Who reads your data point register when you're no longer there?

There is usually one person who knows where the CO2 figure comes from. Which spreadsheet, which tab, which assumption about last year's conversion factor. That person never wrote it down, because it was in her head and that worked. Until she went on holiday, changed roles, or left. Then it turns out that the data point register that did exist was only readable by its creator.

That is not an accident. It is the result of a register that was built as a memory aid rather than as a document. A cell with a number, a column with a source reference that only means something to someone who already knows the context. For the creator that suffices. For a successor, an auditor, or the CFO who has to defend the figure to the supervisory board, it is a puzzle.

What ownership actually means

Ownership of a data point is often confused with responsibility for supplying a number. That is not the same thing. Ownership means there is someone who can explain where the number comes from, which choices are embedded in it, and what changes if the source changes. That is a different role from filling in a field. It is the role of someone who can answer the question without having to go back to an email from two years ago.

In practice, that role is often not assigned, or assigned to someone who has since moved to a different position. The register lists a name that is no longer correct, or no name at all. Anyone who then asks a question about a data point does not get an answer but a search. That is the moment when it becomes clear whether ownership was ever truly established, or merely assumed.

What a readable register does record

A data point register that someone else can take over records three things that go beyond the number itself. First, the source: the system, the spreadsheet, or the colleague who supplies the figure, and not as a separate note but as part of the record itself. Second, the line from source to report: which operations the figure underwent before it appears in the report, so that a reviewer does not have to guess why the number in the report differs from the number in the source system. Third, the owner: not a name that was once filled in, but someone who can currently be called with a question about this specific data point.

That is not complicated to describe, but it is work to record. For each data point separately, and not as a one-off exercise but as something that is maintained as sources or owners change. The question where a data point already exists should precede this: a register only has value if it refers back to a source that truly exists, not to an assumption about where the figure once came from.

What this approach does not solve

A readable register makes handover possible. It does not automatically make figures correct. If the underlying source itself is unreliable — a meter that has been reading incorrectly for years, a spreadsheet with a formula error that no one noticed — then the register only records that error more precisely. That is a different question, one better answered by looking at how many of your data points have a source that is demonstrable and verifiable.

The register also does not resolve which data point is actually needed. Some organisations record dozens of data points for a topic for which a handful would suffice, and then spend maintenance effort on information no one consults. That consideration belongs with the question which data points you actually need, and precedes the question of who should be able to read the register. A register that is complete for the wrong topic is still the wrong register.

And the register does not resolve how much time it takes to get a topic in order. That depends on how many sources there are, how scattered they are, and how many of them already have an owner who can be reached. Anyone wanting an estimate of that will find at how long it takes to get a topic in order a description of what determines that duration, not a number that is the same for every organisation.

Why this usually starts with CO2

Building a data point register takes time, and that time is rarely freed up for all topics at once. Most organisations start with the topic where the data is already most scattered and the pressure to be able to give an explanation is greatest. Why that is often CO2 is described on the page about why CO2 is usually the first topic. Whoever starts there practises the way of recording on a topic well suited to it, before applying that way to the rest.

The scan is under construction

The Data Readiness Scan that records this — data point register, provenance per data point, ownership and quality rules — is still under development. There is currently no tool you can purchase; anyone with an interest in this can sign up for the waiting list and will be notified as soon as the scan becomes available.