Spotify Model & Culture
a true story?

Flashback: Insights from 2016
30 min read
Updated:

Introduction

The following text serves as a personal record of information and a carefully selected summary of a presentation that I recently came across on YouTube.

My intention is to create a resource that I, and possibly you, can refer back to for inspiration and reference.

This is the first article in a series where I aim to connect the past with the present by providing updates and reflections on how these concepts and observations have evolved or remained constant over time.

Disclamer

It’s important to note that the presentation which serves as the foundation for this article was originally delivered in 17th April, 2019. The experiences described by Fredrik Wendt in the presentation were most likely from between August 2016 to February 2017. Given the rapid pace of change in our world, some of the information and perspectives shared in the video reflect a snapshot of that time, capturing the era’s zeitgeist.

As we journey through this piece, remember that while some insights may have evolved, they serve as a valuable baseline for understanding subsequent developments.


“Let’s be clear. Our way of working is unique, but it could be better. We make mistakes and realise that we don’t have the answers to all the questions.
As we’re growing so quickly, new challenges arise every day.
The biggest challenges are to attract the right people and to harness our innovation, agility, and unique culture while welcoming 100 new employees to the team every month.”

~Katarina Berg, (a Chief Human Resources Officer (CHRO) at Spotify)


Experimentation and Hack Week

A vibrant and welcoming scene of a dozen new Spotify engineering employees participating in a Developer Bootcamp

Core to Culture: Experimentation as a Cultural DNA

“Everybody has experimentation in their DNA” — At Spotify, experimentation is a vital part of the company’s culture. Employees are encouraged to constantly test new ideas, learn from the outcomes, and explore new approaches. This fosters an environment where taking calculated risks and innovating are accepted, expected, and celebrated.

Boot Camp Induction: Rapid Development and Delivery

Spotify’s boot camp is an onboarding program for new hires. It immerses them in Spotify’s culture and working methods, encouraging rapid development and delivery within a short timeframe, typically a week. The aim is to get new employees accustomed to Spotify’s pace and ethos of experimentation from the outset.

Hack Week Impact: Creativity and Innovation

“I was overwhelmed… at least 50 of those [projects] were ready to go to market” — Hack Week is an event that allows employees across the company to step away from their day-to-day tasks to work on innovative projects or ideas they are passionate about. It’s a time dedicated to creativity, experimentation, and exploring new technologies or solutions without the constraints of regular work.

Hack Week projects can be individual or team efforts and often lead to new features, improvements, or concepts that could be implemented within Spotify’s products.

Emphasising Cultural Aspects

An image of a bustling small office room during a developer’s working day at Spotify. The room is filled with natural light from large windows.

Practice Variation: “Not every squad works the same”

Spotify has a set of core practices and principles, but it understands that different teams have different needs. As a result, they allow each team to adapt to their unique context, project, and team dynamics. Spotify values autonomy and encourages teams to find out what works best for them instead of imposing a one-size-fits-all approach.

Mission Clarity: Visible and Clear Squad Missions

At Spotify, the missions of different squads are not just abstract concepts but are made tangible and visible within the workspace. This could be through creative displays, whiteboards, or posters, making each squad’s objectives clear to all members. This practice of visually articulating missions serves multiple purposes:

Agility in Different Forms

An image of a bustling small office room during a developer’s working day at Spotify. The room is filled with natural light from large windows

“At Spotify, we’re often testing and launching improvements and new features.
This means you might see something on the app your friend doesn’t, or get a new feature to try temporarily.”

~ Anonymous Why has the app changed? Source: support.spotify.com4

Transparent Strategic Planning

Strategic Agility

“Everybody understands what we’re doing” — Spotify’s approach to strategic agility involves clear and transparent communication of company goals and strategies. This ensures that all teams are aligned with the company’s vision and objectives, allowing for cohesive and strategic decision-making.

Technical Agility

At a Squad level this practice is implemented as a Mission Board mentioned previously.

Strategic Agility

Strategic agility at Spotify also means being responsive to market trends and user behaviours. By understanding and anticipating these changes, Spotify can make strategic ‘bets’ or decisions that position the company for future success. This approach requires a deep understanding of the industry and the flexibility to shift resources and focus as needed.

Technical Agility

Being Responsive to Market and User Trends also has an impact at a more local level where by gathering and analysing metrics quickly, Spotify can learn from user interactions and refine its product in real time. This aspect of technical agility is crucial in a fast-evolving market, where user preferences and technological capabilities are constantly changing.

Evolving Culture and Decision-Making

An image capturing Spotify's journey of adaptability and resilience, likened to a seasoned sailor navigating turbulent waters, have been created

Spotify has a unique reputation for being highly adaptable and responsive to external market forces. The company underwent significant changes in preparation for its initial public offering (IPO). It reconstructed its decision-making processes and organisational structure while remaining true to its core values as it continued to grow and scale its culture.

As a public company, Spotify had to find a balance between financial predictability and cultural agility. To achieve this, it had to redefine roles and responsibilities and implement more structured processes.


Spotify going public: Bevelenent or Machiavellian?

Was Direct Listing a bold move to benefit their team or a clever strategy to boost record label profits?

Join me to find out if they truly did it for the team or just for the record label's profit

Squad, Chapter and Tribe Structure

An image of a bustling small office room during a developer’s working day at Spotify. The room is filled with natural light from large windows.

Squad Formation: Strategic and Adaptive Formation

Spotify forms squads based on their current needs and goals, which allows them to be adaptable to changes in the market. Each squad focuses on specific aspects of the product or service, ensuring that they have the necessary expertise. These squads can be created, disbanded, or changed as needed to keep up with Spotify’s commitment to being agile and responsive.

Chapter Lead Roles

Chapter leads have diverse responsibilities to support their teams. Depending on each team’s needs, they can offer guidance, mentorship, and even hands-on technical work. Such roles help squads work together efficiently within Spotify.

Tribe Leader’s Role

Tribe leaders follow a servant-leadership approach, focusing on facilitating squads’ needs instead of directing activities. This might include budget allocation, providing necessary tools, or removing obstacles squads face. They act as coaches and enablers, empowering teams to make decisions, innovate, and take ownership of their projects. By maintaining autonomous and self-organising squads, tribe leaders create an environment where teams can thrive and contribute effectively to align with the organisation’s goals.

Decision Transparency

An image of a bustling small office room during a developer’s working day at Spotify. The room is filled with natural light from large windows.

At Spotify, the team members communicate with each other on a regular basis and make around 5 to 50 decisions every day.

To keep everyone in the loop, they follow a simple system. Whenever a squad makes a decision, they send an email to decisions[at]spotify[dot]com. This way, the whole company gets to know about any changes or new directions.

The emails are designed to be concise yet informative, including details about what the team learned, the decisions they made, the reasons behind them, and additional resources for those who want more information.

Coaching is natural

An image of the personalized one-on-one coaching session.

“Everywhere you go, you have coaching.”

At Spotify, employees have access to a supportive and nurturing environment where coaches and mentors are always available to lend a helping hand.

The coaches utilise a structured approach called the “four stances of a scrum master” to provide guidance to employees through facilitating, teaching, mentoring, and coaching.8

The primary role of a coach is to assist employees in identifying their essential needs and help them find the answers they seek. Although coaches may not have all the answers, they can help employees by asking the right questions.

To facilitate this, weekly one-on-one sessions with chapter leaders or coaches are available to all staff members. These sessions help employees determine the next steps in their career development and provide a safe space to discuss any concerns or challenges they may be facing.

Post-mortems

An image capturing the essence of a post-mortem discussion at Spotify, showcasing a proactive and collaborative atmosphere.

“Post-mortems are part of their culture.”

When things go wrong at Spotify, it’s common to have a post-mortem discussion. This is not a complaining session but rather a way to acknowledge problems and come up with solutions to prevent them from happening again. It’s a process of learning from mistakes and taking action to improve things.

This article from 2013 delves into Spotify’s incident management, detailing a major outage and the subsequent analysis and remediation efforts. It highlights the importance of learning from failures, identifying root causes, and implementing effective solutions to prevent recurrence.

Metrics and Data-Driven Approach

An image of the room with big screen full of metrics.

To ensure that the service is performing well and meeting the needs of its users, it is crucial to gain a deep understanding of user engagement and the overall health of the platform.

Metrics are a powerful tool that can provide valuable insights into how frequently users interact with the platform, how they navigate through it, and what features they find most useful.

By analysing these metrics, we can identify areas for improvement and make data-driven decisions to enhance the user experience. This information is not only useful for product improvements, but also for developing effective marketing strategies that resonate with the target audience.

Performance Metrics: DAU and MAU

A slide showing Fredrik Wendt talking about desired metrics for numerous features under AB testing

Spotify made this very clear that these are two of the most critical metrics:

MAU — Monthly Active Users
DAU — Daily Active Users

Whatever change you make to our product, these two are not supposed to go down.

It’s really important to stress how significant this simple metric is and how it directs the development of their product.

You will see something like this [table] whenever they make changes, do A/B testing, etc.

This method enables the swift evaluation of the effects of any changes that have been implemented.

Dominic Kalinowski

My Reflection

In my experience, the significance of gathering metrics effectively and efficiently cannot be overstated. Yet, astonishingly, very few companies excel in this area. The complexity of the process can severely delay the feedback loop, which is essential for informed decision-making. Moreover, when metrics collection relies on manual tasks, it introduces an additional layer of delay. This manual intervention not only slows down the process but also increases the likelihood of errors and oversights.

What often happens is that if the process isn’t automated and if there’s no dedicated focus on generating and analysing these feedback reports, a data-driven approach becomes more theoretical than practical. It’s easy for such practices to be sidelined, forgotten, or abandoned altogether.

Efficient metric gathering — characterised by automation, ease of presentation, and minimal manual intervention—stands in stark contrast. Manual tasks, particularly when not the primary responsibility of the person assigned to them, introduce friction and are prone to being overlooked. This neglect is usually not due to negligence but to the process not being integrated seamlessly into daily workflows.

The essence here is that for metrics to drive decision-making, they must be gathered and presented in a manner that is as streamlined and automated as possible. This approach minimises delays, reduces the risk of human error, and ensures that data-driven insights are consistently front and centre in strategic discussions.

It’s about creating a system where data flows freely, insights are readily accessible, and the value of metrics is not just recognised but fully leveraged.

DIBB Framework Utilisation: Data, Insights, Beliefs, Bets

A slide showing Fredrik Wendt explaining the DIBB Framework

“At least ten years ago, I think, they saw that the number of people listening to music on the desktop decreased, and mobile usage increased. But 95% of our team have desktop skills, so they work on desktop stuff.
That’s a bad match.
The insight is that mobile is overtaking desktop, so we should change.
We’re optimising for the wrong thing, so let’s increase the number of mobile developers and retrain people to build infrastructure to iterate fast on mobile.”

This framework guides decision-making by combining several key elements.

Data

Objective facts and figures collected from the platform provide a solid foundation for understanding user behaviour and preferences.

Insights

Interpretations drawn from the data help to uncover trends and patterns that might take time to be noticeable.

Beliefs

The team’s hypotheses or assumptions are often based on data insights, experience, and intuition.

Bets

Strategic decisions are made based on data, insights, and beliefs. These bets are informed risks the company takes to pursue innovation and growth.

Architecture First

An image showcasing efficient highway infrastructure contrasted with a chaotic, neglected cityscape, symbolizing investment disparity in infrastructure and software architecture

Innovation through Architecture

The core issue was the pace. Spotify’s development process was designed around quickly trying new ideas, conducting A/B tests, and iterating based on user feedback. This process was seamless for desktop applications but became a bottleneck for mobile development due to the slow, regulated update process. The quest was to maintain agility in a mobile-first strategy without burdening users with frequent updates, which would be impractical and disruptive.

To achieve this, Spotify leveraged server-side updates, allowing the client application to receive and execute updates in a similar way to web technologies. This clever approach enabled them to bypass the traditional mobile app update cycle, ensuring that users always had access to the latest features and improvements without the need for constant, manual updates.

A Strategic Investment

“Invest super heavily in architecture that supports teams to do the valuable work.”

Implementing this new architecture was neither cheap nor straightforward. It required significant investment in creating the necessary platform and tools that would enable such a seamless, server-side update mechanism.

However, the benefits were clear. This strategic decision allowed Spotify’s teams to maintain their agility. They were able to rapidly iterate and deploy new features and fixes. This level of investment in infrastructure to support innovation is rare, setting Spotify apart as a leader in technological adaptability and user-centric design.

Technical and Product Ownership

An image of three individuals, ensuring clarity in their presentation within the warmly lit room.

Balanced Ownership Roles

Dual Ownership System

When it comes to developing a product, it is necessary to have a solid understanding of both the technical and business aspects of it. That’s why Spotify introduced a dual ownership model with technical and product owners.

By dividing responsibilities among different owners, the decision-making process becomes more refined and informed. Each owner brings their unique area of expertise to the table, which allows for a more holistic approach.

Technical Owners

The technical quality of a product is vital to its success. Technical owners ensure every aspect necessary to facilitate it is considered at the planning level to make the implementation smooth without significant scope creep.

They focus on a range of technical aspects, including code quality and architecture, to ensure they can scale to meet the higher demand in the future.

Product Owners

They concentrate on the market, business, and user acceptance aspects of the product. This includes understanding user needs, market trends, and business objectives and then translating these into actionable product features and improvements.

Additional Role: Road Manager

“Have a “Road manager” for the retro to help organize and drive the effort and schedule various regular check-ins with retro working group, stakeholders, etc.”
~ Henrik Kniberg, 2015 17

In addition, the road manager is responsible for collecting all the data from A/B tests. They conducted a weekly review to ensure its accuracy.

After the review, the Road Manager shares the results with all the stakeholders involved in the project.

Long-term Commitment

An image to symbolize Spotify's approach to achieving long-term goals, paralleled with the nurturing of a plant from seed to maturity, showcasing the stages of growth and the iterative process of development and testing

Spotify has a unique approach to achieving long-term goals. They recognise and accept the challenges that come with them, such as:

Facing those challenges, however, is vital for the company’s long-term success and sustainability. They understand that certain innovations or improvements take time to develop and may not yield immediate results.

To maintain morale and motivation, Spotify emphasises the importance and potential impact of these initiatives. Employees are encouraged to focus on the larger goal and the positive change their work contributes to.

Draft to test, rewrite to scale

When the feature reaches an MVP stage, the team decides whether it should be released to the public. After conducting A/B testing, they gradually introduce the feature to a wider audience.

That often means rewriting parts to allow for global scale because they don’t need to write for global scale right away before they know it’s a good idea.

A team will revisit the code for a long time, which can sometimes be demoralising. However, everybody understands its value.

They accepted it.

Conclusion

In summary, the video has provided additional details about the Spotify Model. However, as it was presented in 2019, we must seek more recent field reports to understand how the concept has evolved over time.

Source: https://www.youtube.com/watch?v=ar4lq1l8pAc

References

  1. Kumar, S. and Expedia Group. (2021)
    How we measure backstage proof of value at Expedia Group,
    Spotify Backstage.
    Available at: https://backstage.spotify.com/blog/measuring-backstage-proof-of-value-at-expedia/
    (Accessed: 14 March 2024).

  2. Da Costa, M. (2023)
    How Spotify uses Skill Exchange to power our Hack Week,
    Spotify Backstage.
    Available at: https://backstage.spotify.com/blog/skill-exchange-hack-week/
    (Accessed: 14 March 2024).

  3. Šmite, D. et al. (2023)
    Decentralized decision-making and scaled autonomy at Spotify,
    Journal of Systems and Software.
    Available at: https://www.sciencedirect.com/science/article/pii/S0164121223000444
    (Accessed: 14 March 2024). 2 3 4

  4. Why has the app changed? (no date)
    Spotify.
    Available at: https://support.spotify.com/uk/article/why-has-the-app-changed/
    (Accessed: 14 March 2024).

  5. @meahtenoha (2018)
    Retirement of our running feature,
    The Spotify Community.
    Available at: https://community.spotify.com/t5/Content-Questions/Retirement-of-our-Running-Feature/td-p/4383603
    (Accessed: 14 March 2024).

  6. Moton, T. (2018)
    Spotify running review: Retirement for Spotify running,
    Tunefab.
    Available at: https://www.tunefab.com/spotify/how-does-spotify-running-work.html
    (Accessed: 14 March 2024).

  7. Blake, J. (2020)
    When should I write an architecture decision record,
    Spotify Engineering.
    Available at: https://engineering.atspotify.com/2020/04/when-should-i-write-an-architecture-decision-record/
    (Accessed: 14 March 2024).

  8. Dulin, M. (2022)
    Facilitating, teaching, mentoring, and coaching,
    Agile Ambition.
    Available at: https://www.agileambition.com/facilitating-teaching-mentoring-coaching/
    (Accessed: 14 March 2024).

  9. Siseman, F., Connor, A.R. and Lemon, C. (2023)
    Agile Coaching (AC) at Spotify: Shining a light on the AC career framework
    Spotify Engineering.
    Available at: https://engineering.atspotify.com/2023/01/agile-coaching-ac-at-spotify-shining-a-light-on-the-ac-career-framework/
    (Accessed: 14 March 2024).

  10. Poblador i Garcia, D. (2013)
    Incident management at Spotify,
    Spotify Engineering.
    Available at: https://engineering.atspotify.com/2013/06/incident-management-at-spotify/
    (Accessed: 14 March 2024).

  11. Byrum, C. (2022)
    Failing forward - how we grow from incidents,
    Spotify Engineering.
    Available at: https://engineering.atspotify.com/2022/05/failing-forward-how-we-grow-from-incidents/
    (Accessed: 14 March 2024).

  12. Goodwater Capital (2018)
    Understanding spotify: Making Music Through Innovation,
    Goodwater.
    Available at: https://www.goodwatercap.com/thesis/understanding-spotify
    (Accessed: 14 March 2024).

  13. Iqbal, M. (2024)
    Spotify revenue and Usage Statistics (2024),
    Business of Apps.
    Available at: https://www.businessofapps.com/data/spotify-statistics/
    (Accessed: 14 March 2024).

  14. Holton, K. (2009)
    Spotify launches iPhone and Android mobile services,
    Reuters.
    Edited by R. Winchester.
    Available at: https://www.reuters.com/article/idUSTRE5861AX/
    (Accessed: 14 March 2024).

  15. Hamburger, E. (2013)
    Spotify announces free streaming on Android and iPhone, but only in Shuffle Mode,
    The Verge.
    Available at: https://www.theverge.com/2013/12/11/5199692/spotify-announces-free-streaming-on-android-and-iphone-but-only-in
    (Accessed: 14 March 2024).

  16. Constine, J. (2015)
    Spotify makes the shift to mobile with 52% of listening now on phones and tablets,
    TechCrunch.
    Available at: https://techcrunch.com/2015/01/10/music-is-a-mobile-linchpin/
    (Accessed: 14 March 2024).

  17. Kniberg, H. (2015)
    How we do large scale retrospectives,
    Spotify Engineering.
    Available at: https://engineering.atspotify.com/2015/11/large-scale-retros/
    (Accessed: 14 March 2024).

  18. Goldsmith, K. (2016a)
    Building a technical career path at Spotify,
    Spotify Engineering.
    Available at: https://engineering.atspotify.com/2016/02/technical-career-path/
    (Accessed: 14 March 2024).

  19. Goldsmith, K. (2016b)
    Spotify Technology career steps,
    Spotify Engineering.
    Available at: https://engineering.atspotify.com/2016/02/spotify-technology-career-steps/
    (Accessed: 14 March 2024).