Hello experiMENTAList, it’s Sundar 👋

I’m a former Head of Marketing Science at Uber where I optimized $1Bn+ in spend across Brand, Performance, and Lifecycle. Now, I share weekly playbooks that help you prove and scale your Marketing ROI.

Thank you for being a valuable part of this growing community of 2.6K+ marketing leaders from Uber, DoorDash, Google, Spotify, and many more.

If you know a colleague or friend that might benefit from this newsletter, send them here.

Now let’s get experiMENTAL!

Connect your email, and you’ll instantly get a CRM with enriched customer insights and a platform that grows with your business.

With AI at the core, Attio lets you:

Join industry leaders like Granola, Taskrabbit, Flatfile and more.

👉 Try Attio Pro for free

First off, I’m back! After celebrating the New Years and then taking some time off in Tenerife, Spain I’m ready to get back to writing.

Second, I thought I’d start the year by talking about two of my favorite topics: food and data. I’m a huge foodie. My wife and I don’t splurge on clothes or other luxury items but when we have the opportunity to we will on fine dining. It’s been a passion of ours since the beginning of our relationship and my dream is to one day own a 3 star Michelin restaurant that features South Indian cuisine from where my parents are from in Tamil Nadu.

So, I thought I’d have fun by sharing what Data teams can learn from Michelin ⭐ chefs because there are some truly striking similarities.

I’m going to start with 2 obvious lessons and then 2 less obvious ones.

Trigger Warning: If you’re on a New Years January diet then this post might not be what you want to read. There will be images of mouth watering food throughout.

When I think of a Michelin ⭐ restaurants, the first thing I think of is quality of ingredients. These restaurants use the finest ingredients available focusing on sustainable, local, and seasonal meat, seafood, and vegetables.

But that’s not where the quality stops.

They then take those high quality ingredients and put them in the hands of the highest quality chefs using the highest quality tools and techniques to create perfection. Quality is everywhere.

Similarly, Data teams needs to take the same approach. We’ve all heard the phrase “garbage in, garbage out” BUT it’s not just about data. It also needs to be applied to talent, processes, communication.

Garbage data → garbage insights

Garbage talent → garbage reputation

Garbage processes → garbage efficiency

Garbage communication → garbage impact

Data teams that lack quality in any of these departments often fall short because that lack of quality comes back to bite them in the ass.

There’s a reason there’s the saying “You eat with your eyes first”.

Michelin Star restaurants have perfected this by transforming simple food into art. Even if you’re not into fancy food, you can’t deny the fact that good looking food is impressive. Some of the dishes I’ve eaten look like art pieces that I don’t want to dig into (for two seconds before my stomach overrides my eyes).

They can truly be breathtaking.

So the obvious thing Data teams can learn here is to learn how to let stakeholders “eat with your eyes first”. Poorly written emails or hard to read slides instantly turn off our minds. It doesn’t matter how brilliant your work is, but if no one reads it it really doesn’t matter.

Now there are multiple steps to get to Michelin level presentation, but start at basics like consistent sizing, fonts, colors in your decks. Make sure you’re using appropriate charts and minimizing words on your slide. Even consistent formatting in your decks will get you to that next level of presentation.

I’ll share an article soon on how to build better decks, but from my experience presentations are what separate the average data scientist from the elite ones.

Now, let’s move on to two less obvious themes.

From the words of the Michelin Guide itself “Consistency is key when awarding MICHELIN Stars, so we need to ensure that customers will receive the same high standard of cooking every time — for example, a Three Star restaurant will serve Three Star meals without fail. Various Inspectors will visit throughout the seasons, for lunch and dinner, both on weekends and during the week. Then we discuss their experiences as a team in order to make a final decision.”

For data teams, there is an obvious corollary in that you want the quality of the work to stay consistent over time, but there’s another type of consistency that is often overlooked.

Do you know what makes a 3 ⭐ mashed potato different than normal ones?

We already covered ingredients, so no it’s not that.

The answer is that after boiling the potatoes, you press them through a tool called a tamis which breaks down the potatoes into almost little fluffy rice like pieces. When mixed, instead of getting lumpy, you get this insanely smooth velvety texture that has an incredible taste and mouthfeel.

That’s it. That’s the secret. Consistency within the bite.

It’s also why Michelin ⭐ restaurants cut vegetables to the same size so they can look visually the same but also cook at the same temperature. Could you imagine if you were eating a carrot and one piece was crunchy and the other was mushy?! A travesty!

Similarly, data teams need to learn to be consistent across:

Imagine you’re working with an analyst and then their first email is great and you feel like everything is on track. Then their second one is confusing and the third is a mess. Consistency within the analysis from start to finish is key.

Data teams can do this by standardizing across the analysis and templatizing. This relieves a lot of headaches for stakeholders and builds confidence knowing that the work is going to be of a certain quality and that it will be consistent.

A few years ago, I ate at a 2 ⭐ Michelin restaurant called Spectrum and across the hundreds of dishes I’ve eaten, this one stood out and to this day it’s the most creative dish I’ve ever had:

These are combinations I could have never imagined putting together and the sheer brilliance of taste and texture is something I’ll never forget.

Now, what the f does this have to do with data teams? Well, the most frustrating answer your stakeholder can ever get is “Sorry, we can’t do that”.

Our jobs are to move the business forward with BETTER decisions. Not perfect. Better. Data teams must learn to get creative especially when there is imperfect data, limited resources, and tight deadlines. I’ve had this situation show up many times at Uber and the way I got out of it was by saying the following:

“We can’t do this BUT here’s what we propose”.

I almost always tried to communicate the viable alternative along with the pros, cons, risks, and uncertainties of my alternative and built a personal brand for being a “creative problem solver”. If you’ve ever heard that phrase, this is where the “creative” part comes from. The creative comes from putting various parts of the puzzle together that people wouldn’t expect you to (just like I was surprised by those ingredients at Spectrum).

Being that creative problem solver positions you and the team as empathetic partners who want to make it work instead of a hardline academic data scientist (which no one wants).

So, remember, if you want to become a better data scientist or build a better data team, think about these 4 things:

That’s it for this week!

Stay experiMENTAL,

Sundar