iFood Serves Up 54% More Checkouts with Error Message Makeover

Insight/Action/Outcome: The product discovery team at iFood, Brazil’s No.1 food delivery platform, used Amplitude Analytics to investigate checkout errors and find ways to fix and solve them. Optimizing the process allowed them to reduce errors by 25%, speed up problem resolution by 60%, and increase purchase completions by 54% — after encountering an issue.

As the top Brazilian food delivery platform, iFood handles a staggering 100 million orders monthly, and dominates over 80% of the country’s market. We deliver meals to millions of homes each day. But even with those numbers, we’re still looking to constantly improve operations—our goal is to feed the future. We believe access to food is a basic right, and seek to make it as easy and affordable as possible.

My role as a Product Manager is to help paint the bigger picture of our product performance and user behavior with data. For example, I’m the one looking at user journeys and considering how we’re structuring our analytics to understand them. Before I became a Product Manager, I already had experience using Amplitude, and I knew it was critical to gaining the visibility we needed.

A natural fit for business growth

Understanding what our customers want is key to our success, but our product discovery team’s old analytics strategy and supporting technology stack just wasn’t cutting it—it dated back to 2018 before major changes, including a few mergers, happened. The team kept trying to modify the software by adding or removing parameters to keep up, but that only increased the complexity of our operations. We kept running in circles trying to stay on top of evolving business needs, and still couldn’t find the answers we were looking for.

When our team received the green light to revamp our data approach, I didn’t have to think twice about switching to Amplitude. Our product discovery team is not an IT-exclusive team, but a mix of marketing experts, product engineers, and others who need quick, clear answers without having to navigate database queries. In this case, Amplitude Analytics suited us perfectly because it’s an intuitive and fast platform.

54% more customers completing their orders

One of the biggest discoveries we made with Analytics was transactional errors and their impact. These issues can cause users to abandon their carts in frustration and walk away from the whole purchase. They may occur when iFood’s platform tries to sync with a third-party app and include payment failures, order processing errors, or coupon and discount errors.

Our team jumped into Analytics dashboards to investigate these occurrences. We went through everything—from error messages that the users saw to how a customer got back on track after getting an error. To see which issues needed our attention first, we tagged errors based on their impact on user experience.

We stumbled upon a few intriguing discoveries right away. Some errors confused users with double messages and conflicting information; others needed a communication overhaul, and several had lost relevancy and should have been removed. But the turning point was identifying the problems that led to dead ends—errors that left users completely stuck, with no idea what to do next. Thanks to Amplitude dashboards, we caught and fixed those, which completely changed how we communicate with our users.

That was a game-changer for iFood. We cut down on errors, which made everything run smoother. Instead of telling users about the problems, we could just fix them. The results were crystal clear—and showed a steadily growing buyer conversion. The number of users who bounced back after hitting an error and completed their purchases soared by 54%. Meanwhile, the time needed to recover after an error and finish the order decreased by 60%!

25% error decrease to improve customer purchasing experience

Another way Amplitude helped us improve user experience was by solving non-transactional checkout errors—like address validation issues, or user interface glitches. We broke down every step of the user flow in Analytics, looking for ways to boost user experience and metrics. While digging into the data, we spotted a major problem: lots of checkout errors.

One major issue was that the app didn’t warn users if a merchant was about to close. Just imagine that it’s 10PM, and you’re exhausted, craving dinner and dessert. You look through the app, finally find something to eat, and right when you’re about to pay for your order, you get a message that the merchant is closed. I myself got this error a few times, which made it even more frustrating!

We initially didn’t consider how much this might be impacting our revenue, but it turned out we were losing a significant portion of customers who just gave up on making the purchase. When we spotted this in Analytics, we added a prompt to alert users about closing merchants, saving them from last-minute disappointment. Our merchant and store partners were happier, too, with fewer complaints and better reviews.

We implemented a similar solution to another error we noticed on Analytics. When customers browse through food and products for grocery shopping, they need to choose a delivery time slot before finalizing their purchase. But sometimes, their desired slot would become unavailable by the time they were about to process their payment.

Customers would get frustrated because they would then have to start their purchase over, sometimes with a cart of 20 to 30 items. They’d need to find another merchant, replace items, and go through the checkout process, only to potentially face the same problem again. This was especially frustrating for grocery orders and significantly impacted buyer conversion rates. Catching this issue and understanding the whole user journey helped us find a workaround.

Before starting the project, our team set a goal to reduce errors. By using Analytics, we mapped the entire user journey and found ways to improve, hitting a 25% reduction in checkout errors—exceeding our initial target by 5%!

Craving answers here and now

Hundreds of iFood team members are empowered by Amplitude—we have 619 monthly active users in Amplitude, and we use it daily to pull data in minutes. Each member is able to perform data analytics themselves, even without specialized IT skills. That is the autonomy we seek in using data, and that’s what Amplitude helps us achieve.

Now, so much more can happen in a shorter amount of time. Before Amplitude, if someone asked a question in a meeting, I’d have to say, ‘I’ll get back to you later.’ These days, I can pull up the data on the spot, making meetings more efficient and speeding up progress.

We have 14 projects going on at the moment, and we aim to use Amplitude to extract every bit of data value and make informed decisions on all of them. We’re even testing integration with Braze, a customer engagement platform, to help the marketing team better target prospects.

Amplitude’s support team has also proven to be a key part of our data journey at iFood. We have regular catch-ups with Luiz Ribeiro Amorim, our Amplitude Customer Success Manager, and Marco Wella in sales to explore how we can optimize Amplitude’s potential and solve issues better.

We’re also rolling out training on using Amplitude Experiment and cohort analysis. Looking ahead, I know we’ll continue making the most of Amplitude to empower everyone at the company. Data democratization is at the heart of everything we do, and we want to make sure teams are comfortable and creative with data analysis to tackle challenges. Our ultimate goal is to help everyone unlock Amplitude’s full potential and give them the tools to excel.