Roles
Who owns your data? Owner, steward and custodian, decided
Three roles, what each is on the hook for, and how to name them without inventing a committee.
7 minute read
Guides
Everything the person handed this job has to work out
Nobody arrives at this knowing it. The usual route is that a board asks a question, somebody senior says they will look into it, and a week later they are reading a 2,000-page reference work written for a bank. These are the same subjects, written for a business of 50 to 1,000 people that has to make a decision this quarter.
Free to read, no form in front of them. Each one says what to do, roughly what it costs in your own people's time, and where we think the received advice is wrong.
13 guides
Where to start
A data operating model for a business of 50 to 1,000 people
Why DAMA and the vendor pillar diagrams stall at this size, and a six-week sequence that does not.
9 minute read
Responsibilities
A RACI matrix for data, with a worked example
Fifteen activities, a filled-in grid for a 250-person business, and the two rows that always cause an argument.
8 minute read
Quality
The six data quality dimensions, and how to measure yours
Completeness, accuracy, timeliness, consistency, validity, uniqueness — and how to measure five of them with SQL.
8 minute read
Classification
Data classification: how many levels you actually need
Three levels beat five. Where special category data sits, and how to write criteria a tired person can apply.
7 minute read
Trust
Trust tiers and medallion architecture: bronze, silver and gold
Gold means refined, not correct. Two ladders that look like one, and why you need both.
7 minute read
Retention
Writing a data retention schedule you can defend
Where the periods come from, the backup problem nobody puts in writing, and a schedule you can draft in a day.
8 minute read
Compliance
What UK GDPR asks of your data: ROPA, DPIAs and DSARs
ROPA, DPIA, DSAR and the 72-hour clock — all of them resting on a register you may not have yet.
9 minute read
Lineage
Data lineage: what it is, and what it needs from you first
What the automatic version really covers, where the trail goes cold, and the cheap version that answers the same questions.
7 minute read
Trusted numbers
Why two reports give two numbers — and how to fix it
Four reasons two dashboards disagree, and the twenty-definition fortnight that fixes three of them.
7 minute read
Master data
Master data management without an MDM platform
Three decisions and a reference-list clean-up get most of the value. When a six-figure platform is genuinely right.
8 minute read
Measurement
The five levels of data maturity, and how to score yourself
The five levels in plain terms, an hour-long self-assessment, and why a score that cannot fall is worthless.
7 minute read
Platforms
Microsoft Purview and Fabric: what they cover, and what they leave to you
What the platform derives on its own, what it needs a human to supply, and the four decisions it cannot make.
8 minute read
Nothing here matches that.
Try a broader word, or look in the glossary — it covers many more terms in a sentence each.
Also worth reading
A written procedure, published in full
Every product in this category ships an empty template. This is one of the seven procedures a new customer inherits, exactly as it arrives, so you can judge the level of detail rather than imagine it.
Still deciding on a platform
Fabric or AWS?
An honest comparison with no recommendation, and the list of decisions that stay true whichever way you go — which is most of them.
The first step costs you nothing
Forty-five minutes with whoever runs your reporting
We tell you honestly whether this is worth doing at all, and roughly what it would take. If the answer is not yet, you will hear that. "Not for us" is a fine outcome, and a better one than a slow maybe.