A
few days ago, I wrote an essay about some of the issues involved in
making deadlines. Since that time, I have been remembering more and
more anecdotes, principles and parables about this worthy but misunderstood topic. In the following, I hope to discuss the interaction between schedule, deadlines, client
management and trust. None
of these totally original observations are, well, totally original. But hopefully some of them will seem reasonable or even insightful now and then.
In the following I will often use the terminology of visual effects. This is because the specific case study that I am thinking about (but usually not mentioned) often involved a "visual effects" like project, if not actually visual effects. I refer to "shots" where a shot is distinct quantifiable unit of film which is to be cut into the movie. Visual effects is bid and scheduled around this concept but other projects are more all-inclusive. A game is likely to be considered "just one thing" and not a bunch of separate but related elements, but even then one will still have discrete deliverable elements, approvals, deadlines for art direction, and so forth.
In
the following, “dWi” stands for deGraf/Wahrman, Inc, my old
production company.
Perhaps the most important of these ideas, and the one that was the hardest to learn, is the first, what I comically number "zero". It is as follows:
0. No R&D on a production project.
The early days of computer animation were rife with this sort of problem. The whole process was R&D so how could you not do R&D on a project? Well, we learned that one the hard way. It will be the subject of another essay. Some of the lessons learned are as follows.
Do your programming ahead of time. Do your tests. Work with your client to define a project based on those tests, not based on your belief of what it can be. People imagine different things even though the words are the same. Work with your clients to develop a look and then, using that look, bid and schedule a project. Nothing major in the realm of the unknown should be attempted on a production schedule. Ever. Period. No way. Not unless you are the client, and maybe not even then depending on how crazy a client you are.
Perhaps the most important of these ideas, and the one that was the hardest to learn, is the first, what I comically number "zero". It is as follows:
0. No R&D on a production project.
The early days of computer animation were rife with this sort of problem. The whole process was R&D so how could you not do R&D on a project? Well, we learned that one the hard way. It will be the subject of another essay. Some of the lessons learned are as follows.
Do your programming ahead of time. Do your tests. Work with your client to define a project based on those tests, not based on your belief of what it can be. People imagine different things even though the words are the same. Work with your clients to develop a look and then, using that look, bid and schedule a project. Nothing major in the realm of the unknown should be attempted on a production schedule. Ever. Period. No way. Not unless you are the client, and maybe not even then depending on how crazy a client you are.
1.
Divide and conquer.
What
Caesar said. Take a big project and turn it into bite sized pieces
and either smash them one at a time, or subdivide them among a team
if you have a team.
2.
Do some of the easy ones first.
There
is something to be said for being 10 percent done, which is much
better than 0 percent done. When you have delivered 20 out of 100
shots successfully then they know that their project is getting
attention and that you know what you are doing.
3.
Do not leave the worst for last.
By
the time you reach the end of the project, you and your people may be
tired. Therefore do not leave the most ambitious to the end, but
neither should you do the most ambitious first. Warm up, deliver
some shots, and then lean into a complicated shot.
4.
If possible, deliver an acceptable shot for every shot in the movie
before going back and improving things.
This
is debatable but I think it makes good sense. You want to be sure
that there is not a “scene missing” in the movie. You do not
want to keep your clients up all night with worry. Deliver something
for everything that must be in the movie and then, if there is time,
go back and make things better.
5.
A schedule without slack is not a good schedule.
Every
schedule needs slack and some of that slack may be visible to the
client and some may be your private reserve. But every project of
any note has had setbacks. People get sick, people quit, computers
break, clients change their mind but dont have any more time or money
to offer, whatever. You need some room to work with. Some of that
slack can be visible to the client, and some may not be. But there
must be slack.
6.
No one should have to kill themselves to make your schedule.
A
little extra work to make something better is never bad, but you dont
want people working 80 hour weeks. It is not healthy and it should
not be necessary. It is great when things are so much fun that
people want to work extra hours to make things better but it should
not be required of them just to get the project done. If it is, it
is a sign of a project out of control or improperly conceived or
scheduled.
7.
Whenever people have to work nights or weekends, the producer should
be there with them.
None
of this, see you in the morning, bye, shit. You make people work all
night, you work all night with us. And people should never have to
work all night alone. Oh, arent there other people on this project?
Cant they stay and help out? What the fuck?
8.
Some people thrive on last minute deadlines. Some people don't.
I
have a friend and client who loves to work all night and make a
deadline. I work like the devil to avoid such situations. I prefer
the “full court press”, basically to work obsessively and get
things done early if possible. If I am early, then I am not late.
9.
Some people seem to do some of their best work while procrastinating
on making a deadline.
This
is a variation of the “clean your apartment rather than study for
your exam” meme but in this case they are not cleaning their
apartment, they are working on an invention or a paper or anything
but what they should be working on. All you can do is gently remind
them of what they should be doing and let them figure out what they
should do. For example, I am writing this post instead of finishing
a biography and resume as part of a job application. Doctor, heal
thyself.
10.
Some people get special satisfaction out of causing chaos and then
saving the day.
These
people need to be executed as a lesson to the others.
11.
Make a deal with your clients
Client
cooperation requires client trust. Trust comes in part from their
confidence that you are doing your best for them given the realities
of schedule and the budget. One aspect of trust or lack thereof is
when clients feel they are being unfairly pushed into a corner and
not given an opportunity to request changes and give feedback.
One
way that I found to generate trust is a system developed at dWi on a
project that had a lot of work and some, but not much, time. The
system, which I doubt is original, it was just original with us, goes
like this: (a) every scene has a kickoff with the client, the
internal art director, the producer and the technical directors, (b)
half the time allocated for the scene is used to give our best guess
/ work in progress before too much client input, but at the half way
mark, you have a review, the client gives feedback, then (c) 1/4 of
the time allocated to the scene is used to implement these changes,
and one has another meeting and a second set of feedback, and then
(d) you go the final 1/4 time hopefully implementing those changes.
However, the project itself has some slack and that slack is the
clients to do with however they want, in good faith. So (e) if they
want to spend more time making that scene or element better with
their slack, we are happy to comply. Hopefully we are close enough
by (d) that this is not such a big deal. I emphasize that all this
is before we need to talk about overages or change of schedule. Of
course one can radically change the scope of work with a schedule
change and more money, but that is not what we are talking about
here.
12.
When you are screwing up, tell your clients as soon as possible.
Early
on in my experience with production, I did not understand this
principle, nor did any of my partners and colleagues. We hoped that
we could fix things enough that there would be no problems, but it
never quite worked out that way. So what I learned is that when you
are convinced you are going to be late, or you think the technique
chosen is not going to deliver what you thought it would, the best
thing you can do is to go to the clients as soon as possible and
explain this to them. They may hate you, they may pull the project,
they may slander you forever, but for some reason it is better that
way then waiting to the last minute when their money is spent and the
time is gone.
13.
Sometimes you have to give projects back.
Sometimes
you are awarded a project and discover, to your discomfort, that
there is something about the project that you really did not
understand. Yes, I know we are all supposed to be professionals, but
things dont always work out. In one case, a very traumatic one, it
turned out that from our point of view a client who had promised to
work with us to make an ambitious project meet their limited budget,
had another meaning to “work with us” than we had expected.
Their meaning was “we are going to fuck you as hard as we can and
make you pay and try to put you out of business and lie as hard as we
can. “ Under those circumstances, the best thing you can do, I
believe, is to apologize, take all the blame on yourself, and
regretfully return the project to them, any remaining money, and give
them any objects or work to date that you believe will be useful to
them. Its painful, but it is better than to suffer from their
charade. They know you want the project, but dont let them take
advantage of you. Apologize, and move on.
14. The first dinosaur may be 90 percent of the work.
This aphorism refers to the unfortunate situation whereby the work necessary to get the first dinosaur approved, its look, how it moves, and so forth, may be most of the work of getting many shots with that dinosaur done. In other words, you have to put a tremendous effort into getting the first shot, but then the other shots will be routine. This is very hard for some clients to understand or accept and even experienced clients may suffer from some anxiety because of this. But the phenomenon is real and it requires special scheduling, effort and management.
14. The first dinosaur may be 90 percent of the work.
This aphorism refers to the unfortunate situation whereby the work necessary to get the first dinosaur approved, its look, how it moves, and so forth, may be most of the work of getting many shots with that dinosaur done. In other words, you have to put a tremendous effort into getting the first shot, but then the other shots will be routine. This is very hard for some clients to understand or accept and even experienced clients may suffer from some anxiety because of this. But the phenomenon is real and it requires special scheduling, effort and management.
That
is enough for now.
No comments:
Post a Comment