The work of data teams often ends in a presentation: slides showing key findings, important takeaways, and suggested next steps. But even the best decks tend to gloss over the days (or weeks, or months, etc.) spent transforming, analyzing, and generally agonizing over data.
And for no-code website development platform Webflow, those same decks left a lot to be desired when it came to collaboration — both inside and outside its data team.
Learn how Webflow used Deepnote to upgrade how its data team worked together and shared its results.
Bridging the collaboration gap
Webflow’s aim is to build the most powerful no-code development platform in the world. The company currently helps more than 3.5 million customers build and host their websites using a completely visual canvas.
Part of Senior Manager of Data Science & Analytics Allie Russell’s job was to determine how the company could use its data to deliver more value to the company and its customers.
Webflow’s first step was implementing Snowflake as its data warehouse, where it collected and centralized all raw data to power various use cases, such as customer segmentation. The company’s data team partnered closely with its engineering, product, and growth teams to experiment and uncover impactful insights.
“We really wanted to figure out how our customers are using our product at a greater scale,” Russell said. “Our goal is to find ways that we can use data to bring value. We built up the source of truth first and are now building out new applications for using data.”
But centralizing data in a warehouse wasn’t enough. Russell and her team struggled to work together on code and share their findings with each other. They needed a place to collaborate, as well as a way to give company leaders and cross-functional partners more visibility into Webflow’s growing data science function.
Data notebooks were essential to the team’s day-to-day work, but with traditional tools, collaboration meant having to share individual GitHub repositories. Team members were operating in their own personal environments, and because there was no easy way to work together, they tended to work in silos.
“No one actively collaborated on their code,” Russell said.
The team needed a collaborative notebook — one that would not only integrate with Snowflake, but also allow everyone to work together, in the language they’re most comfortable with.
Deepnote checked all the boxes.
Unlocking the value of collaborative notebooks
Russell’s expectations were initially low.
“I was personally skeptical about the performance of a collaborative, hosted platform, but when I used Deepnote to collaborate with a colleague and we were able to remotely troubleshoot and try different plots, I immediately saw its value,” Russell said.
Beyond real-time and asynchronous collaboration features, Deepnote’s workspaces make it easy to organize different projects, notebooks, and apps all in one place, allowing Russell and her team to easily jump into one another’s work.
“Currently, I have a team member on leave,” Russell said. “Deepnote allows me to look into her work without having to understand her environment or running into a bunch of errors, and that’s pretty powerful.”
See Deepnote’s workspaces in action in the video below:
And since Deepnote supports all programming languages — including Python, SQL, R, Julia, and more — team members aren’t forced to change their preferred working styles.
“Using Deepnote, our team can conduct analyses in the language most comfortable for them, which makes it accessible to and powerful for everyone,” Russel said.
Deepnote not only gives Russell and her team a way to explore and collaborate on data together, but it’s also opened the door to helping less technical colleagues understand the value of their work.
“There’s Snowflake and there’s our business intelligence tool — which shows the final version of the project — but until Deepnote, there wasn’t anything in between,” Russell said. “Deepnote enables us to bring people into the phase of data science that’s all about experimentation, helps them understand our processes, and encourages folks to leverage data science in even more ways.”
Teammates collaborate on notebooks both inside and outside the data science function, often working together in real time. This includes:
Running statistics
Team members run different types of statistics and make them available to other teams to inform decision-making.
“We use this a lot for experimentation,” Russell said. “Data flows in and gets modeled for experimentation, then we make it interactive and available for product managers and data scientists to use.”
Prototyping
Team members create minimal viable products for different use cases, such as SEO experiments for the marketing team.
“Deepnote lowers the barrier to entry to prototype,” Russell said. “There’s nothing that filled the space in our stack that allowed us to do an analysis and share it in a repeatable way until we adopted it.”
Sharing metrics
Team members build notebooks for colleagues that allow them to see and explore metrics relevant to their domains, such as user activation metrics for product managers.
“The barrier to entry would have been a lot higher if we didn’t have this shared environment,” Russell said. “Using Deepnote, product managers can run their own notebooks and modify different parts of it.”
Making buy-in better
Russell and her team used to summarize their work for different stakeholders in slides, but this created a disconnect between the presentation and what actually went on behind the scenes. With Deepnote, the team is able to help cross-functional partners understand the value of data science, as well as the processes behind it.
“Using Deepnote helps our team bring people into the phase of data science that’s usually unseen — the exploratory data analysis phase — and drive an understanding of the effort and work that goes into the data science process,” Russel said. “Most other tools are made for the final version of a product — they skip past intermediate steps where feedback and buy-in are critical. To be able to bring people along with the data work, especially remotely, is hugely valuable.”
This helps the team de-risk ideas and prove they’re worth further investment, making it easier to get buy-in on the infrastructure work that’s so critical to data science.
Even better, the cross-functional visibility and collaboration Deepnote enables ensures that teams at Webflow are excited about what’s being built and ready to put them into practice, reducing the time to market and helping Webflow deliver better products, faster.
Learn more about Deepnote’s collaboration features
Get started for free to see how Deepnote improves data collaboration for every team member.