SAfe overview
Blog

How our QA team scaled Agile: SAFe overview

Read how our QA team had to replace Scrum principles with those of SAFe and how we managed to achieve win-win results.
17 March 2017
Agile
Quality assurance
Black-and-white headshot of a man in business attire for author profile
Article by Vitaly Prus
Head of testing department at a1qa

Read how our QA team had to replace Scrum principles with those of SAFe (Scaled Agile Framework) upon the customer’s requests and how we managed to achieve win-win results.

The a1qa acquaintance with SAFe started two years ago. A customer, who has been cooperating with us in QA outsourcing for more than 4 years, confronted us with the fact: “As of today, we start working with SAFe, guys.”

However, we had no other options but to obey. We learned the basics, read numerous hands-on articles and started tuning our processes. Looking ahead, we must admit that we did it quite successfully.

To date, we’ve been working with SAFe for almost two years. And our experience is what we want to share. Hopefully, we’ll foretell some of the difficulties that your QA team may come across.

Let’s start with some basics.

When apply SAFe?

Google displayed us the following big picture:

Agile: SAFe methodology
SAFe 5.0 big picture diagram showing portfolio, large solution, essential, and team levels with roles, backlogs, value streams, Agile Release Train, DevOps, and continuous delivery, illustrating when SAFe fits complex enterprise delivery and how QA must coordinate across multiple teams

The core thesis about SAFe is that it contains a number of rules and regulations to ensure smooth agile scaling in a large number of development teams.

  • Are there multiple software products developed and roles in the project?
  • Do you need dozens of approvals to put in place any new suggestion?
  • Are there many development teams who are eager to apply the lean-agile approach?

If the answers are “yes” than SAFe is what can be of help.

What are the levels in SAFe?

SAFe basic structure contains three or four levels depending on the specific needs of the company. They are portfolio, program and team levels. The 4.0 SAFe that has arrived this January has introduced the fourth optional level, called the Value Stream that should be applied in companies with over 120 people working on heavy software systems. In teams of 50-125 people (as it was in our case) the three-level structure is more convenient.

To gain a better understanding of every level, let’s take an example of the project and focus on its levels. It’s important to note that the levels will be analyzed from the QA team’s point of view, not the business one.

Let’s imagine that we have to enhance the e-commerce application aimed at selling books. The current functionality of the app is very simple: the customer uses the search bar to find the book he or she needs and orders it. When the book arrives at the warehouse, manager notifies the customer and he/she collects it from the warehouse. Apparently, the app is too simple to be competitive on the market. We need to strengthen it adding some more features that will make it more attractive.

The portfolio level is the highest level of concern in SAFe. The responsibility of the portfolio level is to discover major initiatives (business epics) that would reflect business priorities. Epics can be functional and architectural by nature.

As for our website, the functional epics shall be the following:

  1. Organize delivery across the country with the opportunity to follow order processing in a user’s account.
  2. Develop an online communication platform for books amateurs.

Architecture epics should be the following:

  1. Integration with GIS (geographic information system).
  2. System migration to the cloud.
Example of SAFe number one
Diagram of SAFe levels with three rows labeled Portfolio, Program, and Team, showing four portfolio epics above multiple smaller purple work items to illustrate how high-level initiatives are decomposed into program features and team tasks

So far it’s pretty easy, we hope. Now we are going down to the program level.

Program level is where business epics are split into features and development team and other resources work on the implementation of the features. A feature is a part of the product that should give some flow of value to the customer or to business.

Let’s split one of the epics on the team’s level. Online communication platform epic can be subdivided into the following features: user’s account, forum, and private messages.

Example of SAFe number two
Diagram of a SAFe hierarchy with Portfolio, Program, and Team levels, showing how the Online communication platform epic is broken into features such as user’s profile, private messages, and forum, then further into multiple team-level work items

On the team level, every feature shall be reformulated into clear and short user stories that can be estimated and implemented within a sprint.

For instance, private messages feature can be broken down into the following user stories: send messages and get messages and save chat history options. Architecture epics have to be broken down as well.

Example of SAFe number three
Diagram showing SAFe work decomposition from portfolio, program, and team levels: epics like delivery across Belarus, CIS integration, and cloud migration break into features such as online communication, then user stories for profiles, forums, private messages, sending, getting, and chat history

In such a manner, with every next level, the tasks get smaller in size and their boarder lines are specified. Estimation also gets more accurate.

In the next article, you’ll learn the workflow specifics on each of the SAFe levels.

More Posts

14 September 2026,
by Pavel Novik
6 min read
A well-thought-out QA strategy for launching gambling software across the globe
Discover crucial components of an effective digital casino game QA approach for achieving global reach with your gambling software.
Functional testing
Mobile app testing
Performance testing
QA in Gaming
Quality assurance
Managing software quality and risks across multiple vendors
4 September 2026,
by AI Engineering Lab
5 min read
When every vendor passes but the product still fails
Individual vendors can meet their targets while the end-to-end service still leaves customers with a broken experience. Strong multi-vendor QA governance connects supplier commitments, testing, risk, and evidence around the end-to-end user journey.
Quality assurance
Test automation
Quality assurance team supporting regulatory compliance
21 August 2026,
by a1qa
7 min read
From audit readiness to continuous compliance: how QA enables regulatory resilience in 2026 
Software, infrastructure and regulatory requirements can all change between formal audits. Learn how QA helps teams retest affected controls, preserve audit evidence, and identify compliance gaps during software delivery.
QA consulting
Quality assurance
How OTA update testing helps prevent large-scale IoT device failures
6 August 2026,
by a1qa
5 min read
OTA update testing: how one bad release can disrupt an entire IoT fleet
OTA lets companies repair thousands of connected devices remotely. When the update fails, that convenience disappears fast. QA determines whether devices recover on their own or wait for a technician.
IoT testing
Quality assurance
How scalable QA keeps super app platforms reliable and secure
16 July 2026,
by Automation Lab
6 min read
Super app testing: How QA keeps complex platforms reliable
Super apps bring payments, shopping, bookings, messaging, and support into one place. Explore how scalable QA keeps every service, integration, and customer journey working together.
Mobile app testing
Quality assurance
Test automation
How software testing helps increase customer retention and user satisfaction
11 June 2026,
by Lena Yakimova
6 min read
Why quality assurance is critical for customer experience and customer retention
Even small product issues can push customers away. Comprehensive quality control ensures those moments never happen, making it one of the most powerful drivers of retention.
Quality assurance
Usability testing
How organizations successfully shift from QA to quality engineering
12 May 2026,
by Mike Urbanovich
6 min read
QA vs QE: How to transition to quality engineering in the age of AI
Quality is no longer something you test into a product. The organizations that understand this are already pulling ahead.
Functional testing
Quality assurance
Test automation
How to maintain high software quality while accelerating Agile delivery
5 May 2026,
by Vitaly Prus
5 min read
How to ensure high software quality that keeps up with Agile speed
Let’s dive into how project teams can combine modern QA practices to ensure that quality keeps pace with rapid development requirements.
Agile
Quality assurance
21 April 2026,
by Automation Lab
5 min read
Modern Selenium Grid alternatives: Playwright, Cypress, and Selenoid in 2026
Teams choosing an automation stack in 2026 need to weigh legacy code, infrastructure load, and the speed of feedback. This article helps them choose between Selenium Grid, Selenoid, Playwright, and Cypress.
Quality assurance
Test automation

Get in touch

Please fill in the required field.
Email address seems invalid.
Please fill in the required field.
We use cookies on our website to improve its functionality and to enhance your user experience. We also use cookies for analytics. If you continue to browse this website, we will assume you agree that we can place cookies on your device. For more details, please read our Privacy and Cookies Policy.