What Leading a Data Team Actually Looks Like Right Now
There are certain problems that seem to never really go away if you’re leading a data team.
Don’t get me wrong. The tech world is currently saturated by AI hype.
Not only because every tool you look at now just feels like a thin layer of ChatGPT, but every conversation you have with the business or at a conference is in the shadow of AI. You can’t escape it.
But if you took a step back, I am sure you’d realize that the problems you are facing daily aren’t just about AI.
Despite CEOs and VPs wanting AI, they still need data teams that actually make sense of the data and help them use it to make better decisions.
In this article, I wanted to discuss some of the challenges that haven’t changed over the past few decades that data leaders still face.
Translating Business Needs Into Actual Outcomes
The tools you use continue to matter less than the outcomes you are able to drive. There will always be the next shiny set of tools, and sure, they can help.
But, like everything in life, every tool has its place.
In fact, despite the header, you’ll quickly realize that when leading a data team, it’s not just about taking the business needs and turning them into some sort of outcome. There is also a whole host of challenges that involve dealing with internal corporate politics, getting naysayers excited about adopting a new BI tool, that hopefully, this time…will actually be different.
All while having to get buy-in for your own projects, pushing back on ones you disagree with, and, well, actually deliver.
You need outcomes and wins your data team can stand behind.
That doesn’t just happen. Even when a project, in theory, has a good outcome, for example, you find a business lever that would drive massive change.
Great. Now you need to go prove it to the business, get them to fund a new initiative, fight against the five naysayers who don’t believe you or just don’t want you to surpass them, and then actually execute.
Easy right?
That still hasn’t changed in 2026. You need to take the data you do have access to, take the technical asks, and translate all of that into outcomes.
Fighting Tool Sprawl
Every tool I’ve worked with, especially the modern ones, only tends to further encourage sprawl. Not because the tools themselves are bad. But because the tools were built with the goal of making things easier as part of what they are.
They are meant to lower the barrier to doing technical work and driving output.
dbt, great, now we can focus less on all the stuff around building a data pipeline and merely translate business logic into SQL.
Tableau, awesome, drag-and-drop to your heart’s content. Build a dozen dashboards a month, and 144 a year. It’s easy.
I am sure now there is some level of markdown file sprawl going on out there as everyone is trying to build their Claude analysts or maybe LLM-generated micro tools. Thousands of them are made daily across a single company.
Dashboard sprawl, ontology sprawl, model sprawl, everything slowly becomes unwieldily.
The thing is, in all of these cases. Things often start out great. The first few POCs or even the first implementation work. The business gets excited and then opens the floodgates.
Suddenly, every team wants to build on top of the new tool. There is no form of governance, no tracking of whether what is being built is really needed. But wow, what productivity!
Who cares about governance right?
We will deal with that later.
Protecting Your Team From Burning Out
Data teams continue to be a catch-all team.
Automations, AI, Dashboards, Analytics, Machine Learning, Data Apps.
They can do it all, folks.
Anything that is somewhat technical but the IT or software team doesn’t handle is probably now being handled by the data team. Well, at least if the data leader didn’t push back and say no.
And now, it’s just going to be harder.
You’re going to have product teams and individuals with dashboards that look 80% of the way there, except they have none of the wiring or layers meant to ensure the data is accurate and the data pipelines are sustainable over a long period of time.
And these ideas can come to your team as fast as people can think.
And you, you have to be the one who pushes back and asks them why.
Why do I need this?
What is it for?
If you don’t, your data team will be overwhelmed and likely burn out.
Growing Your Team
Now more than ever, I imagine data leaders face the challenge of growing their team members. I mean, that’s along with actually trying to literally grow their team size.
How do you grow individuals when the demand from the business is productivity?
It’s not that you can’t both grow and be productive. But major growth often requires mistakes as well as a lack of knowledge and experience.
You need to go from not being capable to being capable.
Businesses hate that. They want their employees to turn a profit for them on day one. That’s rarely been how most professions operate. Even Facebook had its engineers spend the first 3-4 weeks in their own internal bootcamp to help accelerate the time to value.
Nevertheless, with AI, the expectation is to value faster. Build data pipelines yesterday and make sure your first PR is in on day one. Facebook did that too; they also made it clear on how to do it.
So many companies really just want to see their data teams being productive fast and care less about the personal growth. Truthfully, that isn’t really sustainable. We will want to grow talent, make them better and excited to do their job.
But I really would love to hear what y’all are thinking about the junior engineer problem.
Final Thoughts
I am always open to being wrong and time will tell which problems remain for data leaders and which won’t. Maybe, there won’t even be any data directors or VPs in the future.
Everyone will just be player coaches.
That’s what some people are betting on.
But in terms of what I am seeing at companies. Data leaders are still dealing with many of the same problems they have been for a while. Now they just have to also add in how their team is integrating AI.
With that, I want to say thanks for reading!
Also! Don’t forget to check the articles below.
ETLs vs ELTs: Why are ELTs Disrupting the Data Market? – Data Engineering Consulting
NetSuite to Snowflake Integration: Ultimate Guide to 2 Effective Methods
Bridging the Gap: A Data Leader’s Guide To Helping Your Data Team Create Next Level Analysis
The Data Engineer’s Guide to ETL Alternatives
Explaining Data Lakes, Data Lake Houses, Table Formats and Catalogs
How to cut exact scoring moments from Euro 2024 videos with SQL
How To Modernize Your Data Strategy And Infrastructure For 2025