BM: Many people have heard the term ‘big data’ but may dismiss it as a buzzword. Can you describe it in a way that’s relatable?
MF: If I’m talking to a customer or business that’s just getting into this, I’ll often say the services that surround the phenomenon of big data are really the result of asking yourself this question: “What’s the one thing you really want to know about your business, and you know for sure you’ve got the data to get answers, but just can’t get it done?” This is often a really good starting point.
I also like to use the newspaper analogy. In the US, there are two different kinds of newspaper readers – those who prefer the Wall Street Journal and those who read USA TODAY. Readers need to be able to trust that USA TODAY headlines are appropriate, high-quality, and trustworthy, because they don’t have time to read the Wall Street Journal. The same applies to marketers who don’t have time to do all the background research on a system that’s telling them “These are the offers you need to be sending out.” Consume data in bite-sized fashion and be much more productive with the data insights that are being generated.
There’s a human behaviour issue around the advent of big data tools, because people are not quite ready to trust the machine yet. The element of trust needs to be built up between the human and the computer – once you’ve got that, you’ve got an opportunity to be much more efficient.
BM: Your panel discussion will touch on marketers wanting to delight their customers with insights offered through big data, and the challenge of personalizing without getting too personal – care to tell us a little more?
MF: There are going to be some great brands there, including consumer brands as well as software brands.
We’re going to, among other things, talk about that very question. How does, “Wow! HOW do they know that about me?” turn in to, “Wow! How do they know THAT about me?” That’s the challenge, because now it’s now much easier to have intimate relationships with your customers. To what degree do you use that in a way that pushes the limits or breaks convention, and that you’re hoping is good (and very much more, hoping isn’t bad)?
It’s not just about “Are we capable/do we have a data collection mechanism that allows us to do this/do we have processing/do we have a person to interpret the data to get these insights”, it’s “What is our data collection and data governance strategy/do we actually have a plan for this/do we look at this through an ethical lens/are we concerned”. There really is no specific answer, but there are some great stories.
I think that if you’re on a path to delighting your customer, that’s terrific, but there absolutely has to be a listening mechanism so that the moment you’re starting to get into the grey zone, you can back away instantly. If you wait until it’s in The Globe and Mail that you’ve creeped your customers out, you’ve gone way too far.
BM: Do you recall a time when you felt creeped out by a marketing tactic or approach?
MF: I think the most pervasive example is programmatic advertising – you click a link on Facebook and suddenly you’re getting ads for that same product on every single one of your devices. That’s tough, because companies obviously want to deliver targeted advertising. I’m not sure it’s creepy. It’s probably a little frustrating. I think most people would share those kinds of stories.
As far as having been given an offer that has really kind of freaked me out, I can’t think of one specifically for myself. I’m on the winning side of this. I prefer to get offers that are for me. Sobeys is on the panel, but Loblaws does an exceptional job with PC Plus. I’m rewarded for buying more of what I’m already buying. There are two paths you can take to understanding basket analysis and previous purchased data. If you’ve been buying a specific product on a regular basis, you might start receiving special offers for a competitive brand. It might be a really good deal, and it’s great if that retailer is actually getting money to push that because that’s a revenue source, but it’s not always going to generate loyalty. But if I buy a product on a regular basis and I get an offer to buy more at a special price, or to receive more points, then I’m going to be thrilled. In this case the retailer isn’t trying to change my mind, but rewarding me for being consistent.
BM: With targeted advertising, is there any danger in narrowing people’s interests/exposure to new things by rewarding them for consistent behaviours? Is there a risk of not being exposed to new things because everything is tailored to a customer’s history/perceived interests?
MF: I think that there is a risk there. I also believe that as the marketing technologies that are being employed to deliver these targeted offers or targeted advertising increase in sophistication, a consumer’s control over the ability to control their digital footprint increases, too.
I could choose to lock my system, but I choose not to. I think that if you’re an individual that’s naturally curious, you’re probably going to allow yourself to experience all options online. If you’re someone who, instead, just wants the best offers possible, you’re probably ok with the fact that you’ve self-selected a narrow set of choices. I think there’s a little bit of savviness that comes with being able to control your digital exhaust, but ultimately I don’t think your choices are necessarily are going to be limited because of the fact that you’re making the decisions in the first place.
The insights that can be derived from data are continuing to grow, both in volume and also in breadth of detail. And what I mean by that is that there’s an increasing amount of meta-data that’s being collected in conjunction with the primary item that’s being measured. New opportunities to combine products, possibly better products, or something that might be more appealing, will be revealed, so the savvy marketer will look for those opportunities to create unique offers at a greater level of intimacy.
BM: Sociology and behaviour science plays a role in interpreting data for marketing purposes. I’m thinking specifically about Instagram, and the recent algorithm that now displays and suggests content based on your behaviour. This is really interesting to me, because I think (specifically with visual platforms) users aren’t necessarily looking at content they’re interested in, but at content that intrigues/disgusts/scares them, or maybe something catches their eye and suddenly they’re being targeted with similar content they’re just not that interested in.
MF: I totally agree. This is where companies need to do a better job at getting to the real insight of this data because your persona online, when it comes to something like Instagram or Facebook, really has two completely different representations. If your friends are looking, then they’re seeing what you’re sharing – i.e. working out at the gym, a meal at a restaurant, photos of smiling family members, etc. What you’re sharing vs what you’re clicking on is so completely different. Clickstream analysis can provide some serious diametrically opposed representations of what your persona is. So which one is the right one to track? That’s hard to figure out.