Mark Schwarz: It was very clear to the corporate owners and to a select set of stores that they needed to change their transaction prices and make them lower, and that they need to raise operational efficiency to compete with the third-party oil change providers. They work with us to build new ways of looking at their business.
It helps them collect data they never had before. They can cluster those in ways that are novel and more actionable. Then they can roll them into a set of next steps that are reasonable for both the stores and the corporate staff. As I said, in the middle, we have field managers. I see them as analysts. They are experts at dealing with data on a store-by-store level.
We’re highly concerned about what they think about how the business is evolving and how it can be better. Our job is to augment their efforts. We can give them the data they need to make a change. We can cluster and segment it in ways that are helpful. Whether this particular effort is working or not working, we analyze that data to give back to the corporate staff and help evolve the programs so they can change their business faster.
Sramana Mitra: Can you give me an example of some actions that people are taking on the basis of the kind of analytics that you are putting together for them?
Mark Schwarz: A few of the actions are in the express service businesses. In this business, one of the most common actions is just simply socializing the need for change. We have a set of related metrics and a dashboard that help them see if they are underperforming compared to their peers and have frank conversations about what that means. Those typically take several cycles in the field.
In one example, we have a set of reports that show on six different axes how they are underperforming compared to their peers, and allow them to prioritize the first one or two that they will choose to take action with. Thirdly, once those first two steps have been taken across a large number of dealerships, we give the corporate staff what they need to improve on some parts of their business such as the training that they offer or the amount that they charge for some of their services so that they can encourage more and more clear action.
Sramana Mitra: Let’s switch to the next question. When you look around the data needs whether it’s in auto or retail in general, what are the key trends?
Mark Schwarz: There are two key trends that I’ll speak about. One is the frustration with the deluge of data they are given. Many retail businesses also have more data than they can reasonably deal with. That’s a really important part of what the Data Science team at Square Root does. The second one is the inability to enact change across an organization. It’s one thing to have the right signal in a particular metric and be able to change what’s in your purview. It’s a totally different thing to be able to roll out what you like to change about your business to your management and their management. That’s even more difficult to solve.
Sramana Mitra: You see these because of your familiarity with the space. What are some open problems that you see out there that you would recommend other entrepreneurs to look into solving?
Mark Schwarz: While we’re focused on the retail space specifically, one of the core things that we talk about a lot is the ability to drive action from data. A lot of entrepreneurs are focused on Big Data problems specifically and speak less of what exactly will change in a business because of the data. The way that we address that question in the retail space is through a concept we call Observe-Orient-Decide-Act (OODA) loop.
One of the things that other entrepreneurs could focus on is framing actions out of data in a corporate or a non-profit space. While we’re focused on retail vertical in particular, the ability to distribute metrics to a group, and then collect analyses and enact change to the business overall is something that many businesses could do much better at. It starts with Big Data and ends with actionability.
Sramana Mitra: What you’re pointing out is that workflows is different in different businesses. To be effective, you need to understand those domain specific workflows. There’s plenty of work to do in understanding those and creating data-driven actionable paths through those.
Mark Schwarz: Exactly.
Sramana Mitra: Thank you for your time.
This segment is part 2 in the series : Thought Leaders in Big Data: Mark Schwarz, VP of Data Sciences at Square Root