Choosing a project with weighted scoring

Learning outcomes

Build a weighted scoring table from criteria, weights and ratings
Choose criteria that discriminate between the alternatives
Read the result correctly — including when it says the alternatives are equal
Name the ways a weighted scoring can be manipulated

The method

Nutzwertanalyse — weighted scoring — turns a decision between alternatives into something you can hand to somebody else and have them follow.

List the criteria that matter.
Give each a weight; the weights sum to 100.
Rate each alternative per criterion, on a fixed scale (say 0–10).
Multiply, sum, compare.

Criterion	Weight	Idea A	A·w	Idea B	B·w
Team can build it in one year	30	8	240	5	150
Somebody actually needs it	25	6	150	9	225
We learn something new	20	7	140	8	160
Data protection is manageable	15	9	135	4	60
Testable without a real client	10	8	80	6	60
Sum	100		745		655

The number is not the point. The visible reasoning is: anyone can now ask "why is feasibility weighted higher than novelty?" — and that is a question the team can answer, unlike "why did you pick A?".

Choosing criteria

A criterion is useful only if the alternatives differ on it. "Must be software" scores 10 for everything and contributes nothing but arithmetic.

Weak criterion	Better criterion
"Is interesting"	"At least three of us want to work on it in February"
"Modern technology"	"Uses at most one technology none of us has used before"
"Useful"	"There is a named person who will use it and can be asked questions"

Weak criterion

Better criterion

"Is interesting"

"At least three of us want to work on it in February"

"Modern technology"

"Uses at most one technology none of us has used before"

"Useful"

"There is a named person who will use it and can be asked questions"

Set the weights before you rate the alternatives. Otherwise the weights become the place where the preferred answer is engineered.

Reading the result

A clear gap — take the winner.
A gap of a few percent — the method says: these are equivalent. Then decide on something the table does not capture, and write down what that was.
A result that feels wrong — look for the criterion you left out. That feeling is usually information, not stubbornness.

Decisions

Every team documents its project choice as a weighted scoring table in the project proposal.
Criteria and weights are the team’s own. There is no prescribed catalogue — a borrowed catalogue produces borrowed decisions.
Weights are fixed before the ratings are entered.
If two alternatives end within a few percent of each other, the team writes one sentence saying what actually decided it.

Pitfalls

Tuning weights until the desired alternative wins. It happens, it is visible in the history, and it destroys the point of the exercise.
Criteria that do not discriminate. They add work and change nothing.
Too many criteria. Above roughly seven, each weight is so small that the sums are noise.
Treating the number as the decision. The table supports a decision; the team makes it.

Terminology

Deutsch	English
Nutzwertanalyse	weighted scoring / utility analysis
Kriterium	criterion
Gewichtung	weight
Bewertung	rating
nachvollziehbar	traceable, comprehensible

Deutsch

English

Nutzwertanalyse

weighted scoring / utility analysis

Kriterium

criterion

Gewichtung

weight

Bewertung

rating

nachvollziehbar

traceable, comprehensible