When thinking of innovations in data science, many people focus on massive companies like Facebook or Amazon. We know that Facebook leverages data from users to enable marketers to better target their advertisements. Google uses data science to auto-complete words when you’re typing into their search bar. But the reality is that data science and data-driven strategies are being used across modern business for a variety of purposes.
By combining knowledge and analysis of data with business acumen, modern companies can become experts in data science execution.
Some of the strongest data-driven companies are ones you may not associate data science with. These companies are successful in their data-led approaches because of a strong, collaborative relationship between data scientists and business leaders. By combining knowledge and analysis of data with business acumen, modern companies can become experts in data science execution. But this is not easy, and requires a shared understanding.
Fast food companies are often such data science experts. To stay true to their purpose of providing a consistent, cheap product at a fast pace, they must be attuned to inefficiencies within their process. Finding those inefficiencies within hundreds of thousands of franchise locations can be cumbersome without a robust data science-led strategy.
For example, American fast food restaurant chain Chick-fil-A estimates that about 30% of potential customers are driving away because of long lines at the drive-through windows. This inefficiency not only affects the volume of revenue the company can produce but also the quality of each customer’s experience.
It’s not the first time Chick-fil-A has taken business lessons from data. In 2019, the restaurant chain identified a small number of less ordered items that cashiers would have to go to the back of the store to retrieve. This led to longer wait times for customers and less efficient use of workers’ time. By using product-based data to identify these bottlenecks, Chick-fil-A’s data teams determined the optimal amount of these products to keep in small refrigerators underneath the front counter. This new recommended flow almost immediately showed improvements to speed of service, thus resulting in a better customer experience.
Both optimizations were made possible due to their team’s data-driven strategy. The changes led to more efficiency, more positive customer experiences, and thus higher revenue.
Similarly, ecommerce businesses thrive on positive customer experiences in lieu of traditional marketing at brick-and-mortar locations to contribute to their success. One of my favorite examples of ecommerce using data to optimize the customer experience is Stitch Fix.
Founded in 2011, online personal styling company Stitch Fix uses recommendation algorithms and data science to personalize clothing choices for women, men, and kids based on style preferences, size, and budget. They’re committed to their data-driven business model so much so that, when searched, Google labels the brand as a science company.
Due to the many different ways clothes can vary, it’s a difficult problem to solve. By using data science throughout their operation, they can pinpoint subtle traits of each customer to enhance their experience. For example, when clients first request a shipment, they can use data to match the client to warehouses depending on distance to the warehouse and how well the inventories in the warehouses match the client’s needs. The data from that initial request is funneled into intelligent machines which perform a variety of algorithms to eventually end in a curated box of fashionable finds shipped to the client.
The process is highly cyclical. Once a client receives their Stitch Fix box, the algorithms begin again. The client tries on the curated clothes and decides what to keep and purchase or what to return to Stitch Fix. When declaring their final plans for the clothes, the client provides feedback on why they decided to keep the item or why not. This qualitative data allows the algorithm to learn how to not only better serve the client next time, but also how to better serve other clients.
Similar to Chick-fil-A, Stitch Fix’s processes have led to more efficiency, more positive customer experiences, and thus higher revenue.
Oftentimes standard organizational processes and systems are so deeply baked in, it can be challenging to shift them to a data-first approach.
The benefits to incorporating data science into business strategy are clear. But the means to getting there aren’t always obvious. Oftentimes standard organizational processes and systems are so deeply baked in, it can be challenging to shift them to a data-first approach.
I’ve seen this directly in my work helping Harvard – essentially a very large company – think about its own digital transformation. There’s a strong interest in moving to digital systems, but there are so many processes that involve manual entry of data into spreadsheets with manual transformation of that data. Anytime new data comes in, the manual process restarts. Furthermore, these processes are baked in for staff and organizations at every level, which can slow the pace at which new systems can be successfully implemented.
Data science, like many areas of technological innovation, can often be posed as the premier solution to challenging problems like Harvard’s digital renaissance — and it is, to some extent. But what’s happened in recent years is the overpromise that data science alone will solve the problem. Companies too often silo off data scientists instead of thinking about an organizational-wide approach to data and digital success.