Dedicated Teams

The Spotify Agile Model Explained: Principles & Takeaways

With 345 million users, Spotify is a disruptive force in the music scene. Launched in 2008, the platform has grown to become the world’s largest audio streaming subscription service. 

But Spotify is also creating quite the buzz in the world of engineering, particularly in the agile transformation space. The Spotify agile model came into the limelight when Anders Ivarsson and Henrik Kniberg, Agile coaches, released their whitepaper Scaling Agile @Spotify in 2012. That’s how the world learned about the way Spotify successfully approached agility in its teams. After that, quite a few technology companies tried to use the whitepaper to emulate what is now called the Spotify model. So, what’s behind the name, and why is it so popular? What are the key benefits of the Spotify agile model? In this article, I will explore the main elements and principles behind this methodology and discuss the ways you can integrate them into your own company.

What is the Spotify model?

Spotify used to be one of the numerous companies following the Scrum delivery, and it worked well for a number of years. But as it started to scale and the engineering teams grew from one to several, Spotify realized that many Scrum practices no longer served their purpose. That’s when they decided to make some of the standard Scrum practices optional, focusing more on the things that stand behind the rules, i.e., value principles over practices.  

“Rules are a good start, but then break them when needed.” 
Henrik Kniberg, Owner and Agile Coach at Crisp;
Spotify Engineering Culture (part 1)


The Spotify model represents the company’s view on scaling from both technical and cultural perspectives. The model consists of two parts: the first one is the organizational structure, and the second is a set of accompanying values aimed at driving the organizational change. 

At heart is an engineering culture that chooses motivation, community, and trust over structure and control. Autonomy becomes a key to motivating people, with the team deciding what to build, how to build, and how to cooperate while building it. 

As Henrik Kniberg explains, there’s surprisingly little ego in their company: people readily give credits to their colleagues and, consequently, are easily asking for and offering help when needed. This makes work more productive and increases employee satisfaction to an awesome 94%. Here’re more details on the model structure. 

Key elements of the Spotify model

So, how does Spotify make agile work for a large organization with hundreds of developers? Praising autonomy and simplicity, the model identifies the following key elements for structuring people and teams: 

Spotify team model
Spotify team model

Spotify squads

Much like Scrum teams, squads are autonomous, cross-functional groups that focus on a specific area. Commonly made up of 6 to 12 individuals, squads have all the tools and skills needed to design, develop, test, and release features. 

Functioning as a mini startup, these self-organizing and self-managing squads are free to choose their own framework or agile methodology. They’re all encouraged to practice lean startup principles like validated learning and minimum viable product (MVP).

Each squad works on a long-term mission like scaling the backend systems or creating the radio experience. This allows them to build expertise in a specific area.

Spotify squads don’t have a formally appointed leader, but they do have a product owner responsible for setting priorities and maintaining high-level roadmaps. A Spotify agile coach helps squads improve their way of working by running retrospectives or conducting one-on-one meetings. 

Ideally, each squad has direct contact with stakeholders and isn’t held up by other squads.

Tribes

Multiple squads that work in related areas form a tribe. A tribe lead takes responsibility for providing an optimal habitat for the squads. The tribe essentially functions as an incubator for the mini squad startups. 

Tribes subscribe to the “Dunbar number,” which claims that social relationships are harder to maintain when there are more than 100 people, or so. Groups that are too large often see more restrictive rules and tighter management. To avoid bureaucracy, tribes commonly have no more than 100 people.

Tribes hold regular gatherings where squads present their progress and learnings. They may provide live demos of working software, hack-day projects, and new tools and techniques.

Chapters

With squads acting autonomously, the specialists need to find a way to align on the best practices. Chapters make this possible. 

A group of specialists with similar skills and from the same tribe, chapters allow companies to benefit from the joint efforts of the squads—all without sacrificing autonomy. Functioning as a line manager, a chapter lead supports the personal growth of chapter members and helps them overcome challenges.

Each chapter holds regular meetings where they talk about their area of expertise and related challenges. Companies may have a testing chapter, frontend chapter, backend chapter, and so on.

Guilds

Just like chapters, guilds are organized around a specialty. The main difference is that they are more organic and wide-reaching. Anyone who is interested in the topic can join. 

Guilds and chapters have the same purpose. They both aim to promote transparency, alignment, and motivation. But unlike chapters, members from any squad, chapter, or tribe can come together in guilds to share knowledge, practices, tools, and code. Instead of holding regular meetings, guilds usually conduct workshops, which are much like hackathons. 

Trio

Also known as the TPD trio, it’s a unity of a tribe lead, design lead, and product lead. Every tribe needs a trio to ensure that these three perspectives are well covered and accounted for when working on the new functionality. 

Alliance

As a business scales, multiple tribes sometimes need to work together for a common goal. With this goal, three or more tribe trios form an alliance to enable better cooperation between the tribes.

Sounds like you’re playing an MMORPG game, doesn’t it? All these tribes, guilds, and alliances. But we’ll be damned if they don’t work. Spotify’s agile model actually brings many benefits.

The benefits of the Spotify model

The Spotify model is a powerful force propelling Spotify to the massive success that it is today. But its power doesn’t lie in the organizational design; it’s in the Spotify engineering culture. Let’s look at the core principles of the Spotify model and how it stays beneficial.

Less formal processes

Organizing around work, the Spotify model doesn’t give much attention to processes and ceremonies. There are no specific practices for granting more flexibility for the work of each squad. Instead, emphasis is put on aligning squads with each other and helping them become more productive.

Less top-down management, more autonomy

To encourage innovation and creativity, the Spotify model fosters autonomy. Spotify management trusts people to complete their work on their own terms. They decentralize decision-making, allowing teams or squads to find the framework or practices that work best for them. 

A balance between autonomy and alignment

Painting the big picture, managers try to steer squads toward the same direction without telling them how to get there or how to overcome obstacles. They strive to find a balance between team self-management (autonomy) and central direction (alignment).

More trust, less control

Instead of controlling employees, Spotify chooses to trust and support them. They believe that people have the company’s best interests at heart, and this encourages more freedom to innovate. 

Taking on the role of servant leaders, managers ask teams how they can help instead of imposing deadlines or breathing over people’s shoulders.


“Agile at scale requires trust at scale.”
Henrik Kniberg, Owner and Agile Coach at Crisp


Minimized dependencies, enhanced velocity

Spotify introduced an architecture that kept releases independent of each other. The shift enabled frequent releases to production. Functioning as an independent unit, Spotify squads are able to ship software quickly and with minimal friction.

Promoting productivity through motivation

Operating on the belief that motivation has the greatest impact on productivity, Spotify subscribes to this formula: 


Productivity = Effort x Competence x Environment x Motivation


Instead of imposing rules, they push employees to excel in their jobs by finding ways to keep them motivated. 

Placing stakes on improvement culture

Spotify provides dedicated resources to help teams improve their ways of working. Each squad is encouraged to keep improving how they work and create a model for what would make their work better. 

A focus on a healthy culture

Spotify model promotes a healthy company culture based on mutual respect and a common success concept over the egocentric approach. And when the processes within the healthy culture don’t work well enough, people take the initiative to fix problems or broken processes on their own rather than complain or continue to work inefficiently. 

Value over business

Boosting productivity is not about working people to exhaustion. It’s about maximizing the value of their work. In the Spotify model, teams generate ideas, conduct experiments, and measure the results. If necessary, they tweak their approach to further maximize the value of their work.

Emphasis on employee satisfaction

According to Spotify’s Head of People, 91% of employees enjoyed working with the company. But with the bar set high, they don’t consider the percentage satisfactory. Spotify continuously works to make sure its employees are happy.

Defining what awesome looks like

Spotify encourages squads to define what awesome looks like for them. With a clear vision of awesomeness serving as their benchmark, teams are more likely to reach their destination. Through regular ‘team health checks,’ each squad can track the improvements they’ve made over time. 

Failures are encouraged

At Spotify, mistakes are considered acceptable and are a part of people’s push for innovation. They also encourage people to be more accepting of failure: 


“To build something really cool, we will inevitably make mistakes along the way. But each failure is also a learning [sic]. So when we do fail, we want it to happen fast so we can learn fast, and therefore improve fast. It’s a strategy for long-term success.” 
Henrik Kniberg, Owner and Agile Coach at Crisp;
Spotify Engineering Culture (part 2)

How to make the Spotify model work for your business

The Spotify model fosters innovation and productivity in a high-trust environment. Businesses that adopt the model can expect better products, motivated employees, and happy customers. But the model doesn’t always offer the same results to all companies.

When considering the Spotify model, take a closer look at your company’s existing structure and culture. You’ll soon notice that while the model is simple, the environment it’s operating in is much more complex. The key is to draw inspiration from the Spotify model instead of blindly copying it. 

Here are some ways to approach the model while taking into account your own unique organizational context.

Build a culture of trust and autonomy

Backing the culture of autonomy and trust in your organization is critical. Autonomy motivates people to create better products faster, as the teams can make decisions locally and minimize hand-offs. This way, we can avoid bottlenecks with dependencies and coordination.   

To make this work, figure out if you can push some decisions to the team level at your company. Gradually take the decision-making power from top-level management and give it to the teams who Strive for transparency

Look for inclusive ways to gather feedback and promote asking for and providing help within the company. Make the squads’ goals and their way to reach them visible to other teams and inspire employees to share their expertise across different levels, including management. 

Don’t punish for mistakes

The road to innovation is paved with mistakes. Your business will not grow if it’s made up of people who are afraid to err. Make sure you don’t react to failures with punitive measures because this way, you’ll only dull down people's motivation and innovative spirit. Encourage mistakes but make sure your employees draw conclusions from them.


“We aim to make mistakes faster than anyone else.”
Daniel Ek, co-founder and CEO at Spotify


Make releases a routine

With one development team, releasing isn’t a big problem, but what if there are several teams working on one project? Then you have to look for ways to synchronize their work, and it can take months to get a stable version. To avoid heavy lifting and stressful communication, make releases smaller. It will help to release more frequently and easier.  

For that, you need to optimize your test automation and continuous delivery processes. Henrik Kniberg recommends changing the architecture to use decoupled releases.  

Conclusion

It’s not always easy to scale agile methods. But the Spotify agile model provides a well-documented case of a successful implementation. It has helped Spotify and other organizations increase innovation and productivity by focusing on autonomy, communication, accountability, and quality.

But the lessons shouldn’t be about the model itself: the key takeaway here is to learn from good ideas and find what works best for your own company. Instead of copying the Spotify model, take the time to really understand the practices, structure, and mindset behind the approach. You can then tweak the model to fit the environment in your business. The goal is not to become exactly like Spotify but to make use of the model’s benefits. Design, test, and evolve your own model. 

Taking the next step

Are you hungry to learn more about the Spotify model? 

Check the two-part video on Spotify Labs about the engineering culture at Spotify: 

For some more insights, look here:

And of course, it’s always a good idea to consult professionals before reformatting your company structure. So if you need technical guidance on the way to building your own agile processes, feel free to reach out to us at Altigee!

Share:

Read Next

Written By

Altigee helps startups and tech companies build remote engineering teams or extend their existing teams with senior tech talent. For the past 4 years, we helped our clients to hire 200+ engineers.

More from
This Topic

Related read
9 Questions Every Non-Technical Founder Should Ask Themselves