Hello experiMENTAList, it’s Sundar and welcome to another article of my weekly newsletter! I’m a former Head of Marketing Science at Uber where I optimized $1Bn+ in spend across Brand, Performance, and Lifecycle.

Now, I’m diving into the CRAZY world of Growth in Consumer Tech helping you make smarter and more informed decisions.

Thank you for being part of this growing community of 2.7K+ growth and marketing leaders from Uber, DoorDash, Google, Spotify, and many more. If you know a colleague or friend that might benefit from this, send them here.

Now let’s step into the lab!

I've spent the last decade as a marketing data scientist, but as I approach the end of 2025, I've had the time to reflect on my career and realize that I'm no longer in love with being a data scientist.

Below I'll share a bit more about:

I started my career as a data scientist in 2015 before it was cool to be a data scientist. Before all the crazy salaries and competition and "professional fame" that came with being a data scientist. I couldn't have been more excited to become one.

What was even more exciting was how I became one. In 2015, I left the US Treasury after a two-year stint in finance. Before that, I did a two-year stint as a software developer. This was after going to school for finance, so for the first four years of my career, I was lost and trying to find what I was excited about.

I then had a friend who worked at DIRECTV reach out and ask if I knew anyone analytical, because they were hiring a marketing analyst.

I told them, "Well, shoot, I'm analytical. You just need to teach me the business. Everyone is talking about how hot digital marketing analytics is, so I'd love to apply for the role."

I did, and I got it.

After 10 months as a contractor at DIRECTV, Uber reached out and hired me in March 2016. I then had a phenomenal 5-year run where I started as a local rider-side marketing analyst in the DMV and left as their Global Head of Marketing Data Science. I built so many wonderful relationships and worked on some of the most challenging projects of my career.

After that, I went to a few startups and did some consulting before deciding that I wanted to go full-time into working on this newsletter and my podcast.

Overall, I've been extremely lucky and blessed with my career, but as I look at other fields and industries and just the broader nature of work today, I don't know if I would go back and make the same decisions.

Here's why:

As a data scientist, your upside is capped, and no matter what, unless you're in a very, very, very special place, you will never be in full control of the outcome of your work.

That's something that has fundamentally bothered me over the past few years as I've become a data science leader.

Here’s an example:

I pushed my team to deliver an analysis within a very constricted timeframe because I had a great conversation with our stakeholder, and we felt that this would be an impactful analysis that would help us make some concrete decisions. So I pushed them, and I pushed them, and I pushed them. We delivered the analysis, and without any sort of warning, the decision was pivoted away.

We decided to maintain the status quo.

Now, if this happens once or twice, that's fine. You can’t everything your way. But as you're reading this, I bet many data scientists are nodding their head and realizing this is not a uncommon pattern.

That lack of ownership and follow-through is extremely frustrating.

It means that when you start a career as a data scientist, you have a ceiling.

As a person that was raised to be ambitious and never stop growing, this really broke me.

So, to take back more control, I decided to become an advisor and consultant, but that wasn't necessarily any smoother. I struggled massively with positioning myself especially because I wanted to do little to no IC work.

When you're an early-stage company, you're not really looking for a data adviser or data strategist. Generally, you're looking for someone execution-focused. Then, when I tried to pitch myself to scale-ups as a fractional data leader, they would often respond with, "Okay, but we will just hire that," or, "It's not the right fit." And, of course, at enterprise, there's not really a need for a fractional data leader.

I had to reposition myself as either a growth leader or a marketing leader to even sell projects.

The challenge is that unless you do execution work/IC work as a consultant, it's very hard to sell yourself as a data leader. This is not true with other industries. If you’re in product, marketing, operations, finance, and engineering, you can become a fractional CPO, CMO, COO, CFO, and CTO respectively.

As a data leader, I instantly was pidgeonholed into executional work even though so many companies need data leadership.

It means, that post career, you have less optionality.

Finally, I think I've just been burnt out from all of the conversations I see on LinkedIn about data science. The part that most people don't understand is that data science is very little science and almost all art.

My friend Sebastian Hewing talks a lot about the human side, but it is continuously lacking in so many of the conversations.

People so often join data science because they either like to build models or dive into the numbers, but that's not what data science is about. Data science is about influencing decisions, and that human element is something that I've seen fade and fade to the point where people look at data science and expect data science teams to just be data monkeys.

Sadly, it’s a self fulfilling prophecy. Data scientists join because they like being data monkeys and then teams treat them like one.

If being called a data monkey is triggering you right now, take a step back and ask yourself if you’re doing what you can to position yourself as a strategic leader first that is blessed with analytical skills. Likely not.

It's a feeling I can't shake, and I don't have a lot of faith in the future of data science because I do believe much of it can be automated with AI. Those that are resilient and have stronger communication and business skills will be able to survive.

But, that's not an exciting future for me.

Unlike product growth or even marketing where there's a lot of hullabaloo about AI taking over, nothing is as likely to be taken over by AI as data science.

It means, that my future career is going to be limited and more competitive.

So when you put this all together, what does this mean for me? I don't know. And I'm taking the time to truly understand what my professional identity is. It doesn't help that now that I've had my second kid, my heart is completely with them.

So much so that the idea of being career-driven has almost completely fallen to the wayside.

Here's what I do know: I still love sharing my stories of being a data scientist at places like Uber and all the learnings I’ve learned over the years. I love the community I'm building here, and all the wonderful comments and feedback you've all given me around my work. So, I know this is where I'll continue to invest at least for the next year.

Thank you for reading and bearing with me.

It was nice to get these thoughts on paper 🙂

That’s it for this week!

Stay experiMENTAL,

Sundar