hand over the keys

When is a web site finished?

Handing over a web site to a client is always a satisfying feeling but is it always finished?

Of course you should always aim to complete everything on the client's brief and to their specifications but are there times when it is best to hand over the site when you still have some work to do?

Recently I have found more and more that I am handing over web sites that are perhaps only 90% complete or that have temporary solutions in place.

Why?

In the majority of cases it has been down to a conflict between my experience of web sites and the clients expectations. Anyone who has ever built a site for someone else will surely know what I mean.

The client wants a very complex feature coding that you know will either never get used due to its complexity or that they will "re-assess" in a few weeks time and realise you were correct all the time. Should  I do exactly what they say or supply a temporary solution that is easy to implement, easy to use and does 90% of what they initially required.

In one case the client "demanded" an RSVP system be added to their event listing. The simple solution was to add a mailto link but the client wanted all the RSVP to be stored in a database with a variety of reporting options. When I looked closely at the site, it's target user and intended purpose, I realised that very few people would RSVP as there was no reason for them to do so.

So I took the easy option and delivered a mailto link with the client having the option for me to create the full system if the RSVP system was used enough. Guess what, they never came back for the full system.

In another case I was struggling to make the site as beautiful in Internet Explorer 5, as requested in the brief, as it was in Firefox and Internet Explorer 6 & 7. Having studied the usage statistics of their previous site I could see that under 2% of their existing users would be "slightly" inconvenienced.

So again I took the easy option of delivering the site "as is" with the promise that if at any time in the future this became an issue for them I would try again to make the site compatible with an old and dead web browser.

Was I wrong? Is the client always right? Should I have spent the extra time delivering something that would not get used?

The web was meant to be read, not squished.
This isn't the way to test a responsive design.