I think there's way too much mythos about how "easy" a CMS is to use:
Myth 1- CMS is easy to update, all you have to do is install ABC module/plug-in/flavor. This is one of those things that are true in theory, but far from true in practice. Even though I use Drupal, which is considered more clean and sturdy than WP, it often happens that you can install a module and it does one of 4 things:
a) Breaks your entire site and you have to start over again, though this isn't really a common thing and shouldn't really happen assuming you are using Drush and other tools, but I can see this becoming a huge issue for someone with no experience.
b) The module installs, but it doesn't work because it is incompatible with the other modules you are using
c) the module simply does not work or hasn't been updated
d) the module needs 14 other modules to install with it as well, and this could of course loops back into a, b, c.
Myth 2 - CMS is easy to hand off to someone else or take as a hand-off from someone else.
I cry total foul on this one. Just because you know how to use the CMS doesn't mean everyone else does. I've looked at dozens of Magento sites, and even though Magento is famously secure out-of-the-box, I can tell you that about 75% of those I've seen fail at simple security vulnerability checks. If using the CMS was so easy, why is this an issue?
The issue is that in order for the assertion of easy-hand-off to be true, everyone has to be following best practices, but this most certainly does not happen. People "hack core" instead of using .info, .xml, or whatever files they should be using, thus creating even more spaghetti code, destroying the ability to safely upgrade, and yes, destroying the built-in security features of the CMS.
Myth 3 - Easy to learn
Definitely not the truth at all with Magento and Drupal, and I doubt that this is the case with WP as well. If WP was so easy to use and learn, then why do 90% of WP sites look like WP sites, and why would you buy a plug-in or theme instead of simply building your own? Why can't WP just be: hand-code some HTML5 and drop it into some file?
This is the part that stops me from considering CMS development over all other factors. The Definitive Guide to Drupal is 750 pages of small-print with a frustrating low amount of images. Of course, the opening pages admit that this guide is not definitive at all, but really something closer to a primer. Compare this to any RoR, Clojure, hell, even Java book with it's reams of code, and you find that you can do more much quicker with these languages than you can with a CMS, and none of these books come close to 700 pages.
My other issue is that the code is plain awful and tends to be bloated beyond belief. It disturbs me that 90% of the page-load with many CMS's is dangling code that doesn't do anything. Magento has a wopping 3,500,000 LOC, and each page load is how many lines of dangling JS on s site that doesn't use JS? And how do you remove all that dangling JS without hacking core?
I said in the other thread that my friend "develops" Magento sites, and she asked me if I could figure out a small issue. I had zero intention of helping her, but I was interested in seeing what the Magento base looks like:
file abc.php:
file abcd.php:
file abcde.php:
And this chain will go on for dozens of files.
Okay, there are quite a few benefits to CMS devlopment:
PHP is ubiquitous. I can go on all day with a customer about how I can develop a better/faster site in Python, etc, but how much is this going to cost in hosting per month? Why does someone want to pay $40/month for VPS hosting when they can get a shared host for $4/month with CMS pre-installed?
It is secure out of the box.
Generally easy-to-install modules and plug-ins, but I don't think that is all it's cracked up to be.
Built-in Admin interface. This is, in my opinion, the number one feature of a CMS.
I don't think that using a CMS is the golden key to fast development and customer satisfaction because it really isn't. If I was all about earning money, I'd surely spend the time to become an expert in Drupal or Magento development, but I'd rather spend the next 6 months doing something else.
If you have a customer who is semi-thinking show you a site map and a PSD or Vector drawing of what they want, what are you going to do? Three options seems to be most likely with most WP developers: Plug-in doesn't already exist so hack the core, tell the customer that it is too hard and they can't do that, or actually build the appropriate files to create the customized site and do it competently. I can guarantee you that the first too options are more likely. Would you want to be handed that one?