Image by Ian-S via Flickr

I promised this week that I'd stop beating the dead horse of how many keywords to target per page, and I will keep that promise. But the divergence of opinion and the deeply-held beliefs that some of us have on this issue made me realize that I violated one of my own rules with this whole discussion. I gave you my opinion about something and then argued with people when their opinions differed. I didn't take the advice that I so routinely dispense: "Data beats opinions in any decision." Yes, that includes my opinion, too.

I like to say, "Opinions are like necks. Everybody has one." Unfortunately, we know from experience that opinions are not always equal. And, it's normal in business to gather the opinions of the smart people and then make a decision, so this feels normal to us.

But with Internet marketing, we can do better. Of course, there are lots of smart people with opinions that are often worth listening to, but in Internet marketing, we can usually test our theories. We can see what the data tells us and act accordingly.

So, to briefly revisit the keyword per page argument, maybe it's true that different situations require different approaches and everyone's opinion is right under the right circumstances. Or maybe it truly is better to take one of the approaches or the other. But none of that is as important as you knowing what the right approach is to take on your Web site.

And the only way to be really sure is to experiment. Why not try several different approaches in different parts of your Web site and see what happens. In this case, you'd want to see what rankings you get, yes, but also what sales you get. And you'd want to see what happens if you mix it up and do it another way. Over time, the data will inform you as to the best approach, and you can apply that in other parts of your site. Perhaps from time to time you'd want to rerun the experiment to make sure that your original results were still valid.

That's the real secret of these born-on-the-Web companies, such as Amazon and eBay. It's not that they had smarter opinions. It's that they tested every opinion with an experiment and they never stop, even to this day.

So remember that opinions are like necks. And every time you rely on an opinion instead of data, you are sticking your neck out.