A year-end data audit: five questions
The end of the year is a good time to check what state your company's data is in and what to fix before the new planning cycle begins. Five concrete questions for a leader.
December in most companies is the time for reviews and planning for the next year. Numbers are analysed, budgets agreed, KPIs revisited. It is exactly at this moment that it becomes especially visible when data cannot answer the questions leadership is asking.
I suggest using the end of the year not only for results analysis but also for a short audit of data as infrastructure. This does not require a large project - five questions are enough.
Question 1: where does the number we rely on actually come from?
Take any key metric used in management reporting. Ask: where does this number come from? Who calculates it? What methodology?
If different people in the team give different answers - that is a signal. Not a disaster, but a signal: the metric lives without a documented methodology, and different versions can diverge.
Question 2: what happened with data in 2020 that we did not record?
2020 was unusual. There were stoppages, process restructurings, changes in accounting. If those changes are not documented, historical data for this year may create false comparison baselines in the future.
A good question for the analyst: if someone analyses 2018-2023 trends in three years, is there anything about 2020 that would need explaining separately?
Question 3: who knows, and who does not?
Are there "knowledge holders" in your analytics - people without whom you cannot explain how a given metric is calculated? If so, that is a risk. The knowledge should be in documentation, not only in people's heads.
This does not mean "let those people go" - it means "ask them to write down what they know". Better to do this now while the context is fresh.
Question 4: what became permanent this year that started as temporary?
In 2020, under pressure, many decisions were made as temporary. Files, exports, manual processes. Some of them are still running - no longer as temporary but as permanent, without being formalised into a proper process.
Inventorying this "temporary-that-became-permanent" is a good starting point for work next year.
Question 5: what did we want to do with data in 2020 that we did not do - and why?
At the start of the year, tasks are usually formulated: build a dashboard, set up an integration, get proper accounting in place. Some of those did not get done. The question is not "why didn't you do it" in an accusatory sense, but "what blocked it" - resources, priorities, dependencies on other work.
The answer to this question helps plan the next year more realistically.
Why this is worth doing now
This is not an audit for its own sake. It is an opportunity to enter the new year with a clearer understanding of what your analytics is built on and where the risks are.
Five questions - half an hour of conversation with the right people. The answers will determine where to start in January.