Sklar Wilton’s Michael D’Abramo on how to handle data
“Most businesses have more data than they know what to do with. Now what?
Good data. Bad data. Big data. Rich data. Retail is an industry obsessed with numbers, but managing and mining the ever-growing data stream can be a challenge. We recently spoke with Michael D’Abramo, managing director, research at Sklar Wilton & Associates – a strategy and insights consultancy in Toronto – about what it means to be a data-driven company, how data can be used to improve the in-store experience and why delivering relevant offers can be a challenge. This interview has been edited for clarity and length.
Can you explain what it means to be a data-driven company?
Well, I would say the decisions that businesses make are based upon data first and other factors second. And that’s the, shall we say, literal definition. But I think a greater way of looking at it is that it’s a commitment – you invest in the right tools, you commit time to what matters. So, to be data-driven, you have to invest in it. You have to prioritize your data. What do you have? What don’t you have? What do you need? What’s the quality of that data and the resourcing around it?
What does it take to commit?
Data needs to be set free in organizations. Anyone using data analytics needs to be trained to do so. And you also need to commit to the technology, the dashboards and other digital tools that set that information free. We make a lot of dashboards for our clients and that allows them to run simulations and the like, and that’s really powerful. But, unless you have the good data going in and you have people trained to decode it, it’s not going to work for you. You need to get into it and think about it all the way through.
Why, with so much data, are companies having difficulty delivering relevant offers?
I would put it this way, we’re in the adolescent phase. We’re awkward, our voices are cracking, but we’re growing up. The progress for the most part is positive, but let me try to address the macro picture here. Accusing the data of being the problem may be the mistake. We’ve seen, over the years, that lots of people use a shared account – a husband and wife, a family. How are the individuals sharing the account different? Are we connecting our offers to the real people behind this account? Are we missing a mark? And the final thing I’d say is the human element. I spend a lot of time in qualitative research. I’ve done the focus groups, interviews and shop-alongs; I can tell you a likely explanation for a lot of the failures of these offers is customers who are short on time, distracted or forget to check their offers. Yes, it’s important to commit to the data, but you also need to commit to understanding how the customer behaves or understand how the [user experience] of an app works. All those things matter.
What about smaller, independent players? How can they best commit to data?
You do whatever you can to generate access to insights. An independent store can use point of sale, loyalty programs or anything else they have access to. The first thing you do is inventory data sources. What is available to you? A lot of independent stores service a specific physical community or an ethnic community. So, be more ingrained in those communities. This is where, shall we say, a good marriage of classic market research interviews and surveys working with the data you have is going to be more powerful. Use what you have, inventory what you have, use it as best you can and combine it with the other resources you have available.
How can grocers use data to improve the in-store experience?
One of the ways to look at it is using the data to experiment with different displays and layouts. Find a pilot store or a store you can play with to see how different possibilities open opportunities for new sales – stocking products in combination, different side-by-side products, having a section for all the sale items, working with suppliers to better predict demand. There are a bunch of things worth doing, but I’d love to see more experimentation. I think sometimes people think data is some sort of science, but the science only works when it connects to the human experience. Your question at its core is how do I take the science and put it in the real world? I think the answer is… You have to experiment.
Some retailers don’t have loyalty programs, yet. Then what?
Point-of-sale data. Another one, too, is to look at your partners. You’re buying from different suppliers at larger companies. They may be able to provide you some of that information as part of the relationship. That’s another area. But point of sale is a good place to start. And then also looking for secondary patterns. Not just what they buy but the combination of items they buy, the time of day they shop. And then, if you can, do some profiling. Get a hundred consumers to give you a little bit of information about themselves and then research with them and backfill some of the point-of-sale information. I’m being very speculative here. There are many ways you could approach it. It’s hard to know what things you can do but, you want to combine the most reliable data you have with some other insight-generating opportunities like a survey.
How can grocery companies make data work for them?
Once you have a clean organized data set, you need to figure out what you’re trying to accomplish with it. What kind of queries or patterns are you looking for in the data? And this is where data scientists can help reveal some of those things. This is also where democratizing the data comes into play. The challenge becomes when a lot of times this stuff is top down, the data is only shared with the most senior people and that limits its ability to be meaningful to the organization.
What role is artificial intelligence playing in the way data is collected and analyzed?
It’s in its infancy, but I think this is a case where you don’t want to wait. You want to get out and experiment. AI could play a role with the large amounts of data – it can play a role in customer purchasing habits all the way to inventory management. You can create algorithms that analyze shopper behaviour and that can help retailers predict product offerings, create better offers or combination offers that can increase, say, the actual receipt at point of sale by getting people to buy things in the right combinations, promotions and other things. And you can also use AI, and this is something that it’s a little bit beyond the retailer side of it, but there’s probably some supply chain advantages to be built in here as well. So, there’s a bunch of different things that can be done, but it’s still a growing field and this is where we need some ambition to see where it could go. I think there’s some cool stuff out there. There’s no doubt it’s just a matter of taking chances and seeing where it can go. “
*This article is excerpted from Canadiangrocer.com website, published April 3, 2024