Dhanda Stores

[Brinda]

UsingMicrosoft Azure, Azure AI and Azure Cognitive Services, the partnership between the two companies includes immediate and long-term initiatives to digitally transform the customer journey, optimize operations and deliver better products and services. Microsoft 365 is enabling both corporate and store employees today to benefit from a modern and secure work environment.

Moving forward, the two companies plan to work together to leverage Azure and Azure AI and other Microsoft Cognitive Services to transform the shopping experience by eliminating the friction customers experience at the grocery store, such as not finding the products they want, longer-than-usual waits at the deli or meat counters and, of course, checkout lines. For store managers and associates, these intelligent solutions can also help anticipate out-of-stocks and misplaced products.

The new solution, which is being piloted in 25 stores across five divisions, uses Microsoft Azure and geo-fencing to allow customers to complete almost all the tasks associated with paying for gas from within the comfort of their car. The pilot results have exceeded expectations with app downloads continually increasing. Albertsons Cos. plans to introduce One Touch Fuel to more of its fuel stations throughout 2019.

On the fifth episode of Enterprise Software Innovators, hosts Evan Reiser (Abnormal Security) and Saam Motamedi (Greylock Partners) talk with Anuj Dhanda, CIO of Albertsons Companies. As the second largest supermarket chain in the United States, Albertsons is responsible for helping feed much of the country. Anuj discusses the challenges Albertsons faced in the wake of the pandemic, the various next-generation technologies being used to drive the future of grocery, and the thought process behind when Albertsons partners with startups.

Evan: That makes sense. I think when people think about the most sophisticated machine learning organizations, right, they may not think about grocery. Do you mind sharing just kind of some of the ways that you guys apply things like machine learning to drive innovation and to deliver that better customer experience?

Anuj: Exactly. And also it's very localized. So if Oregon is having a heatwave, we might have implications for what we would sell there versus Texas might be different and that yet might be different than Boston or Chicago. So even the local football game makes a difference to the demand. Local weather conditions make a difference.

And of course, local preferences. Every aspect of how we run our business has an implication machine learning could improve. So we talked about demand forecasting and replenishment, but when we think about how we staff our stores, the slot management for e-commerce, how do we fulfill on those slots? How we better predict the specific needs of the customer and to be able to match the offers to them?

Now, in some cases, the lift is significant that you want to prioritize and make that investment. And in some cases, you'd say, hey, it's marginal. But we think there is a big opportunity as we mature both the - for machine learning as well as there is a higher level of base automation in the system.

Also the other thing unique about Albertsons is we have lots of butcher shops in our stores. So we - we provide a lot of these customized services to our customers. So that really helped. We took some restaurant supplies and we were able to convert it into a consumer supply, which if you didn't have a butcher shop, you wouldn't be able to do.

And the third was the businesses spiked. Every aspect of the business, people were just buying whatever they could buy. So we had to scale very significantly, like our e-commerce business scaled about two to three X order of magnitude within days or - and weeks.

Saam: So you mentioned your e-commerce business scaled an order of magnitude almost overnight. What were some of the challenges from a technology and infrastructure perspective that you faced?

Anuj: I think, given that our digital business was already in the cloud, that was a huge help. But really these - the bigger impact on scaling also was not just tech, but also the whole fulfillment of those orders. You know, somebody's got to go pick that order. We have to have labor to pick that, somebody's got to kind of put it in your car or deliver it.

Anuj: First thing I'd say is agility was enterprise, not just tech. Tech was a huge enabler. We were a big part of it, but it was really the enterprise. And in this agile journey one lesson you learn again and again is you gotta make enterprise decisions so business and tech and product have to be like one.

There is no air gap between the three, that showed up in spades. Second I would say is being in the cloud also paid very rich dividends. So like when we had to do the COVID vaccination solution, I think it was three weeks from start to finish, when the team said, this is what we want and when we had our first thing live, because we didn't have time to you know, do the usual process. And so we had to learn. My CEO says that we messed with the DNA. And it is true.

Saam: Absolutely. And it's one of these examples of an external kind of crisis - turning that into an opportunity and, and using that to transform the way you take risk and, and move with agility.

Anuj: Look, the grocery business across the country expanded. But what I will tell you, we are very proud of is in that time period, we also took market share, which suggested that we were doing things in a way and demonstrating to customers that we could supply their needs in a more customer-centric way as well as in a way that worked for them. So the lessons learned is both how we work internally, but how we kind of connect with the customer. And those are the things we are really investing behind in a very significant way.

Evan: So it seems like the way that people buy groceries has changed more in the last three years than in the previous three decades. I know Albertsons has partnerships with DoorDash and Uber - how are you thinking about e-commerce options when it comes to grocery?

And that has certainly increased at a very fast clip, but still, I think we have lots of room to grow. Our view is that customers will decide what works best for them. Some people like to buy their produce in the store and love the theater of looking at the produce, looking at the butcher block and seeing what they want.

And what we find is that even customers who shop online, many of them go to the store, too. If you give the optionality to the customer, can personalize this experience in a way that works for her, that would be the winning strategy for the customer and us.

Anuj: There are things that we are implementing and scaling and using. Then there are things that we're working on and then things we are incubating. The part on robotics, for example, I would say is in the middle column, in my mind there is bunch of robotics that we do in our distribution centers which look like almost like Star Wars.

But each one has to get connected to a customer need. And a way that we solve that particular need. So scan and pay may not work if you want to buy 50 items, it's maybe a little more onerous. For that we want, you know, a different type of solution.

Anuj: So I'm going to use a word which I generally never liked to use, which is to say it's a journey. The precision of the data is expanding very, very significantly. Because now for any particular product, you want to know a lot about the product, not just what it is. You want to know the attributes. You want to know the health attributes. You want to know the provenance of that product and that's all data, so when someone says, hey, do you have SKU-level data?

I think as we step up each function there are many unlocks and the efficacy of data continues to improve. We in fact, partnered with Snowflake early on. They've been a great partner as we are building this whole - our enterprise data platform which is message-enabled and then really lends itself for machine learning and modeling and training models.

Evan: So Anuj, how do you decide when to innovate internally, versus when to collaborate with other partners? Like, how do you strike the balance between, you know, when you buy, when you build, and when you partner?

Anuj: Our general thinking would be as follows: closest to the customer, we want to innovate and own and build that experience as much as we can. And the furthest away from customer, we would say buy, get scale and leverage technology to provide the most efficient, fast platform.

So if you were looking at let's say HR, we would not think about building our own - our own general ledger, but when it comes to the - close to the customer, the app, we would like to have our own, right? So that's the broad state. And then in between what we say is where are we leading the pack? Where would we differentiate?

Where would we see an opportunity to make a real difference? And then we would go sometimes and build. And even when we build, there are many, many tools, and this is again, another value of the cloud part is this whole platform as a service, that has changed the game. So that lots of services we can use, like even the cognitive services, we don't need to start from scratch and say, hey, can I build computer vision?

And we want to make sure that when we roll out something, it can scale and be secure and you know, all those things. So we would default to large companies. Fundamentally changed that. I think, in my opinion, at least, what has fundamentally changed is cloud because with cloud, scalability is, for a small company, not an issue.

Because scalability of a Azure is coming, or a GCP, is standing behind a startup. The security is not an issue because no, again, within reason we have to kind of make sure it is designed right. But those things, which used to be big concerns are not concerns as much. So that has really in a huge way, democratized how we access smaller companies or startup companies.

3a8082e126