Skip to content
MyFitnessPal

How MyFitnessPal built a future-ready QA process for 200 million users

Healthcare, life sciences & healthtech Quality engineering & GRC
w MyFitnessPal built a future-ready QA process for 200 million users

35%


80%


24/7

The challenge

Getting support for growth

MyFitnessPal—a worldwide known health and fitness app—after hitting the 200 million active user mark, faced a challenge that the existing system couldn’t support rapid growth. They needed a scalable QA infrastructure to handle multiple platforms, languages, and device types without sacrificing speed and quality.

The path to faster and more frequent releases of the app while maintaining high-quality standards required a robust testing strategy that would minimize the risk of software defects and ensure a stable, reliable app performance.

The approach

Maximum optimization and automation for high quality at scale

With over 200M+ users requiring special handling and care to maintain integrity, MyFitnessPal adopted a custom QA process called Continuous QA based on industry best practices, CI methodology, and development lifecycle specifics. To boost the development speed of automated tests, Solvd introduced their proprietary mobile test automation framework, Carina, which enabled 80% code reuse between iOS and Android test cases.

The next step was to significantly reduce the time and cost associated with testing on multiple device types. To do so, MyFitnessPal implemented a mobile testing cloud, which enabled the cases to run 24/7 on 30 mobile devices in the multithreaded grid. This solution included both software and hardware layers and provided the complete infrastructure required to execute the automated tests.

Technologies and partners

Tools we use

Selenium
TestNG
Appium
Jenkins
Docker
GoGrid Router
QPSinfra
Carina Framework

About client

MyFitnessPal: the all-in-one food, exercise and calorie tracker

MyFitnessPal is a popular health and fitness app that allows users to track their exercise, nutrition, and overall wellness. As of 2023, the app had over 200 million registered users worldwide, making it one of the most widely used fitness apps on the market today.

Industry

Healthcare, life sciences & healthtech

Headquarters

Austin, Texas

Founded

2005

Related customer stories

Government & public services AI & data engineering
Houston METRO: Making public transit smarter with data
Houston METRO, the fourth most populous metropolitan area’s public transportation authority in the United States, wanted…
AI search
Retail & consumer goods AI & data engineering
Finding the best fitting products with AI-powered search
An AI-powered search solution that helps customers find the best fitting products, including styles, placement and…
Healthcare, life sciences & healthtech Application development
Spire Health: building a remote patient monitoring device for breathing and pulse tracking
Spire Health was a lifestyle wellness device with a companion application that analyzed users’ breathing patterns to make one healthier and improve their mood….
Healthcare, life sciences & healthtech Application development
Mindgram: From idea to product through MVP
The bright idea behind Mindgram, a mental well-being platform with online courses and sessions with coaches,…
Retail & consumer goods AI & data engineering AI advisory
An AI innovations lab for retail CX
A joint AI innovations lab that lets a major U.S. apparel retailer prototype, validate, and ship…
manual qa testing services
Media, telecommunications & technology Quality engineering & GRC
Reddit: Supporting dynamic social media quality standards
Reddit, one of the most popular social media platforms in the United States, was focused on…
Healthcare, life sciences & healthtech AI & data engineering
Teaching a model to see structures inside cells
Joining an effort by biomedical research team to analyze small-sample brain cell data by building transformer-based,…
Retail & consumer goods AI & data engineering
Teaching a vision model to actually read a product photo
Solvd adapted open-weight vision-language models for e-commerce to deliver reliable, fine-grained product understanding—improving attribute extraction, multi-image…
generative user experience and design
Media, telecommunications & technology AI & data engineering Digital experience
Ashoka: Generative AI user experience and design for social impact ecosystems
Ashoka’s operations are based on an organization’s database of social entrepreneurs, consisting of contact information, backgrounds,…