There's a Gartner paper titled "How to Determine the Right Team Size in Data and Analytics". I did some simple maths on its ranges and wanted to check with data leaders out there: how useful are the numbers without knowing what the team has to deliver?
The ranges made me pause over what a fifteen-person data team would mean for a business with $50 million in annual revenue. That is a substantial commitment, but revenue alone cannot tell us whether it is the right one.
One acquaintance described an Australian retailer with nearly $1 billion (AUD) in revenue and twenty people in the data team. It was an interesting comparison, not a like-for-like benchmark. What counts as the data team? How much support comes from partners or people embedded elsewhere? What service levels, complexity and risks does the team carry?
The original post carried the working as an image. The paper itself is licensed, so no copy here. Its public abstract makes a useful distinction: ratios and benchmarks can be a starting point, but building a forecast is a better approach.
Postscript, September 2026, revised November 2026. Gartner's 2026 data and analytics predictions suggest that leaders can learn from AI-first start-ups with "fewer employees with significant ownership", including adaptable engineers and generalists. That gives me another reason to revisit how work is organised. It does not validate either staffing ratio above. The right size still depends on the work, the support available and the outcomes expected.
First posted on LinkedIn. Revised for this site in November 2026.