Sustainability data comes from everywhere. Energy consumption from facilities, staff figures from HR, emissions from procurement, core financial figures from finance. Whoever brings that together in a report will sooner or later ask who owns which data point. That question is often only asked after something has already gone wrong: a figure that doesn't add up, a definition that gets filled in differently twice, a deadline no one felt was theirs.
Finance has a long history with ownership. Every line item in the annual accounts has a name behind it, a control process, a deadline fixed in a calendar. Sustainability usually doesn't have that history. The function is newer, the data comes from more corners, and the pressure to report often arrived faster than the time to build processes. The result: figures that are supplied by someone, but not owned by someone.
That difference is not trivial. Supplying means: I pass on what I have. Owning means: I stand behind its accuracy, I know where it comes from, and I am accountable if it's wrong. Without that second layer, sustainability data remains a collection of separate contributions rather than a controlled whole.
There is no fixed rule stating that finance or sustainability is the natural owner. It depends on the data point. A figure that already sits in the financial administration — energy costs, headcount, revenue per division — often sits closer to finance, simply because a control process already exists there. A figure that only sustainability collects — scope 3 categories, biodiversity indicators, social KPIs outside payroll administration — has no comparable home within finance and requires an owner within sustainability itself.
What doesn't work is assigning an entire reporting category to a department without looking at the underlying data points. Scope 1 and 2 sound like a single block, but consist of data points with different sources, different systems, and different people who have the first hand on the figure. Who owns the definition of a data point in a group with multiple entities is a different question from who owns the process behind it, and both are again different from the question who owns the control over that data point. These three roles can rest with the same person, but don't have to.
The first role is the definition: what exactly does this data point mean, which unit, which scope, which period. The second role is the process: who ensures the data comes out of the source system, on time, in the correct form. Who owns the process underneath in a group with multiple entities is often a different person from who established the definition — the first sits close to the source, the second knows the reporting framework. The third role is the control: who checks whether the figure is correct before it moves on.
Naming these three roles separately prevents a common problem: a data point that is managed a little by everyone and fully by no one. If the definition sits with sustainability, the process with a local entity, and the control isn't assigned anywhere, a gap arises that only becomes visible at the first deviation.
The most common situation is not that ownership is assigned incorrectly, but that it isn't assigned at all. A data point comes in through a spreadsheet someone set up three years ago, and no one has ever explicitly said: this is yours. As long as the figures are roughly correct, that goes unnoticed. At the first discrepancy — a number that suddenly deviates from last year, a check that raises a question — there is no one who can explain where the figure comes from or why it changed. What you do if no one is the owner in a group with multiple entities is therefore not just an organizational question, it is the question that determines whether a report holds up under scrutiny.
Dividing ownership only works if both sides speak the same language about what a data point actually is. That starts with something basic: what a valid value is for a given data point, and when a deviation from that value is a signal that someone needs to look at. Without that agreement, the discussion about ownership remains abstract, because no one knows precisely what the owner is held accountable for. With that agreement, ownership becomes concrete: you are responsible for this data point, within these boundaries, with these rules for when something deviates. How a signal deviation is set therefore belongs to the same decision as the question of who is the owner — one without the other results in ownership without substance.
Dividing ownership between finance and sustainability is at its core an exercise in precision: not assigning a category to one department, but establishing per data point who owns the definition, the process, and the control. That is not a one-off exercise but a register that must remain accurate as people change roles or processes change.
Once that ownership is clear, it also becomes visible how much of the underlying work — collecting, checking, and passing on data — is routine enough to automate. The work scan from FTE TO AI calculates per task which part of it can be taken over by AI, so that owners can focus on the data points that genuinely require judgment.
Vraag maar waar een datapunt vandaan komt. Dat is meestal de hele vraag.
Answers come from this site’s knowledge base. Not tailored advice, and not a scan of your company.