we’re leading the field, but it’s definitely something that we should be looking at that is going to impact the industry. Like integrating [technology] such as IBM Watson into what we do; it’s something that we have on our radar for this year in terms of [improving] dynamic, data-driven digital out-of-home.

The aim is to pull in data sources to make sure the content on billboards is as relevant and as real-time as possible. It’s based on this idea of really thinking of a digital billboard as more like the internet than a standard poster would be; it shouldn’t just sit there for two weeks the same content on it.

So, how can we inform that with data sources? Well, we can say that happens in two ways. The first is the data as a trigger. For example, we could use weather data to say, “okay we know its cold right now, so it would be great if we could reference the fact that its cold.”

Secondly, we could actually use the data as content. That might mean inputting a social media feed, for doing some social listening and understanding. Then, we can engage in sentiment analysis on it, thereby understanding how happy people are in one area — and then visualizing that on the screen, with content that could change on a screen-by-screen basis. We’re already doing a little bit of this using a lot of data sources to make sure the messaging and the content on the screen is as relevant and as targeted as possible. Ideally, of course, we just want to show the right message to the right person at the right time — and data is the way to do that.

Where AI and machine learning is interesting is that we can start to evolve that data set.

And what about the second trend you mentioned, Alexa? How is that impacting the way consumers engage digitally?

It’s [all] related: I think what will be interesting about Alexa, is that we’re starting to see real machine-to-machine communication. We’re seeing small packets of information being passed between our devices, to make decisions on our behalf, which I think is really interesting. As a marketer, we really want to look into the future and see, “what can we do with this new data?” We have to start thinking [that way] now.

In a future world where our devices are talking to each other continuously, how can that data be taken to a macro level and used to improve targeting, to improve creative, to improve planning? I think that’s what really interests us, and I think AI is the way.

For example, on one hand you have a mass of data — this set of data created by the internet of things and machine to machine communication — and on the other side, you’ve got media. To get from one to the other, you need a bit of machine learning, to make sense of [how to] connect it. It is all about predictive analysis, and understanding that data, and then making instant decisions on that data — based on what the algorithms learned previously.

There’s significant potential to leverage this data to personalize digital content, as you mentioned. When it comes to these customized ads and messages, how is Kinetic looking at the attribution side? Are you measuring place visits and attributing them back to which consumers saw this dynamic out-of-home content?

That’s something that we’re really focusing on this year, definitely. We’re looking at various ways that we can do attribution tracking through billboards to in-store — whether that be beacon based, mobile, GPS, et cetera.

We’re looking at ways that we can piece together that journey and understand, if someone walked past a poster, what is their action past that point? The key is [to bridge] it all together: Whether that next action is walking into a store, or going online, or making a search, or opening an app — we need to understand that.


About The Author
Lauryn Chamberlain Lauryn Chamberlain @laurynchamberla

Lauryn Chamberlain is the Associate Editor of A New York City based journalist, she specializes in stories related to retail, dining, hospitality, and travel.