Chapter 2 — SpaceUP 🚀! Or, how we changed the way our Product & Engineering works

Martin Zahumensky
Ataccama SpaceUP
Published in
12 min readDec 22, 2021

Previous chapters: Chapter 1

In August 2021, we at Ataccama decided we need to change the way our Product and Engineering works. We internally called this change “To the Moon and Beyond,” implying that we are far along on our route, but not yet at our destination, and we want to reach much further. I have talked about the motivation, rationale, and desired outputs in the previous article. Now, I want to spend some time explaining how the model looks and works at a high level, the motives behind its creation, as well as shed some light on why we haven’t simply taken something that is already adopted and available on the market.

Existing methodologies and why we decided to go a different way

We were considering several different models such as Scrum, agile at scale, SAFe, two-pizza teams, Shape Up, the Apple model…however, in each and every model we found some parts we didn’t like, or thought wouldn’t work for us. So, we took a step back and started to think about what the cornerstones are:

  • We knew we needed to give enough attention to all products, and have strong teams and expertise in a given area.
  • The “time-restricted mission” type of work had always worked for us — anytime we have a bigger issue we’re always able to build the team around it and get it solved.
  • We knew we need strong leaders with a deep understanding of a given product area, visionaries shaping the product’s future, people who had the ability and the power to execute and build the teams.
  • We had some negative connotations around some past issues that didn’t fully work well for us, such as OKRs and communities.
  • People know the standard models, and if we change something there’ll be a lot of questions about why we aren’t following standards.
  • And mostly, we wanted to make it part of our culture, make it our own.

That’s why we found inspiration and used some key principles from different models, mostly how Pipedrive organizes its engineering teams, and how Apple shapes its organization around subject matter experts. We also took some inspiration for mission planning and scoping from Basecamp’s famous Shape Up. We are grateful for all these inspirations and to all of these companies for sharing their methodologies and experiences. It helped us tremendously in the process.

SpaceUP — From Teams to Spaceports, Missions, Ground Crews, Societies, and…?

Building rockets isn’t the only interesting job one can do. In a deeper sense, what we do is very complicated and requires a lot of knowledge and skills, so we have chosen to go with a Space theme and terminology — we wanted to go beyond the moon, but the space inspiration and how rockets are launched also matched the missionary approach we are so good at. Everybody can relate to it, it’s easy to explain some of the concepts on the Apollo program, and it’s simply cool. 😎 (By the way, in upcoming articles we’ll cover how branding and our own approach is important to achieve success).

So, that’s how the SpaceUP name was born — go beyond the moon and deeper into space 🌕 🚀, working on topics that often go unexplored.

Here are the 4 key areas of the organization you need to understand, and I will break this down in more detail below.

  • Spaceport — the units which encompass a bigger team of 20–45 people. Each Spaceport has its own clear purpose: to own a specific part of our product.
  • Ground crew — the part of the Spaceport crew that remains on land, supporting missions and our customers, and improving what we have.
  • Missions — how we deliver and ship new things. It’s a crew of people formed around a specific goal we want to achieve. Each Spaceport can fly multiple missions, typically 3 to 4 of them at a time.
  • Societies, and Circles — this is where everyone within a common area of expertise connects. Societies and Circles assure we connect people from multiple Spaceports, and that in the end we don’t ship 5 different platforms but have ONE cohesive product.

Spaceports

To ensure ownership of things the organization produces, we needed to change the way we work. We needed to get rid of silos and make sure we maintain a feeling of end-to-end ownership — I built something, I run and support the thing, instead of the previous attitude — I delivered the artifact, now someone documents it, tries to install it, and when it doesn’t work it’s someone else’s problem.

So we created Spaceports — each one has a clear purpose, to own a part of our product. We are talking about end-to-end ownership, everything from defining objectives and programs according to company objectives, figuring out how the Spaceport will work and organize their day-to-day work, and hiring the right talent, to delivering the product, letting the world know we have this product, and making sure consultants and our partners can use it independently, and customers are happy.

Spaceports should have about 20–45 people. They should be relatively stable in time but they can split and connect depending on how company missions and goals change, and the product develops. They should contain all the people they need in order to deliver.

  • Engineers
  • Product designers
  • Product managers
  • Technical writers
  • Any other professions(e.g. data scientist, product marketing) needed to fulfill spaceport objectives
  • Or any tools required for effective and successful delivery.

The Spaceport is your home. This is where you make long-lasting relationships, learn new things, challenge yourself, and build an amazing product. The Spaceport is your hometown, it’s your safe zone where your teammates will always have your back. Spaceports are self-organized. They have the freedom to choose which processes or tools to use if it leads to better results. But they are encouraged to use so-called highways like CI/CD pipelines, monitoring, deployment environments, or any other enabling infrastructure as much as possible.

When we launched the organization we started with 7 Spaceports, divided into two main types:

  • Product Spaceports — they are responsible for the product we ship to customers.
  • Technical Spaceports — these build so-called highways for the product Spaceports, making their lives easier. They work on some shared components of the product such as keycloak (we use as identity, access management module for all our applications), or things like CI/CD pipeline, and also things like delivering our products in the cloud, and running operations.
We have 7 Spaceports — 5 Product Spaceports & 2 Technical Spaceports

Each Spaceport has its own objectives to contribute to the improvement of the given Spaceport. These objectives should contribute to company objectives. By now, you probably have a feeling that these vertical structures will break our aim of having ONE platform, but don’t worry — this is taken care of and you will learn about it in a bit.

Spaceports provides many other benefits:

  • We can find a new job for people if they want to try something different, or something doesn’t work out for them in their current Spaceport they can move. This helps maintain lower attrition.
  • Each Spaceport can be slightly different, so they can share knowledge and best practices. This instills a challenging attitude, propels us ahead, and makes us better.
  • It makes the teams bigger, giving higher flexibility on rotating people around and easier knowledge sharing.
  • It creates “healthy” competition which drives the others continuously forward, to achieve more.

👑 Each spaceport has one Leader. This is someone who is ultimately responsible for the success of the whole Spaceport from brainstorming ideas to building the team, to the successful delivery of the product and fulfilling Spaceport objectives.

💡 Keywords

  • Home
  • Team
  • Mates

Ground crew

A necessary part of the Spaceport. We can’t have everyone from the spaceport flying missions at the same time. We need someone on the ground to keep things running, support customers and mission crews, pick up the calls, perform maintenance, make the Spaceport a more effective and happier place to work in, pay off technical debt which naturally comes from missions, and create a safe place for the landing of future missions.

Each Ground Crew support sets of product components or services from landed missions. There is no single component or service without a backing ground crew. As Spaceports land many missions over time, we need them to take care of components created during missions including the source code, runtime components, documentation, and other assets. This is part of the end-to-end ownership, supporting what Spaceports create. Roughly 25% of the team should be part of the Ground Crew at any given time, but it can differ between Spaceports. In other words, the Ground Crew needs to have a size that provides stellar support to its customers and users.

To promote knowledge sharing and make the work more enjoyable, people will regularly rotate between flying missions and supporting others as Ground Crew. Everybody has the right to go on a Mission! And at the same time, we need to have people on the ground who can support all the components owned by the Spaceport.

🎥 Example: Ground Crew will be tackling technical debt, improving test coverage, fixing bugs, providing support to any party that needs it, being on call for customers, or other teams and missions, etc.

Each Ground Crew has a Leader 👷‍♀️ — This is a person who is in charge and overlooking what is happening on the ground. It’s more of a role than a position. Everybody is encouraged to be a Ground Crew Leader at some time. This role is permanent and mandatory but settled by different Spaceport engineers in time. We encourage rotating engineers in this role each 3–6 months, as this experience helps build your future leaders. You can find out more about these key roles in the next chapter.

💡 Keywords

  • Support
  • Fixing things
  • Knowledge sharing
  • Down on Earth

Missions

Spaceports fly missions. The aim of a mission is to take us one step further to achieve defined objectives or to complete the space program. While the objective or program only defines what we want to achieve, the mission is the actual solution, the “how” part. It should be of reasonable timing and reasonable sizing (weeks rather than months) and needs to have measurable impact and value for customers, and contribute to company objectives.

Missions are important for the entire SpaceUP concept for multiple reasons, and they are the cornerstone around which we build the organization.

  • They make people missionaries instead of mercenaries.
  • They allow people to add required skills to specific missions.
  • They allow people to work on things they like, or want to learn.
  • They allow the use of technologies of interest.
  • They ensure people are doing meaningful things, and they actually know what is expected.
  • They ensure people are rotating jobs over time.
  • In the end, it makes people happy.

A Mission has multiple phases from the pitch, to preflight, space-time, landing sequence, and decommission. We’ll talk about it in detail in the next article about the Mission Lifecycle. A mission is always owned by a Spaceport — always. This is very important as it determines the future ownership of anything that will be created on the mission, and means that this particular Spaceport will be responsible for maintaining it. And yes, some Missions will fly across multiple Spaceports, but there will always be one Spaceport responsible for it.

A Mission needs to have a clear definition, containing information about the expected outcome, boundaries, expected landing date, and also must have an assigned Mission Crew (the size varies based on the characteristics of the Mission) which has all the necessary skills for delivering the Mission. This ensures we have the best possible set-up in place before we start flying, minimizing the chances that we crash. The Mission is delivered end-to-end by the Mission team. In the future, we envision that the Mission’s outcome will be deployed to customer environments as part of the Mission lifecycle. This will ensure we deliver a working solution, in real conditions, and the Mission Crew sees immediate achievements and impact of their work.

When a Mission Crew flies in space, they are on radio contact — this means they are focused solely on the Mission. If something is needed, the Ground Crew can contact the Mission Crew and ask for advice. Given the fact they are in space, they should not be pulled out of the team to do something else, but they can help through the radio. This is very important because removing Mission Crew members can cause the whole Mission to crash. However, it is expected that some roles like product managers and designers have much higher involvement during the Mission preflight and the initial stages of Missions. When the load decreases they should work on preparing the next Missions, analysis, and preparing pitch materials.

After the Mission lands, the members go back to the Spaceport to spread the knowledge. Some members are expected to remain in the Ground Crew. This allows people to rotate, to transfer knowledge from the Mission team to other Spaceport members, and also when flying across Spaceport Missions, people can spread knowledge of how things are done in other Spaceports, thus improving how we work.

👑 Each Mission has one Mission Commander. It’s someone who is ultimately responsible for the success of the Mission. It’s more of a role than a position. Everybody is encouraged to be a Mission Commander at some time.

💡 Keywords

  • Achievement
  • Seeing results
  • Team
  • End-to-end delivery
  • Ownership

Society & Circles

We will talk in detail about Societies and Circles in the next article, but here is a short outline that will help you understand the organization model.

This is where everyone with a common area of expertise connects — it’s the horizontal structure of the organization. As we don’t want Spaceports to be siloed we need also to connect people from multiple Spaceports together, so they can share their experience and knowledge, help others, and agree on company-wide standards. We don’t want to reinvent the wheel, we want to take what worked well in one Spaceport and reuse it in a different one.

As one of the responsibilities of the Societies is to define and improve our standards, they also have their own Missions that will allow us to achieve this.

A Circle is a group of people organized around specific expertise or capability within Societies. Societies are big and there will still be multiple groups, each focusing on a different area of expertise. These specialized groups are called Circles, which you can imagine like communities formed around specific topics and skills. There are two types of Circles, mandatory and voluntary (from the Society's point of view).

We have two main Societies: Product and Engineering.

👑 Each Society has a Leader, someone ultimately responsible for their given Society, making sure our standards are high and people are happy. Also, this person can help you if there is some issue in your Spaceport you can’t solve yourself and need some help, or to escalate something.

💡 Keywords

  • Group of people
  • Craft excellence
  • Education
  • Spreading knowledge horizontally

Wrap up

So, these are the basic blocks of our new organization. There is plenty more to be shared about our Mission lifecycle — how we pitch, plan, execute and land Missions, how Societies and Circles work and what their role is, how we run the Headquarters (HQ), how we executed the actual transformation of 190 people to the new organization, and obviously a retrospective on the whole model. We will share more on these in the next articles.

I hope you enjoyed this, and it gave you some food for thought on how you can adapt this model to help your organization fly to the moon and beyond.

More SpaceUP:

--

--

Martin Zahumensky
Ataccama SpaceUP

Data Management & Visualization enthusiast, working as Head of Product & Engineering @ Ataccama, co-founder & ex-CEO of Instarea, and endurance triathlete.