In the fourth week of The Adwords Experiment things really started ‘moving’. But it took that long and a lot of trial and error to really learn the system. Which is actually much longer than originally thought. On the outset it seemed like a week to learn and then three weeks to collect data. But in the end… Not so much.
Week 4 recap:
So eight clicks over a month. Not exactly what we were looking for when we started the project. And from the amount of money needed to spend it is pretty clear that we cannot get 10,000 hits out of $100… Clicks would have to sell for the one penny starting bid. But here’s what we did get… 10,000+ impressions. The name of my blog was seen over ten thousand times. While there is no tangible benefit to that, maybe they will think more favorably of me in the future. Who knows?
First Hypothesis: Dis proven!
There is no magic number that can bring you enough clicks to pay for your Adwords bid. You cannot game Google.
Second Hypothesis: Dis proven!
There is no magic number that will bring you a profit on your clicks. And that makes sense. If you could they would run out of money and wouldn’t be able to do things like buy Motorola.
That does not mean that using Google Adwords to ‘sell your product’ is a bad idea. This conclusion just means that it’s not a money machine. Using Google Adwords to sell your product is a good idea… As long as you have learned how to do it at the least possible cost. Being efficient is the way a company makes money with the revenue it generates.
On the plus side the whole experiment cost .33 cents. That’s a great value for the learning experience. Hopefully from your standpoint, these blogs are worth way more. The cost to me personally was only $.33, but a company would be facing opportunity cost, plus incurred expenses during such an experiment. Paying someone to do such an experiment, or learn this ‘new technology’ would cost a company much more than it does me. A big thanks to Google for financing this whole experiment!
Now the thing to decide, is do I keep the campaign running and go for more hits? Or do I save the money in the account until another project needs it? What do you think? What I will definitely be doing is writing up a guide based on all the little tricks discovered during this process. Its going to be short, sweet, and to the point. The goal of the document will be to save you the learning time I had to spend to get the Adwords Campaign up and running. This document will cost you… Nothing. That’s right, I’m going to give it away. It will be short, more of a checklist and therefore quick to write. So I won’t charge you for it. More information about that soon.
The Adwords Experiment is up to week 3! Actually going through this process has taught me a lot about using Adwords. If you are going to run an adwords campaign that uses many low competition keywords, you have to create an ad group for each keyword. Ads must have the keyword in the text of the ad. Learning how to do this is like learning how to walk. Clever bidding is like learning how to run. The first two weeks of this experiment was definitely running before walking.
At the end of the third week we see stats that are pretty close to the first two. Something is not right, so more drastic changes need to be made. In week 2 we started seeing Adgroups increasing the effectiveness of each ad… But at the end of Week 3 going into Week 4 its time for a complete overhaul.
During weeks 1 and 2 more Ads in my campaign were created and polished enough to become ‘eligible’ for inclusion on the results for the keywords in the campaign. But not many people were seeing them. And no one was clicking them. The campaign now contains 23 keywords and 19 adgroups which support those keywords. Ads no longer double up on the keywords, which seemed fine at the beginning of the process. There are multiple copies of ads in each adgroup if multiple keywords happened to be used in the same ad.
Going forward each Ad Group will have one Keyword. Even if it has multiple ads. Operating under the assumption that whatever algorithm Adwords uses is smart enough to use multiple keywords with multiple ads in a single Adgroup is apparently not the case. So some major time will be spent resorting the Adwords Campaign, and rewriting the rough draft of the guide. Getting that guide is going to save you so much time. It will even be better then reading about all these mistakes and corrects as I make them.
Till next week! Keep Experimenting… Share your results if you have any! Oh and keep an eye out for my ads.
This week on the Adwords Experiment the bid on each keyword has been kicked up to 2 cents per click. There was some additional tweeking which many help the ads come up in keyword searches. These tweek’s were necessary and should have been done before the experiment started.
Remember before you start an Adwords campaign, make sure that your ads actually contain the keywords they target. Google uses relevancy and bid amount to determine if your ads come up in searches. Apparently they wont let just anyone bid on any keyword with an unrelated ad. To make sure your ad actually comes up, polish the ads to be meaningful and contain the keyswords relevant to your site. For this reason the ‘shot gun’ approach is most likely not a successful one.
Fortunately Google lets you have multiple ads per campaign. Which gives you an opportunity to create ads which target those keywords. This has been a classic example of learning how to run before learning how to walk. The experiment may be in trouble simply because time hasn’t been spent perfecting the ads. But on the bright side it won’t cost anything to keep going… So the experiment can be rerun at a later date.
For Week 3 the keywords and ads are narrowing down into ‘Ad Groups’. There will be six ad groups each with between 1 and 3 ads. They will end up having 3-4 keywords each. You can save yourself a lot of time and effort by setting this up ahead of time. For me this is a fun experiment, so the trial and error approach is fine. But if you are doing this for real money, at a real company, make the Ad Groups ahead of time. You know you are doing it right when Google starts suggesting additional keywords to go with your ads. For me that took till the first day of week 3. If you group well it should happen almost immediately.
There isn’t much reason to post pictures of the graphs… They look exactly the same. Same number of clicks, same cost. But the impressions are rising. So people have been seeing the ads.
This week was pretty much the same as the first week. Which is disappointing, although informative. It means that the amount that you bid is not the only relevent factor in using Adwords. I’m preparing something really special for you guys based on what I’m learning. Anyone who wants to run a good Adwords Campaign is going to want to take a look. Check back with Week 3.