There is an old sight gag that IT managers have pulled on unsuspecting users for years.  When users start piling new and changing requirements onto an application spec, the IT manager will write on the white board "fast, cheap, right" and say "Now pick any two."

What is interesting is that there may actually be a way to get all three.

When polled, non-IT management in companies are unhappy about application delivery.  Forrester reports that “… nearly one-third are dissatisfied with the time it takes their development shops to deliver custom applications, and the same proportion is disappointed by the quality of the apps that are ultimately delivered.” Much of this unhappiness arises from the inability of words to communicated needs and desires.  If you doubt this, try clearly explaining your personal needs and desires to your spouse.  Even among intimates it is not a simple process.

Thus when a user and an analysts initiate that long dance of requirements gathering and documentation, the seeds for disappointment are planted.  It is only after much code has been written that the user has an opportunity to object to the framework, business rule implementation, and style of the application.  Reworking often consumes as much time as the original design,
and by the time IT delivers the application in a barely useable form, the needs
may well have changed.

The hype curve is accelerating on the "prototyping" market.  Tools in
this space allow an analysts, or even a sober end user, to rapidly sketch out an
application’s look, feel, input, and output.  By working through
interactions and outcomes early in the development process, total development
time drops and end user satisfaction rises.

The prototyping market is relatively untapped, and the initial players are
gouging enterprises because they can.  This market is well to the left of
the chasm, and thus early adopters are willing to pay premiums to get the
benefits of rapid and accurate application delivery.  Firms like Axure,
iRise, Serena and Sofea flower this virgin meadow.  They all have varying
degrees of sophistication, ranging from "paint the look and feel" to fairly high
fidelity renderings of the final application.

One of our clients (who I am not at liberty to name) is looking a step
beyond, to making the prototype the actual application.  After all, if a
prototype is of high fidelity, and if you can easily map data into databases,
the added effort to deploy the finished application into production is small. 
If the prototype is web based and there are no client/server issues, then it is
easier still.

This market is too immature to make predictions, but I’ll hazard one anyway. 
With application development in perpetual backlog, and with the need for
business process innovation to drive new revenue and outmaneuver competition,
prototyping, RAD and web apps will merge.  With this combined power, agile
firms will gain market power over competitors by delivering new capabilities
internal and external much more quickly.  Apps will be done right the first
time, done more quickly, and done for less total cost – fast, cheap, and right.