Snowflake connected with Margaret Sherman of Sonos at Data Cloud Summit 2020 to hear how the company is using the Data Cloud to understand customer preferences and enhance listening experiences.

In a world where people are surrounded by a lot of noise, purity of sound in music and other content we seek out in the comfort of our homes can offer a welcome respite. There are lessons to be learned from a company reinventing home audio for today and tomorrow—and using the Data Cloud to do it.

Sonos is a sound-experience company that pioneered multiroom wireless audio and elevated listening experiences to “Amazing!” Understanding how people listen, what their preferences are, and how to deliver the ultimate experience is a journey for Sonos, and data analytics is central to the process.

Margaret Sherman, Head of Data Strategy for Sonos, leads a team of engineers responsible for the data platform and analytics engineering team that supports data insights across the company.

“Our mission is to inspire the world to listen better. Everything we do is in service to that, and data is a fundamental part of it. Data helps us build better products and a better business.”

—Margaret Sherman, Head of Data Strategy, Sonos


Sonos collects very large volumes of IoT data, including structured and unstructured data, from more than 10 million homes worldwide. This data is collected from speakers, applications, and the cloud, and it doesn’t always fit well within a traditional data warehouse.

Before Sonos implemented Snowflake’s platform, extracting meaningful data required a lot of effort and engineering time. “If we wanted to answer a single question about what was happening with the customer experience, we would need to have data engineers write code and spin up clusters,” Sherman explains.

Sonos also found that its large data queries were too much for its SQL systems to handle. SQL might process the data for many hours or time out altogether. Then, it could take weeks to extract the data and get it into a shape where analysts and scientists could work with it.

Today, the same processing takes only a few minutes with Snowflake. “When I saw the time comparisons, I almost fell out of my chair,” Sherman says.


Now, Sherman’s team can serve up more insights that ultimately help Sonos move faster as a company. Sherman says Sonos has significantly improved and accelerated the delivery of data analytics with the Snowflake Data Cloud.  “Our data engineers love it because they’re not sitting around waiting for jobs to finish. They get through their code much faster.”

“Snowflake just crunches through data. We reduced data extraction from weeks to hours, and the business can now act on information very quickly.”

—Margaret Sherman, Head of Data Strategy, Sonos

Sherman also notes that in addition to the power, scale, and performance of Snowflake, it comes at the right price. “While we’re pushing tons of data to Snowflake , we’re paying just for processing costs. This has helped us save thousands of dollars a month this year,” she says.


Data analytics help Sonos deliver a better customer experience in very tangible ways.

“We have to look beyond building products that provide superior sound and deliver products that come with ease of use and freedom of choice,” Sherman explains. “Our customers have a lot going on while they are using our products. Often they are moving between third-party systems and services. It is important we consider this in creating an amazing end-to-end experience. Everything needs to just work.”

With Snowflake, Sonos can gather different kinds of telemetry data to understand where and why reliability breakdowns happen, so it can rapidly address and reduce the number of issues.

“Having all data in one place in the Data Cloud is key to giving us velocity in this process, as is the ability to dig deeper,” says Sherman. 

The dashboards we set up with Snowflake are impressive. We can identify problems we didn’t even know we had.”

—Margaret Sherman, Head of Data Strategy, Sonos

Sherman also appreciates the flexibility Snowflake offers. “We use the VARIANT column quite a bit. It allows us to pull in semi-structured data and work with it in different ways. We don’t need to have a really complex pipeline upstream before pulling the data into Snowflake.”

Looking ahead, Sherman sees the relationship between Sonos and Snowflake continuing to grow and expand. “We really want Snowflake to be the center of our data platform and bring all of our data together. Essentially, we want to live the data dream,” she concludes.

 Watch Snowflake’s Data Cloud Summit 2020 fireside chat with Margaret Sherman.