To truly judge a dividend yield, you have to look at what you paid (or are going to pay) for the stock. Let's say you paid $20/share for Company A, and it's now at $100/share. Let's say their dividend yield is 2%, which may not sound that impressive, so you will get back $2/share annually. However, because you bought in at $20/share, you're really getting a 10% annual return ($2 being 10% of $20). Granted, this is an optimistic case, but it gets the point across.

Dividend yield is one part of the total yield on a stock. The other part is the capital growth yield (that is, the increase in the value of the stock). Don't confuse dividend yield and capital growth yield. You ought to judge dividend yield in comparison to the current stock price, without regard to what you paid for the stock. If a stock priced at $100 per share pays an annual $2 dividend (2%), it doesn't matter (from a purely financial perspective, tax considerations aside) whether you paid $20, $90, or $150 for the share. Your current carrying cost of the stock is $100. If you want, you can always sell your $100 share and invest it in a different stock with a higher yield. You should only consider the expected (i.e. future) capital growth, in combination with your expected dividend yield, in your consideration of whether you will have a good expected total yield. If you did pay (say, two years ago) $20 for your $100 share, then congratulations on your spectacular capital growth (~123% per anum!). If you bought the share 50 years ago, then something has gone very bad (~3% per anum). But these figures do not matter when considering the future yield; and they especially do not matter when considering how much money in dividends per $100 the stock will pay off per year. To truly determine whether you are getting a good dividend yield, you must consider it in relation to the expected capital growth yield. Some companies, instead of paying out their profits as dividends, invest the earnings in new projects (hopefully projects with a positive net present value -- i.e. hopefully not in loser projects). This reinvestment in the company increases the value of the stock & promises larger future dividends. You can compare dividend yields on companies that have the same reinvestment policy. You are getting a good dividend yield if it is higher than that of other similar companies. For example, if two companies with a stock price of $100 both pay out all of their cash flow (that is, there is no capital investment) & if in their industry the average dividend yield is 4%, but one stock pays a $4 (4%) dividend and the other pays a $6 (6%) dividend, you would much rather have the latter. Unfortunately, so would everyone else, so the latter stock would (if it were expected to continue to pay $6 dividends) likely rise in price to $150, reducing its dividend yield to 4%. If you were holding the $100 stock paying a 6% dividend, then you had a good dividend yield (because it was higher than the comparable stock paying a 4% dividend). However, now you are holding a $150 stock paying a 4% dividend & its dividend yield is no better than the $100 stock paying a 4% dividend. If you expect the $100 stock to pay a $5 (5%) dividend in the next period, then you should sell the $150 stock paying a 4% dividend (which would have paid a $6 dividend) and buy 1.5 shares of the $100 stock (which you expect will pay a total $7.50 dividend). If you are correct and the $100 stock does pay a $5 (5%) dividend (and it is expected to continue), then it will likely rise in price to $125, reducing the dividend yield to 4% again. Notice that if you compared the $150 stock dividend of $6 to its historic price (i.e. $100), then you would decide that it has a 6% dividend yield -- being higher than the 5% yield on the other stock -- this is clearly erroneous. Also note that if you can pick stocks that are paying a 5 or 6% dividend when comparable stocks are paying only 4% then you have a gift (maybe a lucky gift, but a gift). Remember that the above example is a comparison between companies that paid out all of their earnings as dividends. If either of the companies reinvested some of the earnings, then the comparison of dividend yield is no longer valid. You should take the reinvestment into account for the total yield of the stock.