The split testing that I mentioned previously revealed some interesting results:
A bit more split testing would be nice. We varied the ad headlines based on a few keywords. The next step would be to vary headlines, maybe using brand names, and to modify the ad description itself. I would like to save some of our free YSM money for other ideas so this will have to wait.
Using this information we crafted a new ad with the best performing headline and URL style. We bid on almost five times as many keywords to get more clicks per day. Some of those were worthless to us because the bid for a decent average position was so high. I ended up disabling some of them. We also created a mini-site using some of the split testing data. Unfortunately, our landing site does not convert well so the clicks and high CTR are going to waste. That is all part of the learning process though.
It was almost shocking how well those niche keywords worked. You will not get a massive number of clicks, and being overly reliant on two keywords is not a good thing, but I have a lot of confidence in the validity of this process now. This could really work well when creating a niche site (just do not pick something too low traffic like we did for Niche Site 1). I believe now more than ever that, with or without PPC, you could replicate the niche keyword to niche site to niche profit model many, many times.
That thought makes me happy.
Follow our adventures on Twitter