Are You Sure You Want to Become a Data Science Manager?

Picture this. You have just delivered a killer project, your team’s buzzing, and then – bam! You are asked, ‘Have you ever thought about leading the team?’ Sounds tempting, right? But hold on – do you really know what you’re signing up for?

As a Data Science manager, I have watched my teams grow from 0 to 12 data scientists and helped scale our DS discipline from 20 to 50+. I have also seen fellow managers leave for new challenges. Both instances created a vacuum to fill: the need of a manager to lead a data science squad. Filling this void can be an amazing opportunity, but I have also seen many colleagues fail to adjust to the transition.

The transition to management is no joke. Sure, it has its perks. But no one talks about the trade-offs. Most early managers walk in totally unprepared – that’s where the frustration starts.

What will you be able to read in this blog?

I will cover what are the main considerations for you to think about before making a move into management.

You will read about:

1. Running a high level self-reflection. Why are you thinking about this move?
2. Debunking the glorified "manager" title. In many instances, lead roles are a lateral move, not a vertical one.
3. 5 main differences when you make the move. I am sure there are more, but these 5 are the ones that I relate to the most.
4. Describing the danger zone. Moving back from manager to ICs is not as easy. Also, avoid the things that drain your energy.
5. But also, acknowledging the doors that a DS manager role open. A managerial role makes you grow in areas that few have a chance to experience.

First, tell me why.

The mentioned gap to lead a data science squad, opened an opportunity for those who wanted to transition from their current individual contributor role to a lead role. However, this transition is not for everyone. I have seen those who have rushed into it, regretting the move some time later. I always ask this question to potential people leads: why do you want to become a manager?

Some answers or thoughts I have heard are:

• Are you seeking recognition? Moving into management comes with increased visibility – you’ll lead decisions, interact with stakeholders, and guide strategy. But is this kind of recognition what excites you? And if so, what would you be willing to trade in exchange for recognition?
• Are you stuck at the Senior Data Scientist level? You have spent many years in a Senior DS role and becoming a Principal Data Scientist is really complicated. The next step may seem like management. But before you jump in, consider: Is this where your passion lies, or are you just looking for a way out of stagnation?
• Is it about the money? In the tech space, managerial roles actually have the same salary ranges as their equivalent IC counterparts. There are other industries where managers do get paid more though. Assuming that there is financial benefit with the new role, it is enough to trade in the technical work you love?
• Is it a requirement in your company? In many organisations, career growth seems tied to management. Maybe you feel pressure from above, as if the only way to advance is to take on a Leadership role. But is management truly the only path to growth, or could you explore other ways to progress in your career (even if it is moving to another company)?

Do you relate to any of the above? If so, see that there is a common denominator in my challenge questions:

What is the trade-off?

Debunking the glorified "manager" title.

Before covering the hidden differences between a manager and an IC, let me touch upon the actual "manager" word.

It comes from the industrial revolution.

The term "manager" has roots in the industrial era, when work was defined by efficiency, control, and supervision. In factories, managers oversaw production lines, ensured workers followed processes, and enforced order. Every time I hear the term "manager" referring to a people lead, I cringe.

But the term is recognisable.

Part of the reason we still use "manager" is simply because it’s a catch-all term. It’s familiar. Everyone knows what a manager is, even if the responsibilities of modern managers go far beyond just managing tasks. In industries like data science, where innovation and collaboration are key, managers are more about facilitating success than enforcing authority. However, the term persists because it’s deeply embedded in corporate structures and hierarchies.

I prefer the term "guide".

First of all, I have always disliked the term "manager" applied to people teams. My subconscious brain associates the word "manager" with authority and power, but in reality, you don’t have that much decision making power. I believe that a manager’s job is to guide and support your team, removing roadblocks, and managing expectations from stakeholders. The "big" decisions often come from higher up, so in many cases, you become more of a facilitator than a direct decision-maker.

Management is more a lateral than a vertical move.

There are people who view moving into a manager role is as a promotion. In some companies that might actually true, but in the tech space, becoming a manager is a lateral move. I have never seen a junior data scientist become a manager. It has always been those who reach to senior data scientists and have acquired enough technical knowledge who make the transition. In fact, it is a lateral move even on pay grade, as many tech companies have the same salary ranges for the same level of IC or manager roles. To make an analogy; management isn’t an elevator ride to the top, it’s more like taking a detour through a maze – with half the map missing and someone constantly yelling about deadlines.

The 5 things no one tells you about when you become a Data Science manager

As an IC considering the move to management, you’ve likely interacted with your own manager daily and observed what they do. Maybe you’ve had a manager and a squad lead, so you think you know the drill – I say ‘maybe’ because perhaps you started as the only data scientist in a startup and grew from there, though that’s rare. If I asked you to list what a manager does, your list might include:

• Do 1–1s
• Some admin stuff
• Plan the weekly and quarterly work
• Interact with stakeholders on behalf of the team

These are true enough, but they only scratch the surface. These are tasks, not responsibilities.

Responsibilities are the things that define what a manager’s role really looks like.

Let’s burst this naive knowledge bubble with five things no one warns you about.

1) Your project focus shifts from technical to strategic work

You still need to be technical, but with a different focus.

The more critical code you write, the more of a blocker you become. The temptation to roll up your sleeves and start coding is strong. Especially when a project hits a roadblock or when you know you can do the work 3 times faster. For example; reviewing pull requests (PRs) or offering feedback on the structure of an analysis in a notebook? That’s perfectly fine. These light-touch interventions allow you to stay engaged technically without bottlenecking the team. However, taking on a full data quality audit or single-handedly building a new model? That’s too much. I can guarantee that what looked like a free coding week, can easily become a crowded Outlook agenda and suddenly there is no more time to write code and you are the blocker. When you do this, you’re no longer leading your team – you’re stepping back into the IC role and taking ownership of tasks that your team should be handling.

On this point, I speak from my own experience and wrote an article talking about how my agenda was a full mess and had to review a way to become more efficient handling bigger teams and more projects.

But, you need to be technical enough to guide your team. A good manager understands the nuances of the work their team is doing. You don’t need to dive into the details of every line of code or optimise models yourself, but you must be able to follow the conversation. You need to understand the "art of the possible" – what can be realistically achieved given the time, resources, and technical constraints. This understanding helps you plan the phases of long-term projects, anticipate risks, and communicate effectively with stakeholders.

2) Your impact is less clear

What got you to success as an IC, will not lead to success as a manager.

You are no longer in the driver’s seat. As an IC, your success is directly tied to your output – your code, your models, your analyses. But as a manager, your performance is evaluated by how well your team performs. This can be a tough adjustment for those used to being high performers. A good manager focuses on enabling the team, clearing roadblocks, and fostering a productive environment.

Stopping work is also part of success. As a manager, recognising when to stop on-going initiatives or not taking on too many projects is crucial. Spreading the team too thin can lead to long-term risks and create single points of failure. Remember, it’s not just about what you achieve but also about how you set your team up for sustainable success.

No performance drop in your absence is the ultimate goal. Imagine that you were to take a three-month leave, but your team maintains their output levels. That is a testament to how well you’ve set them up for success. It demonstrates that you’ve empowered your team to function independently. Your role has shifted from being the day-to-day enforcer to a strategic leader who cultivates talent, raises standards and ensures the team thrives, even in your absence.

3) Your growth doesn’t come from books

Learning to lead takes experience, not just theory

Books might help. As an individual contributor, your learning and development can come from reading technical books and applying that knowledge directly to your work. However, the same cannot be said so easily for management. It doesn’t matter if you read a book about influence or how to have tough conversations; while there may be useful frameworks, the real lessons come from experience. It is only when you have to tell someone that their performance is below what is expected or when you have to influence the CEO to hire 10 more data scientists, that you will figure out how to do these things. And there is no canvas, because everyone is different. Even these 2 scenarios will be experienced completely differently depending on the person that you have in front. These situations require more than theoretical knowledge; they demand emotional intelligence, negotiation skills, and the ability to adapt on the fly.

4) Your task is to help others & the wider Data Science discipline grow

It is not only about 1–1s.

Are you willing to slow down immediate results to facilitate learning and growth for your team members? By having 1–1s, you understand what do your direct reports want and where are they in their journey. But as a manager, you know which projects might be coming in the next quarter or how is looking to move away from a project. These things combined, force you to balance how can people be allocated to help them grow, all whilst, delivering value.

Are you willing to consider hiring as a first class citizen? Hiring can be demanding on your time. You have to do your day-to-day job whilst allocating time for hiring. And hiring is not only the actual interview; it is preparing questions, it is coming up with a standard scoring approach and it is doing a post-interview write up. All of these things might seen like a disturbance. But you will only get back what you put in. If you really put your energy into hiring, this will pay tenfold in the long term. One bad hire can make your life much worse than investing all this time.

Are you willing to go through restructuring? Assuming you are not impacted by a restructure, as a manager, you are the bridge between your team and the company. Whether it’s realigning team priorities, redefining roles, or living through layoffs, you play a critical role in ensuring that transitions are smooth and that the team comes out stronger on the other side.

5) You have to embrace conflict

Because conflict is inevitable.

You can use some types of conflict for growth. There is a part of conflict which is super easy to handle: the geeky brainstorming or challenging other’s solutions. Everyone knows the goal and is generally happy to share ideas. Thanks to the "soft side of conflict", there is an opportunity to facilitate open communication, encourage diverse viewpoints, and foster collaboration. As a manager you don’t only need to be the one listening, you need to create the environment so that this "soft conflict" can happen.

You can’t please everyone. However, there is the "dark side of conflict". Conflicts between your direct report and yourself or between colleagues. In conflict situations, it’s crucial to strike a balance between being empathetic and maintaining your authority. While you want to show understanding and validate team members’ feelings, you also need to ensure that resolutions align with company values. No one prepares you for these challenges – remember, growth as a manager isn’t just about reading books; it’s about navigating the complexities of human interactions.

You can’t shy away from conflict. I know that interpersonal conflict is draining but dealing with conflict is part of your job. In software engineering, when you run into bugs, you can complain and moan, but you go and work it out. And if you don’t work it out and the server keeps crashing, then you get into bigger and bigger problems. The same happens with conflict. If you don’t take it seriously and deal with it head-on, conflict between people can become the reason to attrition, too many shifting priorities or lower quality output levels. It is easy to say, "dude, this is not my problem, go work it out yourself". Bad news for you. Like it or not, it is your problem too.

The 5 things summarised

Stepping into management isn’t just a shift in title – it’s a shift in mindset, priorities, and responsibilities. It’s about how well you empower others, balance growth with delivery, and navigate the challenges that come with leading people. The journey to becoming a successful manager is not a straight line, and no book or framework can fully prepare you for it. It’s in the daily experiences – both good and bad – that you’ll learn the most. (PS: mentors can help too)

Have you thought about the danger zone?

Before you make the leap into management, it’s crucial to understand that isn’t for everyone. There are 2 dimensions I like to cover with those looking to make the transition:

• The battery filling exercise
• The premature leap assessment.

Let’s cover them in a bit more detail.

The battery filling exercise

What energises you?

Here is how the exercise works. Take a sheet of paper and make two columns. On one side, list the tasks and responsibilities that fill your energy – things that you enjoy doing, that make you feel productive and fulfilled. On the other side, write down the activities that drain your energy. Be brutally honest here; the goal is to understand what drives your motivation.

Below you can see what energises and drains me. The goal is that my day as a manager has more of the battery filling than the battery draining.

The premature leap assessment.

Are you really ready?

Management is not just about leading people – it’s about leading technically proficient people. If you haven’t built a solid foundation of hands-on technical experience, you’ll struggle to keep up with your team and might lose credibility when offering guidance.

This is why you typically don’t see junior data scientists becoming managers. The transition usually happens from a senior or lead role, where you’ve accumulated enough technical depth and breadth to understand the challenges your team faces.

The danger zone

If you jump into management too early, you can end up in what I call the "danger zone." Here’s why: once you’re in a managerial role, you’re often bogged down with meetings, stakeholder management, and team leadership tasks, leaving little time to advance your own technical skills. And if you haven’t given yourself enough time to develop those skills before the leap, you might find yourself in a frustrating limbo – you can’t advance as a technical expert, but you also haven’t fully mastered management.

What’s worse, going back to an individual contributor role after you’ve made this leap isn’t always simple. Companies may be hesitant to move you back, and you could face the challenge of having to rebuild your technical credibility. Essentially, you’re in a position where you’re neither fully advancing in technical mastery nor thriving as a manager.

However, becoming a manager opens many new and exciting opportunities

Up until this point in the article, things seem to be daunting and negative. Yes, there are unique demands, but the rewards can be incredibly fulfilling if you lean into them. Becoming a manager can open many doors you haven’t considered.

• A larger impact on business strategy. As a manager, you be the one influencing what comes next. As an IC, you might tend to be given a problem to solve, but not have much to say about what problem to solve.
• An opportunity to shape and cultivate talent. You get to build and mentor a team. I have done this twice over, going from 0 data scientists to 6+ in each team. And the feeling you get reaching these milestones is incredible.
• A pseudo-PM roles. Your work as a manager aligns closely with product management. You never know if becoming a PM can be an option in the future.
• A stronger network. You tend to interact with senior leadership and cross-functional teams allow you to build a powerful network that can open doors for future career growth.
• A unique skill set. With fewer managers than ICs, you develop rare skills in leadership, people management, and process improvement. Your skillset is one of the highest valued out there, so becoming proficient can help you in your future career.

Summary

Key takeaways before moving into Data Science management:

• Reflect on your motivation. Are you chasing recognition, stuck in a role, or feeling pressured? Be clear on why you’re considering management, and weigh the trade-offs.
• "Manager" is not a promotion. In tech, transitioning to management is typically a lateral move. It comes with new responsibilities but not necessarily more power or pay.
• Management shifts your focus. Your role moves from hands-on technical work to guiding strategy, mentoring, and removing roadblocks.
• Success isn’t about you anymore. Your performance as a manager is tied to your team’s success, not your individual output. Empowering them is the real win.
• Experience, not books, fuels growth. Theories can help, but real lessons in management come from dealing with team dynamics, conflict, and decision-making.
• It’s about developing people and the discipline. Are you ready to prioritise team growth, effective hiring, and sometimes handling difficult restructuring?
• Prepare for conflict. Conflict is inevitable – both the constructive, idea-driven kind and the challenging interpersonal issues. Learn to embrace and navigate both.
• Mind the danger zone. Jumping into management too early can trap you in a space where you can’t advance technically, yet you’re not experienced enough as a leader.
• New opportunities in management. From influencing business strategy and building strong networks to potentially transitioning into product management, becoming a manager opens doors you may not have considered.

Further reading

Thanks for reading the article! If you are interested in more of my written content, here is an article capturing all of my other blogs posts organised by themes: Data Science team and project management, Data storytelling, Marketing & bidding science and Machine Learning & modelling.

All my written articles in one place

Stay tuned!

If you want to get notified when I release new written content, feel free to follow me on Medium or subscribe to my Substack newsletter. In addition, I would be very happy to chat on Linkedin!

Get notified with my latest written content about Data Science!

Jose’s Substack | Jose Parreño Garcia | Substack

---

_This post was originally published in the Senior Data Science Lead Substack newsletter_