Ambient data and applied intelligence will help marketers work smarter and create better, more relevant campaigns. Here are two ways to approach data for improved outcome.
There will be no more business as usual once artificial intelligence becomes a reality. But even today there are great ways of developing faster, better, cheaper solutions thanks to data and machine learning. We already know that there is immense value for all businesses in connecting multiple sources of data and applying them in new ways. This is true for the data held in our CRM systems but also for data stored beyond our own repositories. The data that exists beyond our own business, market and consumer targets holds immense potential but only with techniques such as machine learning can we discover how apparently unrelated data and intelligence can be used to inform the planning and timing of marketing and messaging. Don’t think, however, that we can leave it to computers to do all the work. The solutions we identify will only be competitively meaningful if we give them a clear, human purpose. The key is to enable artificial intelligence with context and ambience in order to drive the consumer journey to completion. In essence, we need to make sure that the consumer journey is assessed from all possible and even the most unlikely angles in our quest for the best possible results.
Today we are surrounded by data freely available but much of it is not connected consciously or appropriately to the way we plan and buy campaigns. This data system, ambient intelligence, allows us to identify and track more of the factors that truly have an impact on consumer purchase decisions and use them to help execute campaigns. One of the simplest example of this is the use of weather data to trigger messages that drive impulse purchases: with a sunny weekend coming up, a female razor brand might increase frequency or visibility and a grocery store might put ground beef and chips on promotion.
Adding such data inputs to programmatic buying tools allows us to apply such changes automatically not necessarily to the messaging itself but to the timing of those messages. Machine learning will also improve our ability to model the impact of these strategies, thereby delivering even greater efficiencies. In the near future, we’ll know the precise temperature at which specific offers should be turned on or off in New York vs. London. A more complex example of ambient intelligence is the use of the social psychological data from cultural studies that identify and characterize the mental models of different national groups. Such studies can provide an understanding of the factors that make some creative messages efficient in some markets but fail in other, seemingly identical markets. Cultural studies are a well-understood, quantified field of research but they have never been used in a structured way to inform how marketing messages are shaped. But as an ambient data source that can increase relevance and efficiency of marketing this offers a powerful addition to the more traditional input and assessment tools. Information and understanding beyond the traditional sources are adding a new dimension to marketing and aiding creativity and effectiveness.
The other area we need to consider for creating more relevant campaigns is how we apply data in the work flow and our ability to do what we do in a smarter way; a field we have called applied intelligence. It can seem like a big promise but it can for ease of purpose be approached in two ways: firstly look around to see what data exists to maximize the power of a particular insight – as with the weather examples above. This is the inside-out approach where you gravitate from the main objective and identify solutions that can support that.
The second approach is outside-in with a focus on all the available data around and then testing the ways that each particular source of data might be useful. This second approach provokes innovation and creative thinking and is a way to actively seek efficiency gains and consumer insights.
The question that all businesses need to ask in this approach is, What data is available to me and how can I cross purpose it for this product or campaign? The two approaches allow us to define and investigate the boundary between human intelligence and the efficiency of the machines. The goal should be to get better and better at applying intelligence to maximum effect. This will help us make both the process and the output faster, better and cheaper. The key to success, however, is for humans to give the machines clear guidance on the purpose and goals to be achieved.