Prosjektledelse Prosjektledelse nr. 1 2020 | Page 37

DEBATT
Why?
1. Specific/Simple
We must not simplify for its own sake. You can
lose critical detail. It should be as simple as pos-
sible, but no simpler than what is needed for the
purpose at hand. If you have a gigantic complex
global multi decade project you might need a page
or 20 sub specifications to manage one of the top
ten objectives.
2. Measurable/Operationalized/Quantified
Yes, for all the variable, performance and costs.
Most important initial is quantification, so that
we have a clear common understanding of the
objective. The defined scale of measure needs to 4. Realistic/Relevant … Objectively attainable
reflect the stakeholder value. It must not be just Same comment as above. You cannot fix these
something that unimaginative people find they things as initial requirement specification. You
can quantify (the easy metrics).In my practice I specify what you want, and how cost-effective
also have a lot of sub-dimensions in each scale various levels would be. But you have to dynami-
(who, where, what dimensions) in order to model cally adjust things, as reality reveals itself step-
reality, and to help us decompose the problem to by-step. Agile as it ‘should be’.
the most critical areas. SMART ideas belong to big-bang waterfall, not
Agile Incremental value delivery.
3. Achievable/attainable
When setting value levels, it is extremely difficult 5. Timeable/Timely
to know what is really attainable. This is a matter The delivery time of specific numeric levels of
of latest state of the art technology which could value and qualities is of course important. The
be unknown, or unknown to the requirements earlier the better. For this reason I teach that we
specifier. It is dangerous to back off from levels of need to decompose the larger objective into smal-
performance you would like, but do not yet know ler earlier deliverables for the most critical peo-
how to reach, or more complex to reach together ple-situations-places [7]. This is the agile appro-
with all other objectives and resource constraints. ach. The ‘timely’ here (MW) smells of one specific
So what you have to do is ask for what will be cost date for everything at once. We need a curve of
effective if you get there, and hope some imagina- early frequent prioritised deliveries.
tive designers can find a way to do it.
In practice, with agility and incremental delivery
and feedback and adjustment, you will get infor-
mation about what is really possible as you move
along. No-one knows enough to determine such
things in advance of design, and in advance of
implementation and measurement.
“Thus, the Six Interdependencies
are a combination of already
known, previously independently
considered influences and now
additional consideration of damage and
cross-organizational resource consideration
together:
PROSJEKTLEDELSE • NR. 1 2020 37