The golden triangle is oft quoted in, surprise surprise, given the usual cost over-runs and high prices usually paid, US military specifications.
I was looking at one recently and there was none of that "pick any two" business. Whichever two they got they wanted the third.
The problem with product specifications is getting enough dialogue between the client and the manufacturer to identify what is necessary and what would be "nice to have".
A good product has to hit several key targets:
[li] It has to meet a need[/li]
[li] It has to perform as required [/li]
[li] There has to be the manufacturing and materials technology available to make it[/li]
[li] It has to hit the market price. [/li]
This ideally would be a linear development where everything is established at the onset and every one then works toward meeting the agreed specification.
But the reality is usually one of an iterative development. Inevitably it turns out that it costs more to make than was originally thought and some compromises need to be made.
In the case of a new product there is a marketing expression that should be remembered:
I don't want it right, I want it tomorrow.
This is because the first one to market will often dominate the market but time to market costs a third of the lifetime profits for each year of delay.
Of course, once there is some product in the market that delivers something of what is needed we enter a "me too" phase where new manufacturers have to come to market with none of the concessions made for a first device. They have to bring some compelling technical or commercial advantage to stand a chance.
In their favour is that the existing manufacturer, having secured a market share lives in fear of "cross capture" and wants to capitalise on the original product without spending more money than they have to.
New products can be a real challenge.
The very best target specification I was ever presented with said simply: "4-20mA and we'll pay $3000 a unit."
They would not be drawn any further.
Why?
Because when clients draw up a specification they can make one of two mistakes, out of ignorance about what can or cannot be achieved:
1) they write a wish list which may prove extremely expensive to develop, cost to much per unit and may even prove unattainable and in which case they can't interest a manufacturer.
2) they write a cautious specification which may set standards far too low which is no good to either.
In my case It put the onus on us to really understand the market and the market objectives. The client wouldn't even be clear why they wanted a new technology.
As an outsider we had to not only find a winning specification but also look for unique features that would make life very difficult for the competition to come back with a "me too" design.
Of course, any time you let management into the game you have a problem. In my case management wouldn't let me include all the features I wanted either hardware or software and they didn't want this to cross the boundary from the target market into other markets where it would take sales from far more expensive product (which I solved by creating product "brands").
Once completed the client then wanted all the features I had wanted to include but management would let me include and finally got them.
The thing is that if you are embarked on a one off, life is very very difficult and you really do have to get everything as right as possible to begin with. If it is a new product with suitable numbers to be produced the "I want it tomorrow" approach is better and then covered by the "Continuous Development" Mission Statement which is designed to cover for continuously either fixing problems or adding (and sometimes removing) features.
JMW
![[wink]](/data/assets/smilies/wink.gif "[wink] [wink]")
