However, this is not easy: We're in a niche branch, and even my major kws hardly show more than 2 mio websites. The search-volume of most keywords is so low, that google trends in most cases shows no result.
So there is no "big semantic bubble" I might easily add to the keyword suggestion tool, instead I'd have to investigate and add hundreds if not thousands of niche terms from the long tail of my shop system (most of which would already perform well in the organics, anyways;).
For instance, as a first (bad) test today I added a relatively broad products-group-term and the estimator promised to send me at best 50-70 visitors per day, while the more refined two-word-combinations showed exactly nil. Since these 50-70 will presumably show a relatively bad conversion-rate, I'm now quite unsure what to do in order to NOT waste my time too much searching for money-making keywords.
On the other hand I found that an amazing amount of visitors in the past weeks already came with no referer at all (no, I do not mean the spiders: it was visitors, who ordered) or was searching for our company's name. So I am wondering whether time has come to begin branding, and not only look at clicks and conversion but view the ads-impressions themselves positively.
so:
1) Where's the 101-thread on keyword-research for adwords campaigns?
2) Did starting adword-campaigns influence your organic listings, positively or negatively (google swears it doesn't, but I'd like to hear other opinions;)?
2) Did starting adword-campaigns influence your organic listings, positively or negatively (google swears it doesn't, but I'd like to hear other opinions;)?
I have seen absolutely nothing in the organics, either positive or negative, that I could pin on AdWords campaigns. Most of my long-term AdWords clients rank very highly in the organics - but they've all been around a long time, and we've worked hard on that. Newer AdWords clients don't seem to change position unless we specifically work on the organics - the one exception being that if I work on one or more landing pages, sometimes that help a little. I certainly have never seen any deleterious effects on organics from AdWords. And that's over several hundred sites.
But nevertheless the first hours were a bit disappointing:
First, my account said that 10 cents per add were too low for the ads to show up in the serps. With such a broad term, I doubt I will have more than 1% conversion rate, so that would amount to 10 $$ per order advertising costs, which would be far too much (long term). But of course, I need more data first.
Second, contray to my expectations I cannot identify adwords-customers from the referer-entry as easily as I thought: Naive as I am, I thought it would go like adwords.google.com... just like images.google.com... Seems as pagead2.googlesyndication.com works, but not for the ads shown in the organics. As a workaround I might design particular adwords pages. I might add something like ..?adwords=1 to the pages and then internally parse the get-variable (+probably 301 to the old url), but I'm a bit afraid that would lead to a duplicate content filter or look like cloaking. Any ideas on this?
Maybe, but I prefer my own analysis. For instance, the selfmade tracking-analysis I mentioned in my initial post revealed huge differences to my hoster's stats. I don't like that.
All website-statistics I have yet tried left many many questions open. Normally, I have quite clear questions on my collected data, and to my experience statistics designed by someone else don't cover them. Google analytics may be a brilliant and useful program, but I doubt the google engineers will be able to provide the correlations between my own lingustic analysis of my website content and data on adwords-visitors, which I am planning to investigate.
For the time being, I implemented a workaround. But it is really a pity that the google ads from the organic serps, which are redirected anyways, dont't leave any traces I might exploit.
Oops, hope I wasn't too rude. Thanks for your tip, T_Media.
