The world’s leading publication for data science, AI, and ML professionals.

Step-by-Step Guide: The Art of Winning Stakeholders as a Data Scientist to Drive Impact

From a Data Scientist at Spotifyβ€Š-β€ŠThe killer combo to turning your work into action

The First-Years Chronicles of a Data Scientist in Tech

A Data Scientist’s Guide to Turning Your Insights into Impactful Actions

The killer skill combo to turn your work into meaningful results

Imagine pouring your heart and soul into a project for months, only to see some or all of its content shelved. It’s a scenario many Data Scientists face.

That happened to me more than once. It left me feeling profoundly bitter wondering why that even happened in the first place.

This usually occurs when business goals change.

The project gets deprioritized, and so does your contribution. This is what happened to me in this case. There wasn’t much I could do about company bets shifting.

Other times, it’s because stakeholders do not follow up on your work.

Even when your message is relevant this can happen when stakeholders fail to grasp your insights or _aren’t convinced by them. C_hances are it’s because you probably messed up somewhere. The good news is, there are things you can do to prevent this from occurring.

This also happened to me, and I’m here to tell you exactly how I messed up and what skills I picked up along the line to turn my insights into actions.


The formula to become a Data Scientist never changes. It’s pretty straightforwardβ€”it all boils down to:

Maths + Code + Business Acumen + Soft Skills = Data Scientist Formula

Boom. Nothing crazy.

The only slightly crazy part is the soft skills. They come into play at all stages of the project scope, and even more so when you have to deliver your insights and make sure they are turned into actions. The formula is not straightforward though.

(Also, if you’re a junior data scientist, be sure to check out my article below on the top 5 essential lessons I learned at Spotify as a junior myself)

5 Essential Lessons for Junior Data Scientists I learned at Spotify (Part 1)


The not-so-soft Skills

I don’t get why we call them soft. They’re even harder to build than those technical skills. No textbook formula tells you exactly how.

If they weren’t bestowed on you by your fairy godmother πŸͺ„, you need to get to work on bestowing them on yourself.

Some of them you’ll learn when you get thrown out there in the wild and get your hands dirty. While others will rub off on you when you surround yourself with experts.

Soft skills are particularly important when you want to turn your insights into actions because you’ll have to convince decision-makers why they matter

But we’re not just talking about any soft skill. There is a particularly specific set of skills you need to develop to make sure your insights do not end up in limbo.

1. The Art of Translation✨

One day, I had to explain the analysis of a complex A/B test we conducted, which involved using statistical terms like "chi-test" and "p-value."

Technical jargon sounds like gibberish to most non-tech folks. So to make it easier for them to understand, I focused on whether the results were statistically significant or not. Usually, people know that it has something to do with the reliability of the results.

When you’ve gone through all the technical work involved in your project, all that’s left is to share the fruit of your labor with your stakeholders.

Keep in mind you’re speaking to non-tech people

The ones that will turn your insights or model into a product. These people have no idea what a Random Forest is. They just know it’s not a bunch of trees in Africa.

So learn how to translate complex information into plain language to make sure your work makes it to the next stage

It’s easy to assume that everyone will understand terms you deem to be simple. But just because they also work in or with tech, doesn’t necessarily mean they will, so make sure to speak English. Using plain language is the first step to getting your message across, or else it might end up being all for naught.


2. The Power of Persuasion 🫱🏼 β€πŸ«²πŸΎ

I recently spent months working on deep-dive research. The goal was to understand the quality of a part of the experience and suggest data-backed recommendations on how to improve it.

One key recommendation seemed outlandish to some, but my gut told me we were onto something. So, I reached out to a team working on a completely different aspect of this project. I dug into their past research and gathered any argument I could find that could support my cause.

I ultimately managed to convince my audience why my insight was worth pursuing. Had I not embraced it from the start, I would have never managed to have an impact through my research.

When you want your work to be turned into something real and impactful, persuasion becomes your new ally. It is not an easily learnt skill but it will help you stand out as an exceptional data scientist

Your audience is listening to you. True. But that doesn’t mean they’re convinced. Why should they care?

They also have to go the extra mile to convince you why it’s worth your time to research some stuff. So there is no reason why you also shouldn’t do the same. Everyone’s time is precious.

Here are 4 things you can do, and should do to make them care

  1. Not give up on your robust insights. No matter how far-fetched they sound to others. Trust your instinct and advocate for them, but only when you have confidence in their potential!!
  2. Present compelling arguments tailored to your specific audience. Convince them why your insights are worth pursuing by speaking their language so that it resonates with them. Make sure it’s data-backed. If you’re speaking to product managers then show them how your insights are relevant to the product.
  3. Build a strategic plan. You need to outline the key steps you propose to bring this vision into action.
  4. Conduct thorough research on teams or companies that have successfully implemented similar project. **** Emphasize the importance of not trailing behind and create a sense of urgency.
  5. Repeat, repeat, repeat. A fixed audience needs to hear something several times to integrate it. But also because you need to repeat the insights to different audiences. You never know where/who is the key stakeholder that will receive the information and unlock the next steps.

Be careful 🚨

In your quest for impact, remember that not everything can be artificially forced into being useful

If you find yourself having to convince others too much, it might indicate a misalignment with stakeholders’ expectations and hypotheses. But when your insights seamlessly align with their needs, and your story shines brightly within their context, there’s no need for persuasion – the impact speaks for itself.


3. The Magic of Storytelling 🀯

Remember that deep-dive research I mentioned earlier?

It caught the interest of the VP in my department, so my manager and I had to ensure the story flowed seamlessly and made complete sense before presenting it.

We spent time on this because every detail had to be data-backed – anything we said could influence impactful decisions.

When presenting your insights to decision-makers, storytelling skills makes the ultimate difference between whether or not they’ll end up in the leaderboard

After immersing yourself in a subject for weeks or months, it becomes challenging to distinguish what’s truly valuable to share from what’s not. Learning how to pick out important parts of the data and tie them back to suggestions for the product, and then persuade decision-makers to take action is a crucial skill.

Successful presentations are a result of both skillful delivery and strategic curation of relevant and compelling content

At this point, you know what you want to talk about to your stakeholders. But do you ** know _ho_w** to talk about it?

If you’re a junior, chances are you don’t. Even the most senior Data Scientists in Tech are still honing this skill.

These are the storytelling skills I acquired while preparing my presentation for the VP’s review

First and foremost, make sure that you understand the business goals, main strategic lines, focus areas of the company, and how your project aligns with them.

This will make you create a non-siloed story. Something connected with the living ecosystem of your company. Your research is not a standalone piece of work. It’s meant to contribute to a broader mission and vision.

Clearly articulate in your presentation how your work directly aligns with this overarching goal.

Slide 1 – Executive Summary

It gives an overview of your work and includes a summary of key insights and recommendations. It sets expectations for your audience.

This should typically include:

  1. Key insight, number, or idea that sums up the issue at hand
  2. Research objective + Methods used to reach this goal
  3. One chart illustrating the key statistic driving the research
  4. Bullet points of the key insights and recommendations


Slide 2 – Rationale + Disclaimers/Caveats

  1. Describe that key insight/number with more details and explain the end goal your research is trying to further
  2. Anything worth sharing that the stakeholder should know about to understand your research qualifies as a caveat. This can be a disclaimer on data quality issues for example


Slide 3 – Summary of your Key Insights

Support each key element with data.

πŸ”Š One Tip πŸ”Š : Add _a hyperli_nk next to each point leading to the slide where you develop the insight in-depth. This makes navigating your dec_k ea_sy.


Slide 4 – Summary of the Recommendations

Support each recommendation with the corresponding key number.

πŸ”Š One Tip πŸ”Š : Tag the relevant stakeholder(s). In this part, you want to explain how you suggest putting your recommendation into action to the person who is in a position to do it.


Body – Segment your Story into Chapters + Insert a Recap at the end of each one of them

Each section acts as a chapter to your story. Here is an example of how you can showcase each finding in your chapters.

  1. Make sure that each section prepares the ground for the next. Remember you’re telling a story here, so these need to flow seamlessly.
  2. Include a recap at the end of each chapter. This helps the information stick more in the mind of your audience.


Last Slide – Next Steps

List the actions that need to be taken in the future. An example could be: "Monitor the launch of an A/B test"

Finally, keep in mind that your story needs to speak for itself. This means that anyone should be able to pick up the insights at any time even without your intervention

Storytelling is a continuous journey for Data Scientists. With time, you’ll develop your own style.


Summary

True success as a Data Scientist lies not only in your technical prowess, but also in your ability to effectively communicate, influence, and inspire. You can do this by:

  1. Translating technical jargon to English. Simplifying complex concepts for non-tech people ensures they understand your insights and their value.
  2. Persuading your audience why your insights are worthwhile. Your ability to rally stakeholders to your cause can turn mere concepts into action and create a lasting impact.
  3. Continuously honing your storytelling skills. Success lies not only in the delivery of the presentation but also in the meticulous curation of its content.

Mastering these soft skills will propel your career to new heights and make your work truly impactful.


Coming soon: Unveiling the Ultimate Skill that sets apart Data Scientists who land the Best Jobs in Tech from the rest

  • A skill all Data Scientists in Tech have in common
  • A skill that can unlock the doors to your dream Data Science job
  • A skill so important it warrants its own story

So, stay tuned because you won’t want to miss out.

Okay, now I’m starting to sound like a clickbait.

See you next time πŸ‘‹πŸΌ


I have GIFTS for you 🎁 !

Sign up to my newsletter K’s DataLadder and you’ll automatically get my ultimate SQL cheat sheet with all the queries I use every day in my job in big tech + another secret gift!

I share each week what it’s like to be a Data Scientist in Tech, alongside practical tips, skills, and stories all meant to help you level up – because no one really knows until they’re in it!

If you haven’t done that already

  • Subscribe to my YouTube channel. New video coming up very soon!
  • Follow me ** on Instagram, LinkedIn, ** X, whatever works for you

See you soon!


Related Articles