Most businesses can tell you what their trucks, laptops and buildings are worth. Very few can tell you what their data is worth — or who is responsible for it. That gap is where the trouble starts.
These six principles are the ones we come back to on nearly every engagement. They are not academic. Each one maps to a failure we have watched play out in real companies running real spreadsheets and databases.
1. Data is an asset. Manage it like one.
It is easy to inventory physical assets because you can walk past them. Data is invisible until it fails, so it rarely gets the same discipline. Treat it as an asset with real value: know where it lives, know what it is worth to you, and know what it would cost to lose it. If the answer to "where is the master copy?" is a folder on someone's desktop, you have your first project.
2. Data must be accessible to the people who need it.
Accessible means the day-to-day team, not just the management team. A report that only one person can produce is a bottleneck wearing a disguise. When the answer to a routine question is "ask Sharon, she has the spreadsheet," the business is one holiday away from being unable to operate.
3. Data must be quick to find and quick to answer with.
Speed matters more than completeness for most operational questions. If pulling a straightforward number takes half a day, people stop asking — and then they start guessing. Slow data does not just waste time, it quietly trains everyone to work on instinct instead.
4. Data must be standardised across the company.
Two departments defining "active customer" differently will produce two different numbers, and both will be defensible. That is how meetings get spent arguing about whose figure is right instead of deciding anything. A data dictionary — a plain list of what each term means and how it is calculated — is unglamorous and settles most of it.
5. Available does not mean unprotected.
The same asset that has to reach the right people has to be kept from everyone else. Access should follow the role, not the individual, and it should be reviewed when people change jobs rather than only when they leave. Principle 1 applies here too: you protect assets in proportion to what they are worth.
6. Someone has to own it.
Governance sounds heavier than it is. In practice it means a named person is accountable for a given set of data, there is an agreed process for changing it, and that process gets revisited rather than written once and filed. Data with no owner degrades — not dramatically, just steadily, until nobody trusts it.
Where this usually breaks down
In our experience the failure is rarely a lack of intent. It is that the systems holding the data were never designed for the number of people now using them. A spreadsheet built for one person becomes a shared file with six editors. An Access database built for a single office ends up carrying a whole company. The principles above are all achievable — but not on foundations that were only ever meant to hold a fraction of the weight.
If this resonates and you would like help defining a data strategy — or fixing the system underneath it — we would be glad to talk.
Comments
Be the first to comment on this post.
