You cannot be good at everything at once.
Resources are limited, but improvements are not always zero-sum. Better tools or fewer handoffs can improve several things together. Tradeoffs must be examined, not assumed.
First principles
Six things true of work before any method.
These describe work, not this site. They would hold if you never filled in a canvas, named a pattern, or read the book. Each one forces a question, and the six questions are where every instrument here comes from.
DescribesWork, not the method
Test for the listWould it still matter if every artifact of Work Composition were thrown away?
Leads toSix questions, asked in order
Left outTwo claims, and why
I / The six principles
What we take to be true about work, and what each one makes us ask
1
Work serves a purpose.
Before deciding how to do something, ask what it needs to achieve, and whether it needs to happen at all. A task that is done well and was never needed is still waste, and it is the kind of waste that automation makes cheaper to keep.
The question it requires
What result is needed? What work can we remove?
Where this site asks it
2
Different work needs different capabilities.
Tasks that take the same time may need different knowledge, judgment, skills, or authority. Two ten-minute tasks can be nothing alike, and treating them alike is how the wrong task ends up with the wrong executor.
The question it requires
What does each remaining task require?
Where this site asks it
3
People, agents, and systems have different capabilities and limits.
What can do one task well may be unsuitable for another. The question is never whether a person or a machine is better, but which of them can meet this task's requirement, here, at the standard the result needs.
The question it requires
Who or what can meet those requirements?
Where this site asks it
4
How work fits together affects the result.
Dividing work changes the handoffs, dependencies, and checks needed to finish it. Every cut creates a seam, and the seam needs someone or something to carry the work across it.
The question it requires
How will the pieces come together?
Where this site asks it
5
Conditions affect what works.
An arrangement that succeeds in one setting may fail in another. What it needed in the first setting was often never written down, which is why the copy fails and the original does not.
The question it requires
What must be in place for this to work?
Where this site asks it
6
Capability must be demonstrated.
A claim that something can do the work needs evidence from the conditions in which it will be used. A demonstration under other conditions is a reason to test, not a reason to trust.
The question it requires
How will we know it works?
Where this site asks it
II / The six questions
One per principle, in the order a workflow is worked through
Purpose comes first because the cheapest task is the one you remove. Evidence comes last because it is the only question the other five cannot answer for themselves.
| Principle | Question it requires | Where it is asked |
|---|---|---|
| Purpose | What result is needed? What work can we remove? | Run it, first move |
| Different work | What does each remaining task require? | Fit, the rows |
| Different capabilities | Who or what can meet those requirements? | Fit, the columns |
| Fit | How will the pieces come together? | Canvas, cell 04 |
| Conditions | What must be in place for this to work? | Patterns, preconditions |
| Evidence | How will we know it works? | The gate |
III / Two claims left out
Tested against the list and not kept
Two claims that sound like principles were considered and dropped. Both may still be true. Neither would survive throwing every artifact away and starting again, which is the test a first principle has to pass.
The same shapes keep showing up.
That is a hypothesis to test through examples, and the patterns page is where it is being tested. Work Composition would still be useful if arrangements turn out to be more varied than expected.
The test
It is not enough that removing a principle would break some part of the method. A part of the method can depend on an arbitrary design choice, and that makes the choice load-bearing without making it true.
The stronger question is whether the principle would still matter if we threw away every artifact of Work Composition and started again. The six above pass. The two beside them do not.
Read next
The instruments the six questions became
The canvas asks all six of one workflow. The fit sheet asks the second and third of one cell. The gate asks the last before an agent ships. The thesis is the argument for writing the answers down as structures rather than beliefs.