Are you left at this point wanting a moral? The best decisions are reached when the best information is available. In order to have the best information possible, the right questions must be asked of the right people. In the case of development tools, the developers were "eating their own dog food," meaning that they were making the decisions, expending the effort, and reaping the benefits. This is the ideal. At the other end, an initiative like online bill payment had very little consumer interest prior to a couple of years ago. It continued unabated due to the sheer inertia created by that amount of money being at stake.
In the case of web services, I suppose there is the lesson that a constant danger exists in becoming too familiar with or too accustomed to what's around you. Familiarity may breed contempt, but it's more dangerous when it breeds complacency. Complacency is the enemy of innovation. Developers are always looking for ways to make work less like work, and companies are always keen on finding new ways of reducing costs, but the most revolutionary solutions are often an answer to limitations that the population at large is largely unaware of.
Finally, most drives to innovate arise from itches. If you find that some repetitive task is often linked with another repetitive task, see if there's a way to connect them. If lines of code in different modules are often called in succession, maybe the object model is at fault.
In short, in any effort to replace or combine existing items, you should do the following:
Create an overall strategy for implementing the change
Identify as many of the customers as possible, as well as how they will be affected, or even how they will perceive the effect
Bring thinkers of all stripes into the early stages to maintain perspective and invite creative discussion
Or sometimes do the opposite. Just never accept that something cannot be done, or that nothing is left to be done.