Data And Agility Will Help Retailers Weather Tariffs, Circana CEO Says

Forbes

07-04-2025

Circana is a global data and analytics company that focuses on CPG food and beverage, as well as other categories sold at retailers. Stuart Aitken has led the company since January, and has been in the retail analytics space for years as chief merchant and marketing officer at Kroger and CEO of the grocer's data science and analytics company 84.51° and retail data science company dunnhumby.
Grocery retail is likely to be one of the first industries to deal with tariffs, and I talked to Aitken last week about what today's data is saying and what retailers should do, as well as how the retail data analytics industry can lead to deep insights. This conversation has been edited for length, clarity and continuity. It was excerpted in the Forbes CEO newsletter.
What kinds of things can you learn about today's economic situation from collecting consumer data?
Aitken: It's massively varied, but understanding consumer behavior is critical in times of turbulence.
If you think about tariffs today, if we see significant movement in certain products, where will customers move? Which other products will they substitute for another? As we think about price increases, customers are incredibly flexible. They will move to other products as these things flow through, and the data we collect helps establish what they'll move to and why. Is it price? Is it a substitutable eating experience?
That's just the buying behavior piece. But if you think about the data we collect, it's all the way up through the supply chain. Where's the product being sourced from? What's the price of that commodity that's being sourced, all the way through to the consumer, such that we can help our customers understand all points of the supply chain, where the breakdowns are and how they can pivot across that, including commodity pricing.
Many of us were caught shorthanded during Covid when we had one supplier for [ingredients]. That lesson is still in place today and will be incredibly valuable for people managing through tariffs. If one particular country is hit harder than another, the ability to pivot is much greater. The big difference this time around is the number of countries that are being hit by tariffs, but the ability to still pivot is crucial.
Circana CEO Stuart Aitken.
[The information we collect provides] the ability all the way through the supply chain with data to understand: At the end point, will the consumer switch if the price is too high, and if we're out of this particular product, where will they switch to as well?
The other piece that's incredibly valuable is understanding price sensitivity, especially in this environment. We have the coefficients and all of the elasticities across all items, and we have demand forecasting tools that help our suppliers understand: If we're going to raise prices 10%, 15%, 20%, what will that do to the demand? How much do we actually need to buy?
Food is where you're going to see tariffs flow through the fastest, especially perishables: 82% of vegetables come from Mexico and Canada. Through demand forecasting and through these coefficients, we can quickly see where the consumer will move, and what each individual client or brand needs to do based off of that information.
It's helpful from a pricing standpoint, a promotion standpoint, an assortment standpoint in regular times, but today it's never been more important.
You have been in this niche of the business for much of your career. How has the data you can collect and the insights you can draw from it changed?
Dramatically. It used to be a few niche players were collecting and using data. It started with loyalty cards early on. Retailers were collecting the data, but nobody was really using it and leveraging it to help the consumer, and the retailer and CPG be much more streamlined and focused on the consumer.
The first loyalty cards were happening right after point of sale data in the early '90s, but Tesco was arguably the first retailer to leverage the data through the likes of dunnhumby. Since then, the number of people using it has increased dramatically, largely because of what you can do with the data. Understanding consumers longitudinally is critical.
Let's say I was traveling to see my in-laws in Arizona and we had young kids. We would maybe buy diapers when we were down there, or my in-laws would buy diapers once or twice during Christmas. Understanding that that was an anomaly in the data is critical, such that you don't start giving offers for baby food, or carry that product in the store at a significant rate. Understanding consumers longitudinally is vital for every aspect of the supply chain, for suppliers and for retailers.
From a pricing standpoint, understanding where you cannot take price increases is so critical. Over the last five years or so in the U.S., private label has become a massive growth engine for many retailers. If you look at new entrants into the U.S. marketplace, most of them are private labeled—80% to 90%. The American consumer is beginning to see there is incredible quality and value there for them. One of the things many CPGs are working with us on is trying to understand how to offer better quality product to compete with private label right now, especially as the economy struggles.
How do you take the data that is collected and use it? How do you make it into something actionable?
Let's pick something like the toilet paper shortage we had during Covid. We're likely to run into some of those shortages again now. The reason it's so important not to look at just your own data, but market share data in general, is because as the chief merchant, I could quickly tell through the market measurement data that supply was going to competitors. If I was losing share in an environment where I'm out of stock every other day, and I'm losing share very quickly, I can pick up the phone to various CPGs and say, 'Hey, why am I out of stock and losing share? Am I getting my fair share?'
From a pricing standpoint, there's a massive misnomer in the industry that an item has elasticity. (He holds up a pen) Everyone believes this pen has an elasticity. That's not true. This pen has an elasticity because we have bought this item. Each one of us have a different elasticity to this pen. The pen's elasticity is a regression to the mean from each of us buying it. How do we group customers together and create pricing segments, store pricing processes? Is it only price-sensitive people buying the product? From a pricing standpoint, it has massive implications.
In retail, people talked about price gouging for years, though Covid margins were still [at industry averages] between 1% and 3%. That didn't change; it was genuinely upstream. How can we help find ways to reduce costs upstream as well? Data helps in that endeavor, too.
From an assortment standpoint, data helps massively as well. We see a massive trend coming in proteins. You have many retailers working on protein-enhanced products. It's going to help the GLP-1 lifestyle, but it's going to help much more than that. The aging population needs more protein. Protein is a great long-term attribute. You're going to see this big onslaught, but it's going to have longevity that you might not have seen in some other spaces.
Being able to look at these trends, understand these trends, and then share it with our partners is how we use the data. If you think about using data, it's across all four Ps of marketing: pricing, product placement in store, and the product itself from an innovation perspective.
How has this space changed in the last couple of years with the rise of generative AI?
I have a computer science degree. In fact, my first job was teaching neural nets and rules-based engines in the '90s, before compute was ever even a figment of our imagination.
Today, the engine that drives AI is data. The notion of garbage-in-garbage-out is very true in an AI world. You think about the data set we sit on—$5.8 trillion of annual consumer spending—and being able to look at that data longitudinally over time. We have 60 years of data and purchase behavior. Sit an AI engine on top of that, and we are truly at a precipice, a point in time that's going to fundamentally change the space over the next five, 10 years. You look at the number of analysts companies have, the number of merchants people have, the elasticity of this pen, the complexity of trying to understand that, and then drive that down to 2,500 stores and every store having 50,000 SKUs. How can any human possibly manage that really well? Machines will.
Think about five years from now and where we will be. You're conversing with an AI agent to better understand behavior, to the point where it's going to be questioned whether AI will actually come back with recommendations based off of what it is seeing for the individual to go and execute. That's the future. And that's exactly why I'm here. Too many companies are talking about AI and how it's going to do this, that, and the next thing.
[A famous quote says] data is the new oil. It's the new oil for AI. It's what I'm so excited about. The data set we have isn't just point-of-sale data, it's also panel data such that we can see across retailers what the consumer is doing. The best part is our survey data, which is the why behind the behavior. We can go and ask those consumers why they are switching between various retailers, between various brands, et cetera. For me, that is probably the most exciting part of our industry right now: How much change is about to be upon it.
How does all of that help in this moment, with the tariffs starting to come online and still a lot of uncertainty?
Just this week, I've had five retailer phone calls, four CPG phone calls, with COOs or CEOs all asking the same question: How can you help me right now? One of the CPGs asked me to talk to their board because [they are] wanting to know what is happening and be on the forefront of this.
With the data assets we have, and the fact that CPGs and retailers have gone through much of this before, there are pivots and moves we can all make in the industry to adapt to what is happening in the macro environment. The pivots in the supply chain that were made previously—such that you didn't have a single-source supply—exist today. How you leverage that and pivot to that quickly is going to be critical from a pricing standpoint.
Using the data to understand switching behavior. Knowing that price is going to increase this much means my demand is going to go down by Y. When it goes down by Y, I only need that much supply. Don't oversupply me. If you increase your prices dramatically [and] keep your supply the same amount, that's going to be an issue for you.
Another example is then saying, okay, we believe customers will switch to these two commodities. And if those two commodities aren't as impacted by tariffs, then up your demand on those. We have a demand forecasting tool that has never been in more demand than it is right now as they test these scenarios. Think about scenario testing through our demand forecasting system: Being able to better understand what products are going up. Scenario that. Where would they switch to? Scenario plan that.
Here's what I believe you'll see: Less promotions happening because of the volatility in supply, the volatility of cost. You're going to see that on the perimeter of the store first—vegetables, fruits, et cetera—and then you'll start seeing that migrate into the center of the store. The key is understanding what's going to have tariffs, what's not, where's the demand going to shift, and how do you make sure you have enough supply.
What kind of advice would you give to any business that has something that they are selling now, at retail or not?
It's such an overused word, but be as agile, as nimble as possible, and surround yourself with as much data as possible, such that you're looking at it from every angle. Don't be looking at just your own data. That's the biggest mistake I see companies making. Bring in as much external data, so you better understand the environment that's going on around you.