How to Justify the Value of Data Teams

In a fantastic recent post, Tris J Burns argues that the litany of problems facing data teams stems from failure to clearly and convincingly convey their value. Here's the economics-based playbook I've used for years to solve this problem... As an economist trained in non-market #valuation, I took the approaches I used in environmental litigation cases and adapted them to produce solid $ estimates of my data team's contributions. The basic principles: 🔸 Data teams produce value in two main ways: 1) by increasing the positive outcomes assoc w/ with the org's objectives (e.g., higher sales, increased patient recovery rate, higher contract win rate, etc.); or 2) by decreasing the costs of achieving a given level of positive outcome. 🔸 The analysis is focused on two scenarios - the "But For" Case (what outcomes likely would have been achieved had the data team NOT done Initiative X) and the Actual Case (what outcomes were actually acheived in the presence of Initiative X). 🔸 Proactivity is key - once you implement X, the "But For" circumstances no longer exist, and the ability to gather the necessary data may be forever lost. As soon as you become aware that Initiative X MIGHT be implemented, begin gathering that baseline data. 🔸 The two people(s) you need to get to know well, and that your leader needs to facilitate your access to: 1) the person(s) responsible for budget execution and timekeeping; and 2) those responsible for reporting on the org's key performance metrics. If your focus is reduced costs, work with the budget execution folks to accurately characterize the "before" state of relevant costs (the ones Initiative X is likely to affect). Then measure that same basket of costs post-implementation, making sure to adjust for any factors unrelated to X that may have changed in the interim. Measuring the change in outcome measures attributable to Initiative X is more difficult, because we often have less direct control over these, and there are many more confounding factors to account for. Again though, the key is working with the performance measurement folks to get accurate pre- and post-Initiative X data. 🔸 Specific methods for estimating the delta vary (regression- and time series-based methods are common), and the final estimate will be build upon a set of assumptions with some uncertainty inherent (but so will any arguments against your numbers...). The target you want to hit is estimates that are logical, methodologically sound, easy to explain, and make reasonable assumptions. 🔸 Keep in mind that once people start nitpicking your assumptions and approach, you've already won - you've successfully advanced a substantial, positive value assoc with the data team's efforts - now you're just arguing over the magnitude of your win. Also, in your work with the budget exec and KPI teams, look for ways to improve their processes with your data skills - they're powerful friends to make and keep... #career

A lesson I learned early on is that no one cares how you get their data when the company’s stock price is down or the P&L is flashing red. And for many companies their P&L has been flashing red for a while and now even more so. If that is your company...no one’s asking if your data pipelines are DRY or if your warehouse follows best practices. Don't get me wrong, those things are important. But it's also crucial that we, as data teams, understand our stakeholders' goals. Because this would be a bad time to suggest: - A 50k investment into a new tool - A 9-month migration project - Hiring two new data engineers As Joe Reis put it, many data teams are being converted into wartime data teams right now. The economy is being squeezed, interest rates are high, and executives are going to look over their numbers far more closely. So if you're a data leader you need to make sure you: - Have clear wins and outcomes that tie to the business - Prioritize projects that improve decision-making or reduce costs - Speak the language of risk, ROI, and tradeoffs, not just pipelines and models Because when times are tough, the data team isn’t just a support function,  it’s a strategic asset if it can prove its value. Let me say that again, if it can prove its value. This isn’t the moment to show off technical sophistication. It’s the moment to show impact.

Most often, brilliant data teams spend 80% of their time on tasks the business doesn’t care about. And that’s exactly why they struggle to secure a seat at the strategic table. When I joined Voya, I saw this firsthand. One of my first moves was to host a full-day strategic offsite with my new team. A key part of the agenda? Inviting skip-level leaders to present their work. I wanted to see how they framed their contributions—especially in front of their new boss. Then, one presentation left me surprised. A leader from a highly technical team took the floor. His depth of knowledge was undeniable. He spoke fluently about Data Management, ETL, and complex systems. But every time he started on a use case, he followed the same pattern: "we got data from that component and delivered it to this component." This went on, again and again. Finally, I interrupted and asked,  "What did this help you achieve from a business perspective?" His response? "I don't know - we never talked to the business, just to the IT teams." And that’s exactly why the data team was seen as a cost center instead of a strategic partner. - The team had no visibility into how their work connected to business outcomes. - Worse, the business had no visibility into what the data team was delivering. To break that cycle, I made business literacy a core requirement for everyone on the team. One simple initiative: inviting leaders from across the business to lunch & learns – giving my team direct exposure to product lines, strategies, and customer challenges. And the impact was immediate. An unexpected bonus? This endeavor got the attention of business leaders who understood (many for the first time) how the data team was fueling their success. The key to a high-performance data team? Every single member must understand and articulate the business value they bring to the table. When people understand the “why,” their work becomes more meaningful—and their commitment to it deepens. What's your experience with building business literacy in technical teams?

Sometimes I hear people ask, “will [insert career] be replaced by AI?” And I think that’s probably the wrong paradigm. The more important question—whether we’re talking about data engineers, data analysts, or even CTOs and CDOs—is, “am I adding value that can’t be replicated by an AI?” And particularly for data teams, I think that answer is a resounding YES! Right now, AI is programmed to deliver specific, useful outputs based on carefully phrased prompts. And that can certainly be helpful. But that’s a tool, not a solution. The real work of data engineers involves a high degree of complex, abstract thought. This work — the reasoning, problem-solving, understanding how pieces fit together, and identifying how to drive business value through use cases — is where true creation happens. And it’s that abstract and creative problem solving that’s the true value of a data engineering team. AI might make you more efficient—but it won’t make you obsolete. Here are a few steps you can take today to ensure you’re delivering value that can’t be automated away: 1. Get closer to the business: One of AI’s greatest limitations is a lack of business understanding—which means you’ll need to uplevel your own. Build stakeholder relationships and understand exactly how and why data is used — or not — within your organization. The more you know about your stakeholders and their priorities, the better equipped you’ll be to deliver data products, processes, and infrastructure that meet those needs. 2. Measure and communicate your team’s ROI: Particularly as more routine tasks start to be automated by AI, leaders need to get comfortable measuring and communicating the big-picture value their teams deliver. 3. Prioritize data quality: AI is a data product—plain and simple. And like any data product, AI needs quality data to deliver value. Which means data engineers need to get really good at identifying and validating data for those models. Ultimately, talented data engineers only stand to benefit from GenAI. Greater efficiencies, less manual work, and more opportunities to drive value from data. Call me an optimist, but if I was placing bets, I would say the AI-powered future is bright for data engineering. Check out the full Medium article via link in the comments! #genai #dataengineering #dataengineeringjobs

💡 Struggling to Get Your Data Engineering Work Noticed? Here’s How to Change the Game 💡 In many organizations, data engineering can feel like thankless work. You’re the backbone of the data pipeline, yet your contributions often go unnoticed compared to more visible roles like data analysts/BEs. If you’ve been fighting for resources or recognition without success, you’re not alone. The truth is, non-technical leaders often don’t see the value of data engineering until something breaks. That’s why it’s crucial to align your work with clear business outcomes. Here’s a strategy that could help: 1. Understand Business Pain Points: Spend time talking to stakeholders. Understand their problems and where data could provide solutions. Offer quick wins that directly address these issues—this will help demonstrate the immediate value of your work. 2. Showcase Tangible Results: You need to tie your work to outcomes that leadership cares about. When your efforts lead to increased efficiency, cost savings, or better decision-making, make sure these wins are highlighted. 3. Consider Your Environment: If you find it nearly impossible to change the perception of data engineering in your current role, it might be time to explore opportunities where data is seen as a critical asset—where data teams work alongside product teams, or better yet, where data is the product. Remember, you’re not just managing data—you’re enabling the entire business to operate more efficiently and effectively. Your work is critical; it’s just a matter of getting others to see it that way. 💬 Question for You: How do you ensure your data engineering efforts are recognized within your organization? Share your strategies in the comments! #dataengineering #businessintelligence #dataanalytics #careeradvice #leadership