I was early Marketing at Uber spending 5 years building Brand, Performance, and Lifecycle Marketing Data Science teams. I got to work with & learn from world-class marketing, data science, and martech leaders. Over the course of my career there I've measured over $1Bn in Marketing Spend and estimate that I've helped drive $200M+ in incremental revenue and cost savings . 

Then I moved on to a few early stage startups and scaleups (both as a FTE and Consultant) where I learned how to be even scrappier when building Marketing. 

Now I help Marketers make smarter decisions through my newsletter & podcast on Consumer Tech Marketing with how-to guides, case studies, and stories. 

I'm doing this because I love the marketing stories and people behind it in a world where Product seems to get most of the credit. There's so much magic in Marketing and I want to share more about it. 

Check it out here → https://experimental.beehiiv.com/

Outside of my newsletter and podcast, I love spending time with my 2 beautiful kids and my incredible wife in Amsterdam. We used to love traveling and eating out but you know... kids ...  so now I love cooking and exploring all the fun things little kids can do!

Sundar Swaminathan

One of the hardest things about having kids has been reconciling my professional identity before vs after. 

Before, I cared so much about my career. 

Being raised Indian and American, what I did defined who I was.

Even as a kid, you’re asked “what do you want to be when you grow up?” 

“Doctor. President. Astronaut”

What you wanted to BE was what you DID. 

Early career, I’d chase promotions and titles and cared about so many small things. 

Post kids, I couldn’t give a sh*t. 

And I don’t mean I want to do sh*t work. 

I still have pride in my work. 

I mean I don’t care whether I get that next promotion or whether I am “changing the world” or working on something cool. 

I don’t have career ambitions. 

I just want to hang out with my kids. 

And I am working to live not living to work. 

But the gap between how much I care about my kids vs my career is widening. 

So much so that I now have an aversion to the idea of going back to a job. 

Like actually disgusted. 

Lol my friends have described this as me just having a “mid life crisis” but I think it’s something deeper.

It’s an existential one. 

One where everything I had been working for 15 years towards now carries zero meaning and significance in my current view of life.

Former Head of Marketing Science at Uber where I optimized $1Bn+ in spend across Brand, Performance, and Lifecycle → ✏️ & 🎙️ on Growth and Marketing in Consumer Tech

No alternative text description for this image

I’ve been using A.I. for over 30 years.

I use it for literally everything:

- Writing this post 
- Finding new recipes
- SQL when I was an FTE
- Speeches when I present 

The list goes on and on. 

But the weird thing is every time I tell someone I’ve been using A.I. that long they question me. 

They call me a liar.

That I’m just saying it for the likes. 

It hurts. 

But I can’t do anything to change. 

I was born with Average Intelligence. 

I was born with A.I. 

It’s who I am. 

Note: this post was written with A.I.

Former Head of Brand Science at Uber | Optimized $1Bn+ of Marketing | Author and host of experiMENTAL ✏️ & 🎙️ on Marketing Strategy and Science

CEO: Simple question.Should we do brand marketing?

CMO: Simple answer. Will you give me 6 months without pulling the plug?

CEO: That's a long time.

CMO: That's likely how long it takes to see brand impact compound. If you'll panic before then I'm not going to even bother. 

CEO: What happens at month 6?

CMO: Your performance CAC will do one of two things:
1. We either keep the same volume and lower CAC 
OR
2. We keep the same CAC and do higher volume. 

Depends on what our strategy and goals are.

CEO: How do we know it's working before? 

CMO: Proxy metrics, share of search starting to go up, increased "direct" traffic.

CEO: Those sound fluffy...

CMO: This is exactly why I asked for 6 months of you backing off, but here's another question for you that might change your mind. 

Are you happy with the current performance?

CEO: HECK NO! 1. We're burning profit 2. Our CACs are skyrocketing 3. The quality of users is really going down. 

CMO: And why do you think that is?

CEO: Idk.

CMO: I do. We ran out of people that know about our product after we picked all the low hanging fruit. 

CEO: Ugh that sounds right. We've been in market for a few years.

CMO: Exactly, but our product is awesome and we love it and we need more people to know about it.

CEO: Hmmm that makes sense to me.

CMO: Here's the best part... our competitors aren't doing this which means we can get a head start and start building a moat that both improves our growth AND slows theres down.

CEO: Ohhhhhh siiiiiick! Okay, you're right! Let's f*cking go! 

CMO: Heck yes... just remember...6 months. 

--
🔔 Follow Sundar for Marketing Strategy & Science
♻️ Repost if you think others will find this useful
⬇️ Save so you can revisit this post later

CMO: We need to invest in marketing data science. 

CEO: Why now? What changed?

CMO: We did. At Series A, we had 1 channel and 1 market. Now we have 3 channels across 10 countries.

CEO: So just add more rows to the dashboard.

CMO: Not good enough. The measurement approach that got us here breaks at our current scale.

CEO: What do you mean breaks?

CMO: At Series A, I could eyeball attribution and make calls. Now, channels are cannibalizing each other, and I can't tell which ones are actually driving incremental growth.

CEO: Give me an example.

CMO: We're spending in paid social and paid search. Both show great ROAS. But when we scale one up, the other's conversions drop. They're fighting over the same customers.

CEO: So what do we need?

CMO: Incrementality testing. Holdouts. Geo experiments. The stuff that actually measures causation, not just correlation. But, I can't do it myself because I don't know how to. We need to bring in data science.

CEO: That sounds expensive.

CMO: Way cheaper than wasting budget on channels that aren't working. At our scale, a 10% measurement error is a seven-figure mistake.

CEO: Why didn't we need this before?

CMO: At Series A, our biggest risk was moving too slow. At Series C, it's moving confidently in the wrong direction.

CEO: How do justify upgrading our measurement?

CMO: The cost of being wrong is bigger than the cost of certainty. We're past that point.

CEO: Alright. What do you need from me?

CMO: Budget for proper testing infrastructure including data engineering and patience for answers that take weeks not hours.

CEO: Done. Now why are you texting me at 2AM?
 
CMO: Because wasted marketing spend never rests. 

--
🔔 Follow @Sundar for Marketing Strategy & Science
♻️ Repost if you think others will find this useful
⬇️ Save so you can revisit this post later

CEO: I need 30% more keeping CAC flat by Q3. 

CMO: Sorry, that would break physics. 

CEO: What do you mean?

CMO: Between high volume, fast speed, and low cost, you can only pick two. 

CEO: Explain more.

CMO: Sure thing. If we want:

High volume + fast speed → High cost

The only way to pump up volume quickly is inject more spend but it'll be inefficient and so cost will spike.

High volume + low cost → Slow speed

To get high volume at low costs we'll need to do everything from better campaign structures, experimentation, data pipelines, and even upper funnel campaigns. That'll take time.

Fast speed + low cost → Low volume

If we need to get to lower costs quickly, the easiest way to do that is to cut spend which means low volume. 

CEO: So there's no way to get high volume, fast speed, and low cost?!

CMO: No. My hands are tied. It's literally Marketing physics. Anyone, including your other CEO friends, that says they can do this is lying or hiding something. 

CEO: Crap. Are there other laws of Marketing physics?

CMO: Yes. At least four more. You should read this article by Sundar. He writes about Marketing Strategy and Science from his time at Uber.

CEO: Sounds great. Send it over.

CMO: Here you go → https://lnkd.in/gG3UkXnm

Jr Marketer: Hey, I'd love to become more data driven 
Manager: *slacks 10 articles below*
Jr Marketer: *1 day later...* Whoa. 🤯. IQ +1000

--

1. Everyone tells Marketers to be more data-driven, but not everyone knows where to start. Here are 10 of the most common analytics terms that's a good starting place.
https://lnkd.in/gZA9-wce

2. We need to get smarter about measuring marketing and here are 5 different tactics we can use to measure what we're doing.
https://lnkd.in/gw3mXHGh

3. MAU is a KPI for us but we've been measuring it all wrong. This is one of my favorites on hwo to properly calculate MAU.
https://lnkd.in/g9MEku7h

4. As we transition to increasing personalization and need smarter segmentation, I really reading how Uber did it with their 600M+ customers.
https://lnkd.in/gHXiYMSg

5. We're coming up in 2026 planning and this one helped me shape how we're going to be doing planning and target setting for our team. 
https://lnkd.in/gmH8VV8T

6. Given we don't always have analytics resources at our disposal this one is great for when we need to run a pre / post analysis ourselves. It's not about perfection but just slightly smarter decisions.
https://lnkd.in/gbbw4yz7

7. We know how important LTV is to everything we're doing but turns out we had been doing it all wrong. I took this article to our Finance team and we've rebuilt it.
https://lnkd.in/gxUqcxcw

8. This one is just WHOA! This article is why we've added a measurement strategy to our campaign briefs and why our data science team loves working with us again.
https://lnkd.in/gPrr-tAF

9. This is a complement to the LTV article above but it's how we're defining CAC moving forward. It helped me align with Finance and why we're now BFFS.
https://lnkd.in/gmvsqxvF

10. This is just a fun article on how Duolingo turned around slow growth and grew 900% by being more data driven
https://lnkd.in/gWDnDj9X

Former Head of Brand Science at Uber | Author and host of experiMENTAL ✏️ & 🎙️ on Growth and Marketing in Consumer Tech

CFO: Our attribution model is amazing!

New CMO: Nice! I won’t be using it much. 

CFO: What? This cost us a ton of $$

CMO: It tells us what happened, not what we caused.

CFO: What’s the difference?

CMO: Attribution says Facebook drove 1,000 conversions last month. It doesn’t tell us how many of those 1,000 would’ve converted anyway without Facebook.

CFO: Wait. You’re saying the model is wrong?

CMO: I’m saying it’s precise and useless. It gives us confidence in the wrong answer.

CFO: How is that possible?

CMO: Because customers see ads everywhere. Attribution gives credit. Incrementality measures impact. Facebook might get last-click credit, but the customer already decided to buy after seeing our TV spot.

CFO: So we’ve been optimizing for the wrong thing?

CMO: We’ve been optimizing for who gets credit, not who deserves it.

CFO: Then what should we be using?

CMO: Holdout tests. Geo experiments. Incrementality studies. They’re messier, take longer, but actually tell us what’s working.

CFO: Why didn’t we start with that?

CMO: I don’t know but I just got here. Most of the time we confuse precision for confidence.

CFO: I’d rather be roughly right than precisely wrong.

CMO: That’s the mindset shift. We’re gonna fix this! 
–
🔔 Follow Sundar for Marketing Strategy & Science
♻️ Repost if you think others will find this useful
⬇️ Save so you can revisit this later

Who reads my newsletter on Marketing in Consumer tech? 

Marketers in consumer tech. 
Also data scientist.
And founders. 

When I set out to write this newsletter, the biggest doubt I had (which is most people’s fear) is that no one would read it.

I’d just be writing into a void and while it gave me joy, it wouldn’t be a business. 

Almost 50 articles later and a year into it, I know that’s no longer true. 

50+ articles. 
2,800 readers.
60% open rate. 

And most importantly?! 

The audience I’m writing for is who reads it.

Marketing leaders across the funnel from Brand to lifecycle at companies like Uber, DoorDash, Apple, AirBNB, Spotify, Google and so many more. 

It’s just motivating to be able to take a step back and see the daily efforts translate to something. 

Anyway, Im excited and think my best work is yet to come. 

Thank you to everyone that supports me! 

Onwards and upwards.

CMO: We need to A/B test this launch!

Me: We can't. But here's what we can do...

Propensity matching.

CMO: Wait, why can't we A/B test?

Me: I don't think we want to hide the launch of our loyalty program from 50% of our users. 

CMO: Fair point. But, I'm being asked to show the ROI of the loyalty program. Will we be able to do that?

Me: Yes! We can use propensity matching.

CMO: Say more fam.

Me: So in a normal A/B test, we randomize users before and then measure the difference. 

Propensity matching is the opposite.

We start by looking at the users we want to analyze, which in this case is everyone who signed up for the loyalty program.

Then we look at everyone who USED to behave the same but didn't sign up for the loyalty program.

Then we measure the difference in their behaviors. 

Here's an example:
User 1 starts as an X then becomes a O.
User 2 starts as an X and stays as an X. 
We measure the difference between O and X. 

X → O (Treatment)
X → X (Control)

CMO: Woah! Basically it's a way for us to compare more apples to apples without an A/B test.

Me: Yes, exactly. 

CMO: This is awesome. Thanks for saving my a*s

Me: No problem. If you want to read more check out this article I sent to 2,800+ Marketers → https://lnkd.in/gcHsSpqy

CMO: Damn, we are not data driven at all.

CEO: What do you mean?! I've tried to be since day 1!

CMO: Because no one changes their mind based on the data.

CEO: I don't understand. We hired data scientists. We have dashboards everywhere. How are we not data-driven?

CMO: Again, because no one changes their mind based on the data.

CEO: That's not true. We look at reports every week.

CMO: Looking at data isn't the same as being driven by it.

CEO: What's the difference?

CMO: Last month, we tested two pricing models. The data showed Model B was 8% better. We still went with Model A.

CEO: We had strategic reasons for that.

CMO: Exactly. That's being opinion-driven with data decoration. Nothing wrong with it, but let's not pretend it's something else.

CEO: So what does data-driven actually look like?

CMO: It means you make the uncomfortable decision when data contradicts your intuition. Most companies can't stomach that.

CEO: Give me another example.

CMO: Our best performing acquisition channel according to dashboards? I wanted to cut it. You pushed back because "it's always worked." That's not data-driven.

CEO: But you didn't have proof it wasn't working.

CMO: Our team had a solid analysis that suggested it was plus experience. We wanted to run an incrementality test proving it out. You wanted more proof. That's the problem. Data-driven companies trust the data AND run tests to confirm hypotheses. You just wanted a narrative that felt right.

CEO: This is uncomfortable to hear.

CMO: I know. It’s not easy for me to says That's how we know we're getting somewhere. We say we want data-driven culture but actually want data-supported decisions.

CEO: What's it take to actually become data-driven?

CMO: Start saying "the data changed my mind" more than "the data confirms what I thought." Been reading Sundar's posts on this. He calls out the theater constantly.

CEO: This Sundar guy is on to something. Send me his LI profile.

CMO: Sent below 
--
🔔 Follow Sundar for Marketing Strategy & Science
♻️ Repost if you think others will find this useful
⬇️ Save so you can revisit this post later

New CMO: Why do we separate brand and performance?

CEO: Because they’re different things.

CMO: How so?

CEO: Brand is long-term and hard to measure. Performance is short-term and drives revenue. We split the budget 30/70 to keep it clean.

CMO: That’s the problem. You’re acting like they work in isolation.

CEO: They don’t?

CMO: Our performance channels work better when brand awareness is high. And brand campaigns drive some immediate conversions. They’re not separate games.

CEO: So how should we think about them?

CMO: Stop asking “is this brand or performance?” Start asking “what’s the full impact of this investment?”

CEO: That sounds complicated.

CMO: It is. But pretending they’re separate is why companies think brand doesn’t work.

CEO: Explain.

CMO: We run a brand campaign. Performance metrics don’t move much immediately. You call it a failure. But we didn’t measure if it lowered our CAC or improved conversion rates on paid channels.

CEO: How would we measure that?

CMO: Geo holdouts. Test brand in some markets and not others. Then measure the impact on all channels, not just brand-attributed conversions.

CEO: Why doesn’t every company do this?

CMO: Because it’s easier to keep them in silos. Brand team reports on awareness. Performance team reports on ROAS. No one connects the dots.

CEO: We’ve been doing exactly that.

CMO: I know. Time to change it up. 
–
🔔 Follow Sundar for Marketing Strategy & Science
♻️ Repost if you think others will find this useful
⬇️ Save so you can revisit later

experiMENTAL Newsletter & Podcast  has 500 downloads! 

If you’re a consumer tech marketer looking to learn what works class teams are doing, I biasedly think this is the best place to start.

I cover everything from brand building to deep data science with leaders from Uber , Spotify , Anthropic , Wayfair , and so many more amazing companies! 

I’m now up to 10 episodes out too so you don’t need to just listen to one episode and then get sad you can’t keep bingeing. 

Start listening now: https://lnkd.in/gFGjqJ3e

Or watch : https://lnkd.in/g8VPT8MD

Here are 4 stories from my time leading Uber's Brand Data Science team that will change your mind about the impact of Brand Marketing:

All of these stories are empirical with data backed causal studies and not theoretical. 

1. DoorDash kicks our b*tt
In 2018, UberEATS and DoorDash were neck in neck in the US. 

Now? It has 66% marketshare.

The most readily obvious reasons are selection and focusing on suburbs. These are definitely primary reasons but they also outspent us on Brand Marketing early. 

They were on TV and OOH faster and bigger which is exactly where suburban audiences consume and it made them the entrenched preference.

The UberEATS team is world class but once that brand moat sets in it becomes a tough uphill battle. 

2. Needing to be always on

We ran a brand campaign in the UK where we were testing a new concept for Uber as a whole.

During the campaign, awareness went up massively. 

Think XX%. Signups did too across Rides and Eats.

But we then had to pause the campaign and within a month that awareness came plummeting back to the pre campaign base line. 

And signups? You guessed it. 📉

We then advocated for always on budget. 

3. Know where your audience is

In India, we were launching a new campaign around women's safety in Uber Auto.

The classic perf marketing channels work but not as well as other channels especially in dense cities in India and we wanted the message to be more public.

So OOH and TV were where we invested and we saw Auto trips go up by 6% within a short period. 

4. In Australia, we went all in with a campaign called "Tonight I'll be Eating" for UberEATS that featured Kim Kardashian and a well known Australian comedian. 

The campaign took off! 

We were hitting record bests and the impact was immediate cementing our place as the #1 app in Australia.

I've got a dozen more stories, but the point is as someone who was lucky enough to causally measure these campaigns, I got to see the impact of Brand at a scale that no other marketing type can touch. 

Brand campaigns have massive impact.

You just need to figure out how.
--
🔔 Follow Sundar for Marketing Strategy & Science
♻️ Repost if you think others will find this useful
⬇️ Save so you can revisit this post later

Sebastian Hewing 🚀 is my next guest on experiMENTAL Newsletter & Podcast! 

He's a data and AI advisor to VC and PE-backed startups after working at some of the fastest growing startups ever with Rocket Internet SE.

On today's episode we chat about: 
• how to build data setups that empower, not babysit, your stakeholders
• a three-step framework that’s helped 40+ companies
• best traits of high-leverage your first analytics hire
• why solution-first thinking fails data teams

My favorite quotes from our conversation:

"I always implement a role that I call the super user, which
is a role that that I think is lacking in a lot of data setups and causes all kinds of problems."

"I think it's a very poetic point. The data advisor is not coming in for the data analyst they're coming for to help out another function."

"I think there's the pre AI problems and then there's the pre AI challenges and then there's the additional challenges that come with everything that's happening around AI."

Sebastian is by far one of the best data leaders I've had the pleasure of chatting with and if you're either a data leader looking to uplevel or a stakeholder looking to learn how to best implement data practices, this one's for you.

Tune in wherever you love listening to podcasts:
Listen: https://lnkd.in/ebSGnypU
Watch: https://lnkd.in/gPacUgjP
Subscribe: https://lnkd.in/gPacUgjP

CEO: Revenue is up but margins are down?!

New CMO: Yup, we’re just renting growth.

CEO: Explain. 

CMO: There’s a difference between earning growth and renting it.

When you started, CAC was $50 and customers had strong word-of-mouth. We were earning growth. Now CAC is $120 and referrals are flat. We’re renting it and paying the paid tax. 

CEO: So we’re spending more to acquire customers?

CMO: Worse. We’re spending more to acquire lower-quality customers. The easy audience is tapped out.

CEO: How do you know they’re lower quality?

CMO: Day 30 retention dropped from 60% to 45%. LTV is down 25%. We’re hitting diminishing returns on paid channels.

CEO: What happens if we keep scaling spend?

CMO: Revenue goes up. Profitability goes down. Eventually, we’re growing unprofitably, and the board pulls the plug.

CEO: So we stop spending?

CMO: No. We need to slow down marketing. We fix the unit economics first. Better onboarding, improved retention, stronger referral mechanics. We also need to invest in Brand Marketinf after. Then we scale again. 

CEO: Why can’t we do both at the same time?

CMO: Because resources, priorities, and bad growth masks the problem. If revenue is climbing, no one feels urgency to fix retention. Then one day, the channels stop working and we realize we built on sand.

CEO: How do we know when we’ve hit this wall?

CMO: When CAC is rising faster than LTV. When our payback period gets too long. That’s the signal.
–
🔔 Follow Sundar for Marketing Strategy & Science
♻️ Repost if you think others will find this useful
⬇️ Save so you can revisit this post later

Measuring marketing is NOT about the latest and greatest tools.

A successful strategy has 4 components: 

1. Goals
2. People
3. Culture
4. Channels

Goals 
⮑ What is the business trying to accomplish? 
⮑ How does Marketing fit in?

People
⮑ What skillsets do you have?
⮑ What's the hiring roadmap look like? 

Culture
⮑ Are you willing to experiment?
⮑ Does leadership understand the risks?

Channels
⮑ What are the main channels used?
⮑ Are we launching new ones?

If you find yourself struggling to prove the ROI of Marketing, take a step back and figure out which of these 4 places is lacking. 

The most successful Marketing Science teams start at the top. 

The tools are never the answer. 

What’s the number one thing you focus on to implement an effective measurement strategy? 

--
🔔 Follow Sundar for Marketing Strategy and Science
♻️ Repost if you think others will find this useful
⬇️ Save so you can revisit this later

This visual perfectly captures what building experiMENTAL Newsletter & Podcast has felt like.

After my short stint as an FTE this year, I was super stoked to go full steam ahead on my newsletter and podcast about Marketing in Consumer Tech. 

It let me chase the joy of sharing everything I've learned in my career while having the ultimate flexibility to hang out with my kids and wife.

But then you run your family's financial model.

And reality sets in that your passion has to be a business.

Writing an article goes from shipping for fun to shipping and tracking if people opened, read, and converted. I'm microanalyzing every survey response or piece of feedback. 

It sucks the fun out of writing and I couldn't put my finger on why until I saw this chart from Peter Walker.

Simply, the largest change in happiness happens after "Initial Enthusiasm" when it becomes "Reality sets in". 

It has by far the steepest slope (1st derivative for you math nerds). 

And that's where I'm at.

It's scary. 

And the only thing that gets me through? 

Unwavering faith in my work. 

Faking it til I'm making it. 

Experimenting. Pivoting. Grinding.

And now I get why founders have this insane ability to force visions and missions because that's what's required.

Not only do I want to believe I produce the best content on Marketing Strategy & Science in Consumer Tech, but I have to. 

I have to believe my sh*t is worth reading.

I have to believe it's worth buying.

I have to believe it's worth it.

Because then everyone else will too.

Love, 
Sundar

Dear post reader, 

Happy Thanksgiving from 0-3 weeks ago! 

I can not control when this surfaces, but I hope you have fully recovered from the binge eating, career sh*t talking, and your football team’s loss. 

I also hope that you were able to celebrate with everyone you were thankful for and that you recognize the good things going on in your life even if it’s surrounded by some other bad. 

Happy Thanksgiving! 🍽 🦃 🥘 🌽 

Sincerely,
Sundar

My next podcast guest is Mattia Santin, Senior VP of Marketing at Contentsquare (ex Hotjar, Uber)! 

On today's episode we chat about: 
• the career leaps that shaped his approach 
• what happens when CMOs get their hands dirty 
• why too many marketing teams mistake motion for progress 

and much more.

My favorite quotes from our conversation:

"The first thing I do is spend as much time as possible talking to people"

"You have to go down, into the trenches also in the numbers and understanding, making your own assessment of what, what you're seeing"

"Everybody was pushing me to spend more. It, it didn't feel right, in part because I never had that experience, in part because I didn't feel comfortable in our measurement and our ability to understand truly ROI."

Mattia is a brilliant strategist AND executor knowing how to drop in and build impactful Marketing teams. If you’re looking to get a leg up on how to navigate strategy, scale, or structure inside a fast-growing company, this one’s for you.

Tune in wherever you love listening to podcasts:
Listen: https://lnkd.in/gFGjqJ3e
Watch: https://lnkd.in/gueY5bYq
Subscribe: https://lnkd.in/gxR5c4Xd

‼️Launching accelerator: babY Combinator

Focused on seed and preseed stage for founders aged 7 and below. 

The new generation is born with AI from day 1 and understand the technology better than anyone.

In addition, when they build products for their peers, that demographic will have the highest LTV with an average of 75-80 years ahead of them. 

The math is undeniable. 

There of course will be some challenges:
- Additional facilities for diapers
- Cofounder tantrums 
- Naps 

But we believe we can over come them and create the next generation of truly generational founders. 

It is never too young to ride the AI wave. 

DM me if you want to be an LP. 

If you’re sold, skip the convo and Venmo me

Sebastian Hewing 🚀

Data Action Mentor

Building masterclasses and communities for data leaders and future data leaders who want to create outsized business impact with data & AI.

Founder

Sebastian Hewing Consulting

🚀 I build impactful, AI-first data infrastructure for VC/PE-backed consumer companies.

Previous clients include Zalando, Lazada, Global Fashion Group (Dafiti.com, Lamoda.ru, Jabong.com, TheIconic.com.au, Zalora.com), Linio, Jumia, Lieferando, eBay, Telefonica o2, HUK Coburg, Axel Springer, Lendico, OCC Assekuradeur, Creadi AG, Boerse Stuttgart, Bergfreunde.de / Backcountry.com, Sanity Group, Takeaway.com, Wasoko, Westwing, Beat81 and more

Owner

Locafox GmbH

Raised EUR 10+M from established Investors such as Holtzbrinck Ventures to digitize local retail and build Europe's largest omni-channel shopping marketplace

- Built up Business Intelligence and Analytics Team from Scratch
- Established a true Data Self Service Environment (Around 90% of Analytics Requests could be handled independently by Functional Departments)
- Managed and Led Market Entry to Spain
- Spain contributed 1/3 of the Total Revenues after 6 Months

Director Business Intelligence & International

Rocket Internet SE

Rocket Internet is the world's largest Internet incubator. Our team has been building online companies since 1999 and has created over 100 market leading companies in 50+ countries, dozens of which have been exited successfully. Examples: Zalando, Lazada, Delivery Hero, HelloFresh.

- Implemented and customised 7 Data Warehouses (DWHs) for Rocket Internet’s global portfolio
companies.
- Product Owner for Analytics Applications such as split-testing, catalogue
sorting algorithms, product recommendation algorithms, marketing attribution
modelling and predictive modelling.
- Built and led the global Data Team. Disciplinary lead for 10 central
employees and functional lead for 40+ employees in 7 Portfolio companies.
- Increased Revenue per Visitor of 3 Portfolio companies by 10-12% with the
help of a custom-built product recommendation algorithm.
- Management of the central Analytics Budget for 50+ Portfolio companies.

Head of Business Analytics

SHS VIVEON AG

One of Europe's leading boutique consulting firms for Data Analytics, Data Warehousing and Customer Management.

- Managed a variety of projects with focus on customer analytics; e.g. customer
segmentation, churn prevention and development of customer scorecards.
- Accountable for the development of the Multi-Channel Marketing practice area,
specializing in the field of Marketing-Performance-Monitoring and -Optimisation.
- Fast career progression, receiving semi-annual promotions.

Consultant

InsightAsia Research Group / DemandROMI (JV)

built up a pan-Asian consulting division focusing on Marketing & Sales Performance

Co-Founder

Accenture

conducted emprical study about success factors of Multi-Channel Management in Banking as part of final thesis

Business Analyst

Founder Data Action Mentor | Swiss Army Knife for Data & Analytics | Ex-Rocket Internet

I'm a mix of data engineer, entrepreneur, product manager and marketer. 🛠️🚀🧑‍💻🎯

I help tomorrow's data leaders create outsized business impact by building AI-first, 10X data teams.

-> Find out more at www.dataactionmentor.com

In the last 17+ years, I worked as a Data Practitioner and Data Leader for Startups, Corporates, and in Consulting across all continents. 

As Head of Business Analytics at Rocket Internet, the world's largest venture builder, I built and led Data & Analytics Teams and Infrastructure for 7 portfolio companies in 15 countries.

Rocket Internet scaled companies such as Zalando (Europe's leading online fashion platform), Delivery Hero (a leading food delivery platform), HelloFresh (the world's leading meal kit delivery company), Lazada (the Amazon.com of SouthEast Asia, exited to Alibaba) and Jumia (the first African company on the NYSE). 

As a consultant and advisor to Founders and CEOs, I built or re-built the first data team and data infrastructure for 50+ leading consumer companies.

P.S.: My biggest passion next to Data & AI is Traveling. I live on the beautiful island of Taiwan 🇹🇼 but I typically spend 5-6 months per year traveling to hang out with my awesome global clients. I have been to 100+ countries and territories.

Founder & Mentor

Solve all your Data- & Analytics Challenges and connect with experienced Data & AI Professionals for a price and time investment that every startup can afford

“Let’s define a data strategy.”

Translation: let’s argue about tools for 6 months.

Here’s the thing nobody wants to admit:

Most data strategies are just about data stacks.
Not actual strategy.

If you want a strategy that survives the next reorg, AI hype wave, or exec sponsor change - start here:

1. Problem: What are we solving?
2. Users: Who are you serving?
3. UVP: What's our data team's secret sauce?
4. Solution: What are we building and what are we NOT building?
5. Distribution: How will we get this in front of users without chasing them down?
6. Systems: What helps this scale without breaking us?
7. Outcomes: What will users actually do differently?
8. Costs: What’s the real investment? (Spoiler: it’s not just the software)
9. People: How will we grow a team that wants to stick around?
10. Vision: What guides our decisions when priorities clash?

Answer those 10?

You’ve got a data strategy.

Ignore them?

You’ve got a dashboard backlog and a stack no one uses.

♻️ Repost if you've ever watched a “data strategy session” turn into a Power BI vs Tableau debate.

👉 And join 3,000+ data leaders who read my free newsletter for weekly tips on building impactful data teams in the AI-era: https://lnkd.in/gt_2Gc4s

Sebastian Hewing

10x ROI with data & AI foundations for VC/PE-backed companies | Hands-on Data Strategist | Solopreneur & Travel Addict (100+ countries & territories) 🌏

We’re in the middle of a dashboard pandemic.

Symptoms include:

→ 173 dashboards across 3 tools
→ Nobody agrees on the numbers
→ “Can I get this in Excel?” - every day
→ Burned-out data teams trapped in JIRA tickets for minor tweaks

Diagnosis?

Chronic dashboard overproduction with decision fatigue.

Because somewhere along the way, we confused outputs with outcomes.

We thought: more dashboards = more insights = more value.

But here’s the thing:

-Dashboards don’t drive decisions
-Dashboards don’t change behavior
-Dashboards don’t fix broken processes

They visualize. That’s it.

Want to really move the needle instead of pumping out prettier charts?

Start with this:

→ What decision are we trying to support?
→ Who owns it?
→ What happens if we don’t build it?

It’s not easy.

But it’s the cure.

♻️ Repost if your dashboards are multiplying faster than decisions are being made

👉 And join 3,000+ data leaders who read my free newsletter for actionable tips on building impactful data teams in the AI-era:
https://lnkd.in/gMSUaAsx

Putting together a bunch of tools is NOT a data strategy!

A data strategy is this: 👇 (refined over 18 years)

1. A problem to solve

Start by uncovering and deeply understanding stakeholder problems (not with tools and data stacks!)

2. A human who has the problem

Learn about the jobs-to-be-done your users are facing. Sit on their lap, if necessary.

3. A unique value proposition (UVP)

Clearly communicate why your users should drop Excel to use your services.

4. A solution

Craft solutions that solve your user’s problem(s) and take away their anxiety to try them (be accessible)

5. A distribution strategy

The best solution is worthless if your stakeholders don’t know about it. Let the world know why your data team rocks!

6. A systems strategy

The best data teams spend more time building systems than being a slave in a system. Don’t be a dashboard factory! Build a problem solving factory!

7. Outcomes

Most data teams only produce output - tickets solved, dashboards built yada yada. You need to produce dream outcomes for your stakeholders!

8. Cost Structure

A necessary puzzle piece, but a lot of data teams worry about it too much. Worry more about creating value than reducing cost!

9. People Strategy

Have a hiring and people management strategy that reflects your principles and values.

That’s all.

Easy peasy, lemon squeezy, right? 🍋

♻️ Repost if you’ve felt the pain of building dashboards while your stakeholders export to Excel.

👉 And join 3,000+ data leaders who read my free newsletter for actionable tips on building impactful data teams in the AI-era: https://lnkd.in/grgzCkUf

Building impactful data teams for the AI-age | Data Strategist & Engineer | Advisor to C-Level in VC/PE-backed companies | Solopreneur & Travel Addict (100+ countries & territories) 🌏

The best CMOs would crush it running a data team.

Because they get three things many data leaders struggle with:

1. Speed over perfection
2. Business impact over tooling
3. Influence over insight

I used to think marketing is all gut feeling and creativity.

But the best CMOs I’ve worked with proved me wrong.

→ They were fast.
→ They were analytical.
→ And they were obsessed with business impact.

Tomorrow, I’m breaking down the 6 traits that make top CMOs unstoppable - and how data teams can borrow those traits to become true business partners instead of dashboard factories.

Here’s a taste of what I’ll cover:

✅ How marketers act with speed despite messy, conflicting data
✅ Where data teams lose trust and how to fix it
✅ And what it really takes for data teams to earn a seat at the table

If you want to influence the business, you need to understand how high-performing marketers operate.

And tomorrow I’ll show you exactly how.

I’m sharing with 3,000+ data leaders how CMOs win - and how you can adopt their playbook to earn real influence as a data leader.

📩 Sign up for my FREE newsletter if you don’t want to miss it:
https://lnkd.in/g3WJTejm

They don’t hate dashboards.

They hate feeling stupid.

You built the scalable pipeline. The modular dbt models. The layered dashboards. The forecasting model.

And what do they do?

👉 “Can you export that to Excel?”

You think:

“Why do they ignore everything we built?”

But here’s a better question:

“Why did we build something they don’t want to use?”

Let’s be honest.

The villain isn’t the stakeholder.

It’s the complexity we’ve come to worship.

→ We throw AI at problems a simple SQL statement would solve.
→ We hide insight behind dropdowns, tabs, and filters - then wonder why nobody clicks.
→ We act like we’re building for Google.
→ But we’re a 200-person scale-up with 3 analysts and a CEO who checks Slack at 6am.

Complexity looks smart.

But it kills trust.

Simplicity is the real flex.

♻️ Repost if your CEO’s favorite BI tool has a green X icon.

👉 And join 3,000+ data leaders who read my FREE weekly tips on building impactful data teams in the AI-age:
https://lnkd.in/gHvQuqMc

“Our data stack has 16 tools.”

But stakeholders still don’t trust a single number.

It’s like owning 16 pairs of shoes and still feeling like you have nothing to wear.

You don’t have a tooling problem.

You have a clarity problem.

Data leaders keep falling into the same trap I once fell into:

→ Hire more engineers
→ Buy the fancy tech
→ Integrate all the data
→ Build a beautiful, scalable pipeline

…and still get weekly “WTF is this number?” emails from the CEO.

Here’s why:

🧠 You start with a tech stack, not with your users’ actual problems
🧩 You build infrastructure, but no one knows how it maps to business goals
🛠️ You design systems your team loves but your users hate

All of that makes your data team technically impressive and practically invisible.

Because trust isn’t built in dbt.

It’s built in moments like:

- “Ah, I get it now.”
- “That’s exactly what I needed.”
- “This matches what I see in Hubspot.”

If your data products don’t trigger these reactions, it doesn’t matter how “modular”, “real-time”, or “compliant” they are.

The goal isn’t to impress your peers.

The goal is to help your stakeholders live easier, calmer, better lives.

Want to become a true business partner with your data team?

👉 Join 3,000+ data leaders who read my free newsletter for weekly  tips on building impactful data teams in the AI-era:
https://lnkd.in/gETTkzRd

♻️ Repost if you’ve ever watched a stakeholder open Looker… and immediately go back to Excel.

The “single source of truth” is NOT a myth.

I used to think it was a fantasy too.

One clean, trusted dataset that aligns every team.

Like:
→ unicorns 🦄
→ dragons… 🐲
→ ...and matching revenue numbers across teams.

But here’s what changed my mind:

I realized most metric debates aren’t about the data.

They’re about perspective.

The “truth” is built from:
→ A measure: e.g. ARR
→ A filter: e.g. New ARR
→ A dimension: “Finance ARR”: invoice date, “Sales ARR”: deal closed date.

Sales wants to show momentum → “Closed deal date”

Finance wants precision → “Invoice date”

Same Measure.
Same Filter.
Different dimension.

Here's how I think about the single source of truth today:

Treat your metrics like Legos. Provide Measures, filters and dimensions in a single source of truth.

Then every domain can build their version from the same bricks.

You get governance + agility. 

Metrics aren’t magical. They’re just combinations.

♻️ Repost if you’ve ever lost sleep over Revenue definitions that almost matched

👉 And join 3,000+ data leaders who read my free newsletter for weekly tips on building impactful data teams in the AI-era:
https://lnkd.in/gJUCHAGz

Everyone wants to use AI to reduce headcount.

But here’s the inconvenient truth no one wants to hear:

👉 The biggest driver of bloated data teams isn’t lack of AI tools - it’s lack of foundations.

→ If your single source of truth is chaos…
→ If definitions and metrics change depending on who you ask…
→ If data governance resembles a police checkpoint (or doesn’t exist…)

Then AI won’t make you lean.

You’ll just end up automating the mess and hiring more people to clean it up.
Let’s get specific.

Here's a realistic, right-sized setup for a late-stage scale-up, assuming solid foundations:

- 173 Users (consume data via dashboards & reports)
- 17 Super Users (can self-serve simple insights) → 10:1 ratio
- 6 Analysts / Data Scientists (handle complex questions) → 3:1
- 2 Analytics Engineers (curate data for analysis) → 3:1
- 1 Data Engineer (builds the technical platform) → 2:1
- 1–2 Product Owners / Data PMs (the glue across domains)

That’s 11 people in the data org, or 5.5% of employees.

Let’s recap:

🚫 AI won’t save you from foundational debt
✅ AI can amplify your impact - if your data isn’t broken
Otherwise?

You’ll be automating confusion and scaling chaos.

♻️ Repost if someone just told you AI will replace analysts and fix your data problems magically.

👉 And join 3,000+ data leaders who read my free newsletter for weekly tips on building impactful data teams in the AI-era: https://lnkd.in/gVEHHj_T

taplio

The Battle

Sundar Swaminathan

Sebastian Hewing 🚀

Followers

Posts

Engagements

Likes

Performance ScoreSundar Swaminathan's LinkedIn Performance

Performance ScoreSebastian Hewing 🚀's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Sundar Swaminathan

Sebastian Hewing 🚀

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Sundar Swaminathan's LinkedIn Performance

Performance ScoreSebastian Hewing 🚀's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreSundar Swaminathan's LinkedIn Performance