Is This a Pig
or What?
By Randy Dufault
I was pondering a SOW the other day…
The Computing of Business
Now I suspect many of you, particularly if you reside
in the prime pork production region of the USA as I
do, may well be assuming (more on assumptions in a
bit) that I speak of a domesticated boar that has borne
piglets, but alas, I do not.
I speak instead of an ess oh doubleyew, the often-used
acronym for a Statement of Work.
To get the computing of our businesses done we
all need to use contractors in some fashion and the
contractual vehicle most often used to finalize a
relationship between said contractor and you, for
a given set of things to be done, is an SOW. The
contractor prepares one, ships it to you, you sign it and
the project is a screaming success. Simple, eh?
If only it were.
Have you ever read one of those things in detail? In my
time I’ve had the opportunity to see some really good
ones and some really not so good (e.g. bad) ones. I’ve
also had the opportunity, sitting on the contractor side
of a transaction as I most often do, to write some really
good ones and the occasional bad one.
An SOW is intended to describe a reasonably well-defined, fixed set of things to be done. If done properly
it should protect both the contractor and the contractee,
and give both parties confidence that what is desired
to be done, gets done in a controlled and satisfactory
fashion.
So what should you be looking for in that SOW?
The first thing is some sort of discussion on the part
of the vendor of what he or she believes you want
to accomplish. Sometimes such a section is titled as
a Statement of Requirements, or it may be a simple
introduction describing the business challenge/
problem. Regardless of what it is called, after reading
it make darn sure you are convinced the vendor fully
understands what you are trying to accomplish.
Typically next is a list of work tasks, or Statement of Tasks,
the vendor is proposing to execute, in order to solve the
described challenge.
If your requirement is for a complete new warehouse
management system and the vendor proposes a single
task reading: “create, test and deliver a warehouse management system,” you best start looking for a different
partner. Tasks need to be reasonably granular—not too
specific, yet not too broad. I’m not sure I can describe
what is a well-defined task, but I think you can typically
get a feel for the right and reasonable depth, in the
broader context of the whole project.
Included with a task should some sort of completion
criteria. Completion criteria should describe how
both you and the vendor know when the task is done.
A statement like “This task is complete when the
contractor departs the premise” likely is a bit high-level.
A task may or may not have estimates (dollars or hours)
associated with it. If you have asked for a fixed price
proposal, don’t expect the vendor to provide estimate
detail on each task—he doesn’t have to. On the flip side,
a time and material contract absolutely must include
detailed estimates.
Somewhere in the SOW there needs to be a set of assumptions (see I told you there would be more). In any
transaction there always are assumptions and any supposition falling outside the realm of normal common
sense and is not explicitly recorded, has the potential for
making a good business relationship go bad.
Much as with tasks, it is hard to say what is an
appropriate and complete set of assumptions. If the
list looks like a canned set the vendor throws in every
SOW, there may be an indication that not quite enough
thought was put in to what may or may not happen on
the project.
If the list is very detailed—maybe too detailed—it
may be an indication the vendor is not completely
comfortable with what the are being asked to do. A
little more discussion may well be in order.
Then there is the price section or Statement of Costs.
Of course there is going to be a number, maybe many
numbers. Does it all make sense? What about travel
and living expenses? Who is responsible for those?
What about taxes? Who pays those? Are there controls
and protections in the pricing both for you and for the
vendor? In the end, are you confident after reading
the SOW and understanding the pricing, that you know
what the project is going to cost you? If not, maybe it’s
time again for more discussion.
More negotiation and a revised SOW are always an
option. Don’t rush the process.
Contracting relationships are a fact of business life and
that means that we all need to deal with these SOW
things. Just make sure the next one you deal with truly
makes sense both to you and to your vendor.