I can't draw a straight line even when using a straight edge as a guide so the only enjoyment I can derive from the sites I build are how they perform.
Not only in advanced functions and features that they provide visitors but also generally good to great rankings for the site owners.
Also, along with dburdon, the metrics and feedback you can get regarding your methods is much greater than "old school" meat world marketing and also much greater than server access logs alone can provide. It is like an entirely new layer of data one can use to work with.
What I like best about
SEO though are clients that prefer long term strategies over short term and likely only short term gains. Developing an
SEO strategy for them is usually the easiest because it allows me to concentrate on search engine basics to build performance and reliability over the long haul. For clients that want results last week, trying to track down and understand the latest fad some client heard about while not knowing long term effects can be nerve wracking. All to often explanations as to why something might not be the best route to take fall on deaf ears.
Oddly enough or maybe not so odd, I find sites I work on for clients interested more in the long term doing better both long term and short term than sites where clients insist on things being done their way.
Maybe the biggest reason
SEO interests me is that if one approaches it from the point of view of trying to understand what search engines' primary job is and how to best support them doing what they do, it often is not that hard.
Somewhat related, I find many of the things important for search engines are also important accessibility issues as well so doing one well often leads to doing the other well also without any extra work.