The Story Behind the Visualizations

Why We Chose This Topic

How a group of food‑obsessed students ended up exploring food delivery trends.

Everyone in our group loves food — cooking it, trying new restaurants, and following whatever the internet is obsessed with that week. Our original idea was to explore viral food trends, like the rise of matcha or the sudden popularity of TikTok recipes. We wanted to understand when these trends exploded, how big they became, and which platforms drove their popularity.

But after searching extensively, we couldn’t find datasets that captured viral food trends in a usable or consistent way. So we pivoted. Instead of looking at what people talk about, we shifted to what people actually do: ordering food. Food delivery is something almost everyone interacts with, and it reflects real behavior rather than hype. That pivot led us to explore the landscape of food delivery in Canada — how often people order, what they order, and how they feel about the platforms they use.

What Problem This Project Addresses

Helping people understand the food delivery landscape they use every day.

Food delivery apps are deeply woven into everyday life, especially in Canada. Many people order food regularly, but very few actually know what the broader food delivery landscape looks like. How often do people order? Which cuisines dominate different cities? How do platforms compare in sentiment and popularity?

Our goal was to create a website that informs people about these patterns in a way that’s both engaging and easy to explore. We wanted the experience to feel fun, surprising, and interactive — something that sparks curiosity rather than just presenting numbers.

Who This Project Is For

Designed for people who use food delivery apps — occasionally or often.

Our primary audience is adults who order food through delivery apps, whether occasionally or multiple times a week. These users already have firsthand experience with the platforms, so the insights we present are immediately relatable.

Because the visualizations focus on real behaviors — ordering patterns, cuisine preferences, platform sentiment — the content feels familiar and relevant. The interactive elements also make the experience more engaging for people who are curious about how their own habits compare to broader trends.

Limitations and What We Learned

Challenges we faced and the insights we gained along the way.

This project came with a number of unexpected challenges — some technical, some logistical, and some simply related to timing. One of the biggest hurdles had nothing to do with data or code at all: it was time. The most demanding week of this project overlapped with major final assignments in our other courses, including a client‑based project that all three of us were involved in. We knew what needed to get done, but there were only so many hours in the day.

Finding the right datasets was another major obstacle. At the beginning, most of the free and accessible datasets we found were based in Asia, which didn’t align with our goal of exploring food delivery in Canada. After extensive searching, we eventually found datasets that were not only relevant but also geographically centered in North America. No single dataset contained everything we needed, so we combined several sources to build a more complete picture.

Once we had the data, we ran into GitHub’s 100MB file size limit. Our combined datasets exceeded that limit, which meant we couldn’t push or pull them normally. To solve this, we implemented Git LFS (Large File Storage), which stores large files on a remote server while keeping lightweight pointers in the repository. This allowed us to keep everything in one place, but it introduced new complications: frequent conflicts, re‑tracking issues, and the need for constant coordination between team members.

On the technical side, converting Observable notebooks into standalone JavaScript was another significant challenge. Observable’s structure is powerful, but it didn’t match the design or layout we envisioned for our website. Rebuilding the visualizations in plain JavaScript required rethinking how data flowed through each chart and recreating interactions from scratch.

Despite all these challenges, the project taught us a tremendous amount — not just about food delivery, but about data storytelling, visualization design, collaboration, and problem‑solving. We learned how to adapt when things didn’t go as planned, how to work with imperfect real‑world data, and how to transform raw information into an engaging narrative experience.