Inside Data at Wolt: How Data Science and Applied Science Work Together

Data powers nearly everything at Wolt, from what you see in the app to how fast your order arrives. Two core teams bring this to life in very different ways.

The Data Science team focuses on guiding decision-making. They run experiments, analyze user and operational data, and deliver insights that help teams across Wolt prioritize and act smarter. The Applied Science team, on the other hand, builds the machine learning systems, AI, and other algorithms like optimization that power our real-time experiences. These experiences can include predicting delivery times or personalizing recommendations. Together, they bridge analytics and AI to power a platform used by millions across 30+ countries.

Getting food or groceries delivered might look simple, but under the hood, it’s powered by a complex ecosystem of data, decisions, and intelligent systems. From forecasting demand across dozens of markets to tweaking product flows that improve retention, data helps us move faster, scale smarter, and deliver better experiences for customers, courier partners, and merchants alike.

In this blog, we’ll take you behind the scenes to explore how these two teams work, the problems they solve, how they collaborate, and the impact they make across Wolt’s products and operations.

Data Science

The Data Science team at Wolt plays a crucial role in helping us understand and improve our products and operations. We turn complex business problems into clear, data-driven questions and dive into patterns, behaviors, and trends to uncover insights that matter.

Our team includes data scientists, product analysts, and business analysts, each bringing a different perspective to the table:

• Data Scientists build models to solve complex problems that power product features.

• Product Analysts study user behavior and product performance to inform design and development.

• Business Analysts spot trends and opportunities that help guide Wolt’s strategic and operational decisions.

We’re embedded across teams in Product, Engineering, Marketing, Strategy, and Operations. Whether it’s uncovering the reasons behind user churn or testing the impact of a product change, our goal is always the same: help Wolt move faster, focus on what matters, and improve the experience for everyone who uses our platform.

Here are a few ways our work shows up in the real world:

• Driving growth and engagement: We continually look for opportunities to keep users engaged with the platform. For example, we tested alternative verification methods during sign-ups, which led to improved sign-up rates and conversions of new users.

• Service efficiency: Our machine learning models help suggest more accurate prep times to merchant partners and guide courier partners to demand hotspots, reducing travel time and improving fulfillment.

• Product insight: We identify parts of the customer journey where users drop off and work with product and design to smooth out those flows.

• User segmentation: Using our spatial mapping platform, we plot and analyze customer and venue data to provide valuable context for business decisions.

And while we love numbers, we’re not just about models and metrics. Our job is also about making sense of the noise. We tell stories using data in a way that brings clarity and helps others move forward with confidence, especially important when decisions affect millions of users.

How the Data Science Team Works

We start by clearly defining the problem. Then, we zoom out to understand the context, dive into the data, and work backwards towards the solution. We use experimentation to validate changes and ensure what we do actually moves the needle.

We also build frameworks and dashboards that let other teams answer questions faster on their own. Empowerment is key. We want our insights to scale.

Applied Science

The Applied Science team at Wolt brings machine learning, optimization, and AI to life in our products. We sit at the intersection of research, engineering, and product. Our models run in real-time, production environments serving millions of users across diverse markets.

Our team includes Applied Scientists, Machine Learning Engineers, and Applied Science Leads:

• Applied Scientists work end to end designing and implementing algorithms into production.

• Machine Learning Engineers work closely with applied scientists on ML related engineering tasks, scaling our deployments.

• Applied Science Leads guide and coordinate applied science efforts while staying hands-on with day-to-day team work.

We're embedded in cross-functional product teams and collaborate closely with PMs, engineers, and designers to turn ideas into scalable systems. From predicting delivery times to personalizing what users see in the app, we help shape both the user experience and the operational efficiency of our platform.

Here are a few ways our work shows up in the product experience:

• Courier assignment: We solve optimization problems to match the right courier to the right order, improving delivery times and efficiency.

• Real-time predictions: Our models generate over 57 billion delivery time estimates daily, each one directly

  impacting user experience.

• Personalization: Our real-time recommendation systems account for intent, location, and context to deliver a personalized experience on the Wolt app for each visitor.

Demand forecasting: We build predictive models to forecast demand and automate inventory replenishment in our Wolt Market stores.

How the Applied Science Team Works

We take end-to-end ownership, from prototyping and training models to deploying and maintaining them in production. Every decision our models make affects millions of users in real time, so we design our systems for speed and reliability at scale.

We also collaborate on experimentation efforts, helping teams evaluate the real-world impact of our solutions through rigorous A/B testing and data analysis. Our dedicated machine learning platform accelerates our workflows and lets us focus on solving complex problems, not managing infrastructure.

How do Data Science and Applied Science teams collaborate?

One of our most important collaborations has been improving Estimated Delivery Times (EDT). Here’s how we tackled it together:

• Data Science analyzed historical delivery data and built dashboards to pinpoint where our previous estimates were failing and what was driving the inaccuracies.

• Applied Science developed a new machine learning model that predicts delivery times more accurately and collaborated with engineering to integrate it into production.

• Together, we designed and ran experiments to evaluate the real-world business impact of the new model.

The outcome? More accurate ETAs for customers, fewer support tickets, and better alignment between what we promise and what we deliver.

Summary of key differences

Here’s a quick overview of how the two teams differ, and where they meet:

Conclusion

At Wolt, we’re solving meaningful problems with real-time impact. From the moment you open the app to the moment something arrives at your door, data plays a role in making that experience seamless.

Data Scientists help us make smarter decisions, faster. They dig into the numbers to figure out what’s going on, why it matters, and how we can do better. Applied Scientists build the intelligent systems that power our platform. They work on machine learning, AI, and optimization, turning cutting-edge research into real, scalable solutions. 

If you’re passionate about solving problems with data, collaborating with kind and sharp colleagues, and scaling impact across borders, we’d love to hear from you.

👉 Check out our open roles in Data Science and Applied Science.