Technical analysis is one of the most controversial (and misunderstood) topics in all of finance. Like
@pingu pointed out, there is definitely some voodoo portions of TA (and that satire article is hilarious!), but if you believe in momentum/trends, then you also believe in TA.
At it's core, TA is the use of hard market data to make investment decisions. Historically, this was mainly price and volume data, but has expanded over the last 20 years or so (hedgers/speculators, put to call ratios, implied vol, etc..). There is an overlap now on certain topics/techniques that can be considered TA or quant. At a high-level, there are really two camps of TA:
1) Discretionary Chartists
This is the old school version of TA that includes chart patterns, support and resistance lines, elliott wave, etc. A lot of the voodoo portions of TA is in this space. This area is somewhat dying (as I think it should be), but some people still swear by it.
2) Systematic Implementers
This mainly deals with trend following, but could also include shorter-frequency and/or mean reversion strategies. Trend following has been around for a long time, and it is the second big camp of TA. A huge portion of CTAs / Managed Futures shops employ various varieties of trend following strategies. The only differences amongst them are; securities traded, time horizons, and risk levels. Trend following strategies are some of the best diversifies to an overall investment portfolio, actually. They essentially just capture multi-asset momentum risk premia.
Regarding the CMT designation, I think of it as a 3rd level designation. 1st level would be CFA. 2nd level would be FRM and CAIA. The CMT isn't going to get you a job (unless it is a niche TA job like
@financeguy mentioned), but it doesn't hurt to have a a secondary designation (if you want to spend the time). I don't think it has a lot of value by itself, unless a TA job is your sole goal. The good part about the program, IMO, is the curriculum is aligning itself more and more with the systematic trend following camp than the old-school chartists camp. Time series analysis, statistics, portfolio management, and behavioral finance are becoming larger and larger portions of the curriculum; all of which is a good thing if it is going to be more broadly accepted in the industry.
Those are my thoughts. As background, I work on the buy-side with multi-billion dollar pensions/endowments/foundations.