This was not the case even 5 years ago. The tools are now out there to make a reasonable-looking website with minimal training or experience. Sure, it looks very much like the other sites made with that tool, but (apparently) most businesses don't care.
Custom web dev will continue to exist only at the low and high ends -- hobbyist websites made by people who enjoy the process, and a few large, complex specialist sites.
People were saying exactly the same thing in relation to FrontPage 20 years ago. After that it was Dreamweaver. WordPress. Wix. etc.
The tools for people to go off and do it themselves have been with us for a very long time, and they've only ever squeezed out the very bottom of the market. The types of organisations that DIY appeals to aren't in the market and don't have the budget for an average developer, and the organisations that spend money on developers wouldn't consider DIY.
The whole industry is moving forward. The bottom of the freelance market 20 years ago was plain static pages and that's what FrontPage competed with. The bottom of the freelance market now is a content-managed dynamic site and that's what Facebook pages compete with.
I don't see this. Take a look at the Wix testimonials, for instance: http://www.wix.com/stories/ – they are individuals and tiny businesses, not the mid-market.
Previously you contracted a freelancer to put together your website piece by piece, now, thanks to improvements to the tools available you contract a freelancer to customise a cms for you. You then hire a marketer to manage it going forward.
The work is still there, it's just the freelance work has dried up because you no longer need them to put copy on the site.
There's still a disconnect on the part of some small business owners...they think that their website is important. But a lot of them are realizing that nearly all of their inbound leads are coming from other sources and that the website is just an expense that isn't really necessary.
After a small investment on his part in advertising, we were generating many more leads than he could handle at a very good ROI--seemingly because the nice site was doing a good job differentiating him. So at least in his case, having a decent website did bring a real competitive advantage.
I got to see the numbers of a mid-sized web design agency, and they were really an ad agency, but they didn't realize it (and didn't market accordingly).
The web is substantially more complex than it was even a decade ago. There are way more apps, platforms, and experiences to develop for—more and more companies have their very existence tied directly to their digital technology investments.
For those companies, there is no reason to hire a "web designer" (understood as someone who knows design and a bit of programming). They hire UI/UX designers and developers.
On the flip side, for businesses where technology is not fundamental to the competitive advantage the tools have become commoditized. A decade ago, they might have hired a "web designer" to build out a basic WordPress site. Now they can just use Squarespace.
If you can actually do web development, there's more work than there has ever been. Software has just eaten the middle out.
[1] No offense. I've made plenty of money doing web design work in the past.
We use freelancers a lot to supliment our inhouse skillset which I think is the industry standard
In other words, the technology cycle is coming around again. We went through the same thing in the early 2000s, as it became nearly impossible to get a non-maintenance job as a desktop software developer and the web freelance market took off (just on the heels of the dot-com bubble bursting). Also, just as the shift in web-development from well-funded dot-coms to individual freelancers and small startups led to the "Web 2.0" renaissance, we might see a "Mobile 2.0" renaissance with a new generation of technologies (Swift, MBaaS, Dagger, RxJava/RxCocoa) that changes how we do mobile development.
(I'm not sure we'll see a "Web 3.0" renaissance - "Desktop 1.0" was the PC, CP/M, MS-DOS, Apple Basic, etc., and "Desktop 2.0" was Macintosh and Win32 development, but what was supposed to be "Desktop 3.0" - OS X, desktop Linux, and .NET on the client - remained fairly niche specialties. I wonder if React/Polymer will go the same way.)
That is the definition of "anecdotal."
The second point backing this up was the reduced number of conferences. Assuming there were more people who have work then there should be more conferences - even if there were more people out of work.