You're right about the flaw being lower diversification and with it you are taking an active bet on a small group of companies which may perform better over time and just as easily may perform worse.
The main problem is this. Dividends are not some sort of free money. They come out of the total return, so when you say you may need less money to get the same income, that is a fundamental misunderstanding that will really screw yourself over if you don't come to understand it.
So firstly, "dividend investing" is helpful only psychologically, and may be harmful by choosing things with higher dividends to "feel better" and could come with lower total return.
Secondly "dividend GROWTH investing" at least is not quite so bad as "dividend investing", the idea is to choose quality companies and in theory is a good idea if the dividends grow enough so that you can ignore the total return in the future and not freak out in a downturn and thereby let you invest with a higher AA in equities, but whether this theory is based on reality is another thing. And on top of that, again, the dividends are not some sort of magical free money, the more dividends you get, the lower your growth will be. Dividends come out of total return.
I believe during the GFC many dividend growth funds also cut their dividends, so again you will still need bonds or CDs. My concern is what would happen during a prolonged bear market. My gut says that this smaller bond allocation will get used up with less return for an extended period and the "good feeling" you get by assuming you don't need to look at the return and are somehow safe with your dividends will turn out to be very dangerous. I have not back tested this though.
In general, the dividend growth investing idea is very alluring to me, but I have a real worry relying on it. Having said that, There are some closed-end funds especially in the UK and Australia with 50+ years where dividends have increased every year (or almost every year) and they keep some cash for bear markets and I think investing in those seems reasonable, but I would certainly not invest all my money with them due to the sake of diversification. Maybe whatever I was going to invest in those markets I would invest in them, and whatever I was going to invest in the rest of the world I would stick to an index fund. I think it is easier to outperform and cut out the shitty companies in smaller markets so I would worry about someone using this in the US for example.
In response to "what is your strategy", my philosophy drives my investing, and my philosophy is 'I don't know more than the market and neither does anyone else", and as such I want to be 100% world diversified in my equities with no tilting to anything. Every time someone comes along with something interesting like min volatility, dividend growth investing, moving averages, REIT's, my question always comes back and the answer is always the same - I and the massive majority of people have no fucking idea even when their arrogance sounds like they do, and I have no idea how to find the last 1-2% that actually can find outperforming companies, so the answer to that is 100% cap weighted equities and no more guessing is needed. Draw down as needed in retirement any excess that is not met by dividends by selling off some stocks. It's not a great strategy but it excludes the bullshit and guessing which is basically every other option.