Home

Case studies

The first modern real estate platform

QA for real estate platform

DeviQA's team takes full responsibility for testing a real estate solution consisting of a web app and mobile apps.

2k+

Test cases added

85%

Test coverage

1.6k+

Automation test scripts created

80%

Automation test coverage

2.5x

Faster time of regression testing run

800+

Bugs reported

400+

Bugs catched by automated tests

4k+

Feature tickets verified

About project

Compass has been developing a modern real estate platform that uses cutting-edge technology to create an intelligent and smooth experience for buying, selling, or renting real estate. It also features a web-based customer relationship management tool with a user-friendly interface that helps real estate agencies efficiently oversee and manage diverse communication activities.

Before DeviQA

2.5 weeks to run regression testing
7 days to run regression testing
40+ minutes to run automated E2E testing on each pull request
50% of the application was covered with test cases
40% automation E2E testing coverage
Global API endpoints were not covered by automation
No load test scripts
No mobile automated tests

With DeviQA

2 days to run regression testing
10 minutes to run automated E2E testing on each pull request
85% of the application is covered with test cases
80% automation E2E testing coverage
500+ API auto-tests created for multiple endpoints
Load testing scripts are executed simulating behavior of up to 1000 concurrent users
200+ mobile test scenarios have been added
4 test suites created from scratch for multiple newly developed features
Integration of automated suites with multiple third-party services to increase application coverage

Our contribution

Team

8 full-stack QA engineers, 1 manual QA engineer

Project length

Since 2016

Technologies and tools

Jira

Confluence

Twilio

Datadog

TestRail

Postman

PostgreSQL

GitHub

JavaScript

TypeScript

Cypress

Cypress Dashboard

Supertest

Mocha

Python

Locust

CircleCI

Our engagement

On this project, each of our QA engineers works independently to support their own development team and collaborates with a global QA team when needed. The project is extensive, and multiple development teams focus on their own specific tasks. DeviQA’s mission is to support these teams by providing comprehensive testing solutions. Our QA engineers actively participate in Business Requirement Specification (BRS) discussions and design testing strategies at the early stages of software development. All team members have the opportunity to contribute to feature implementation to ensure high-quality testing.

Throughout our work on this project, we have implemented and maintained various solutions, including the following ones:

1. Configuration of a CI/CD tool (CircleCI) to automate test execution for specific microservices upon PR creation, schedule regression and smoke tests, and notify the team in Slack about job results, while also providing detailed statistical reports.

2. Execution of integration tests covering our application and third-party services, such as Twilio Flex, Gmail, Google Sheets, and Google Voice.

3. Development of a custom framework used to execute API tests for services.

4. Automation of UI tests that simulate real user interactions with the system.

5. Manual testing of iOS and Android mobile applications.

All testing solutions for those parts of the application that we are responsible for are developed and maintained exclusively by our team members.

Services provided

E2E automation testing

The DeviQA team developed over 1600 automation test scripts for smoke and regression testing. These tests were integrated into a continuous integration process, with most of them being executed against new PRs for the respective microservice. Approximately 80% of the application was covered by automation testing scripts, which led to a substantial acceleration of the regression testing process.

API testing

We established and maintained an automated API test suite using a custom framework based on Mocha and Supertest. This suite consisted of over 500 automated API tests, covering various endpoints and scenarios for endpoint usage.

Load testing

Using the Locust framework and closely collaborating with developers, we wrote Python scripts and implemented load testing during deployments to test environments. These tests simulated the real-world load on the test environments during deployments. This approach helped us identify and address deployment-related issues that typically reveal themselves only in production due to the high load from real clients.

Distributed QA team

Our QA engineers were distributed across several development teams and guided by Dev Leads. Each QA engineer had their own specific feature and test suite to focus on and was seamlessly integrated into the development process and discussions.

Web/mobile testing

Our team is in charge of verifying new features, and we've successfully tested and confirmed over 4000 Jira tickets. In addition, we conduct testing as a part of regression execution. We also provide native app-specific smoke and regression testing.