Does it make sense to add an exact match keyword for every single possible phrase, even if there are 5000+ of them and many/most only receive 1 click per month, etc?
The only "problem" is that you'd then have tons of exact matches that get very few impressions/clicks. For example, say you have a term that gets only 50 impressions per month. After 3 months you'd only have 150. Even if you had 50 clicks, would AdWords know how to "properly" adjust your bid based on your 33% CTR, with only 150 impressions?
I'm developing some custom tools, and using the AdWords API it would be relatively simple to programmatically analyze and build out your entire AdWords account.
For example, you could easily monitor every single click you receive, and automatically add an exact and phrase match for every term if you didn't already have them.
You would always have an identical group of broadmatches that didn't get many impressions, but rather would serve to help you build your db of new phrases your system didn't already know about.
Each day your system would generate a report showing all the clicks you received from terms that were broad matched. You would go through and categorize these as not relevent (add it as a negative) or relevent (add it as an exact and phrase match).
Eventually the goal would be for your broad matches to receive little or no impressions. You would have a massive negative keyword db and receive very few impressions for non relevent terms, and the CTRs for your exact and phrase matches would be as high as humanly possible.
Actually, once you achieved that you would get rid of the broad matches entirely and start all over using your phrase matches as you used the broad matches in the first system.
Eventually you would have only exact matches for every possible search phrase you could ever want to bid on. Is that the ultimate goal? Or is this all going too far?
The big problem I'm trying to combat are impressions and clicks that are not relevent. Unfortunately there are a lot of people searching for various widget information that is just not relevent to my widget site.
There are an almost unlimited number of possible terms that contain the words and phrases I want to bid on, and there's just no way to negative match them all - unless I literally built up a negative keyword list in the 5,000-10,000 range which would take some serious man hours.
It would be very well to have only exact matches! But is this possible? I mean how many keywords would you need? And how many ads? And will you be able to track all these keywords?
On the other hand, every impression would be very well targeted. You won't need negative KWs, your CTR will be higher, therefore your CPC will be smaller.
The question is: Are all these possible to be done? Is there anyone that has only exacty matches and covers all targeted searches?
It would be very well to have only exact matches! But is this possible?
I often do this, and by doing it, far outreach anything which other PPC management companies can do for people, because they're lazy or don't know any better (or don't care)
I mean how many keywords would you need? And how many ads? And will you be able to track all these keywords?
As many as you have to have, as many as you need and yes, you can track them all -- it's just a matter of doing the work.
On the other hand, every impression would be very well targeted. You won't need negative KWs, your CTR will be higher, therefore your CPC will be smaller.
Bingo. Remember to charge by the hour.
Is there anyone that has only exacty matches and covers all targeted searches?
Very few, that's why it works so well. :)
Oh, and if you're wondering why I'm telling you all this, it's because I know that very few people will be bothered to put the work in, in spite of knowing that it works. ;)
I'm actually starting to think that the best way is to focus primarily on negative keywords and work it from that angle. If I eventually had 5k or 10k negative keywords it would pretty much achieve the same thing, and it might be easier to do now that I think about it.
