MachineIntelligence.jpgDavid Raab is a consultant specializing in marketing technology and analysis for clients in financial services, retail, communications and other industries. His current work is focused on designing marketing architectures and selecting marketing tools. Raab has written hundreds of articles for DM Review, DM News and other industry publications. He blogs regularly here.

UnboundID: Recently you wrote about the shift from transactional to relationship buying in the consumer economy. How is this playing out today in marketing, sales strategies?

Raab: Think of the Nest thermostat. We don’t often think much of thermostats. We never had a relationship with the company that makes thermostats before. But now you get this thing and you train it and it communicates with you. It knows more about you than you might imagine. The consumer must think more carefully about whom they are entrusting all that data with and consider, what if the thermostat maker is sharing data with some I don’t want to have it. So, trust is key here, in that you aren’t sharing data on your customers without asking first. Transparency is also really important. It doesn’t mean you have a document that is 20 pages long and no one reads. It means you have a simple document that explains what you are doing. People should be able to choose to some degree how you use their data, through opt in and opt out capabilities. Finally, trust is a matter of branding. Trust in the company is more important to me than whatever technology the company is using.

UnboundID: At the same time, you wrote that many decisions are being relegated to machines. How do you strike a balance between having a more personal relationship with customers while also automating as much as you can?

Raab: I suspect that there is really not a conflict between automation and trust. Many people have a better relationship with their phone than their spouse. Computers might actually do a better job of listening. They don’t get tired and bored, and as they get more socially adept they will respond appropriately. It’s not that the computer is reading my mind, but it is anticipating my needs. My thermostat knows when I get home every day, maybe from data transmitted by my car, and it will turn up the heat for me automatically when I arrive. As another example, we are seeing more of these chat tools online–intelligent agents that ask questions and help solve problems. You don’t always know if it’s a human being on the other end of the connection.

UnboundID: What do you expect from the next iteration of marketing technology systems? Where is the innovation needed?

Raab: In BtoC companies, we will see machine intelligence baked into everything. That automation will lead to huge simplifications, such as in setting up a campaign which is today extremely complicated with most tools. Doing omni-channel marketing is a large undertaking. This means that you are limited on touches and segments you can reach. In the future, machines will do most of this campaign management work automatically. Another area is in attribution software, which involves understanding the impact of various marketing activities. In the past, we’ve lacked the data to know who read what or viewed a particular video and what was the result. As this technology advances the marketer will know which promotions worked and then can predict the impact on future behaviors. You can understand how a particular advertisement affected sales of the product more accurately. The next challenge is to track individuals over time, not just segments.

UnboundID: Is it realistic for large consumer brands to drive for the integrated customer profile, the single version of the truth, as many startups are proposing?

Raab: Yes, although this doesn’t have to be perfect. You are always chasing the data. But the point is to reach a critical mass where there’s enough information available on enough people, so that you can begin to make useful judgments and decisions. I think it’s extremely important for companies to build up these customer tracking systems and profiles.

UnboundID: Is there a particular way that’s best to do this – such as physically integrating the data into one database or providing a virtual view of all the customer data in one place?

Raab: Most companies will use a mix of methods to obtain this integrated profile. It’s possible to know at any given moment where somebody is and the weather in their location, but you’re probably not going to store that type of data. That’s high volume quickly-changing data that’s probably not going to be super helpful. But, any data that requires processing or delivers a historical perspective, really any direct customer interaction with the company, you’ll want to import and store that on your systems. We know how to do this and have the technology available, so it’s really up to the company if they are going to make that commitment. Hopefully, the companies that don’t do it will be punished in the marketplace. But with some of the semi-monopolies that we have in transportation and utilities, it doesn’t always work out that way.

LIVE WEBINAR: Navigating the Customer Data Privacy Minefield