Home › How we work

Skills that aren't a technology

How we work

The skills every role uses, whatever the platform: how we think and decide, how we find out what people need, how we run projects, and the role that holds a project's technical shape together.

How we work

Thinking principles

How we decide: the habits behind every design on this site.

Questions first, then design

Every page here starts from the questions to chase, then the decisions; the Data systems pages show the pattern.

Critical thinking in data projects

Five habits for claims, numbers and designs: what exactly is claimed, what is the evidence, what else explains it, what would prove it wrong, what does it cost to be wrong.

Short ideas

Coming: More principles: reuse first and build only the gap, design end to end before building, one home per fact, and others.

How we work

Business analysis

Turning what people need into something a platform can promise.

Source system questionnaire

What to ask every source owner, with the reason for each question.

Know your consumers

The consumer register: who uses the data, what for, and what they need.

Where the answers go

Turning answers into contracts, settings and tests.

Short ideas

Coming: Requirements that can be tested against data, metric definitions with owners, and acceptance criteria.

How we work

Project management

Running data and AI projects so they finish, and finish right.

Discovery: fail fast for the whole project

The discovery document, the beliefs to test, reviews and sign-off.

Project review

Review a project's documents for gaps at its stage.

Top project management mistakes in data projects

Why data projects break normal project management, the mistakes we see most, early warning signs and a healthier plan shape.

Estimation mistakes in data projects

Twelve mistakes that make data estimates wrong, why each happens, and how to avoid it, with a worked estimate.

Where data engineering time goes

Most hours go to understanding source meaning, not code; the non-technical work and how to cut it.

Short ideas

Coming: Planning in tiers, estimating, risks and dependencies, status that tells the truth, and stakeholder updates.

How we work

The technical anchor

The person who holds a project's technical shape together, across roles.

The technical anchor

What the role does, and why projects without one drift.

Project review: the checks

The full list of checks, by stage.