In most of my posts on dividend investing I tend to preach about a strict method for equity selection that focuses on valuation, dividend yield, dividend sustainability and a minimum number of annual dividend increases. I also typically mention the same stocks in most articles, which exhibit the type of dividend stock I am looking for.

Yum! Brands (NYSE: YUM): The reason why I purchased this stock is because it has a strong presence in China. In addition it has a strong opportunity for growth in the world’s most populous nation, beating rival McDonald’s (NYSE: MCD).

The common characteristic behind all of these stocks is that they were purchased in a clear violation to my entry criteria. The most notable exception was that they were yielding below my minimum threshold of 2.50% at the time of purchase. I only initiated a small position in each of the three stocks mentioned above, mostly because of their low yields.

The main limitation behind this sample of course is that I selected three growth stocks during a period where the market has gone up by 100% since its March 2009 lows. Bull markets typically tend to favor growth stocks, as investors bid their prices up, while defensive sectors such as consumer staples might not rise as much. It is important to note however that while the stocks mentioned above were purchased at yields which were lower than my minimum threshold of 2.50%, their valuations were reasonable based off P/E ratios below 20 and double digit growth rates in EPS.

The point I am trying to make is that one should not get married to their positions. It is important to have a strategy, but it is also important that not all strategies will work all the time. While I do believe that Johnson & Johnson (JNJ), despite its recent difficulties, is a great dividend growth stock to own, it should be one of at least 30 in a diversified dividend portfolio. Diversification is important, not only to reduce the risk of concentrated bets on a sector that implodes but also to provide investors with a greater chance of including sectors and stocks which could provide decent total returns over time. I am striving to have a dividend portfolio which is comprised of three types of dividend stocks. So far the higher yielding and the lowest yielding stocks have delivered the best total returns, while stocks in the sweet spot have done only OK. It is also important to keep a long-term view on your investments as well, as your investing portfolio could go through several market cycles before and after you retire on it.

Another important thing to remember is that dividend investing is not a black box style investing method. Sometimes a company with great quality characteristics which is just shy from your entry criteria might make more sense, despite the fact that your data analysis says otherwise. While it is great to have predetermined criteria before purchasing a stock, sometimes it might be wise to break some of your rules in order to enter into the right position.

At the end of the day, pure dividend growth investing is mostly about identifying companies with rising earnings and dividends, and then riding this trend for as long as possible, despite current yields. Adding additional filters could protect investors at times of irrational exuberance, but could also limit their upside during economic rebounds. As a result, investing should be viewed not just in a quantitative but also in a qualitative way. Evaluating company qualities on a stock by stock basis, while highly subjective, could be the decisive factor behind your investment success.