Well, sometimes you do. It really depends on what your goals are. If you want to keep it simple and fast, then you probably don't (although maybe there might be a little bit of work involved if you want to link everything to your own production environment). But then again, maybe you're going after the major accounts. Each of them will have very specific demands for their web2print front site. Have you considered how they are going to be realized?
Scope
It all comes down to scope. Using one of the generic web2print clients (that come with the FLO Suite license), you won't have a lot of work getting started. And a lot of our customers stop right there. Happy to have a good, reliable, easy to use web2print solution.
But most of our customers want to go further. They want to use the FLO Suite to target large accounts. Rather than be satisfied with the small stationary orders, they use those orders to make the customer happy, and go after the larger jobs (catalogs, menus, point of sales materials, and so on).
The IT department
But that doesn't mean that all of them have their own developers or even project managers. About a third of our customers go all the way and do entire projects on their own, with a minimum of outside help in periods of load. Another third relies more heavily on outsorcing for their custom projects. While most of them do have project management capabilities (or good sales people that can do the task), they don't have inhouse developers. This is outsourced to one of our integrators. The outsourcing might consist of the entire project, or it might just entail the development, based on a comprehensive analysis of the project.
Focus
A lot of the resources needed to set up web2print solutions also depend on the focus the customer puts on it. If they create (or use) a single basic application and afterwards just make small changes to it for each account, they might still benefit from the flexibility that they need to convince the big end users. But it might save a lot of time in training and skill development.
But then again, a tight focus might loose you jobs. If the end customer wants Google Maps integrations, or e-books, or PURL campaigns, or even complete PIMS projects, then helping them out might mean more projects in the future. They are all valid choices, as long as you have them.