I'm interested in creating a keyword that is not entirely a broad match and doesn't exactly fit the dynamic title function either (as far as I can tell).
I want to make keyword with a "wildcard" i.e. "wildcard word or phrase, big widgets" where the wildcard word or phrase can be absolutely anything BUT "big widgets" must be typed in. Just typing in "big" or "widgets" will not cut it. For that matter, I don't want "big widgets" typed in without the wildcard (too expensive for the term I am interested in). I want it to be "any word or phrase" + "big widgets".
Can this be done?
There aren't any current wildcard functions in AdWords.
It sounds like you already understand the 3 matching options, and that's all there is to work with at present.
I even typed in "asdflkjsdfa; big widgets" and he/she showed up. Am I missing something?
This would phrase match everthing that includes the phrase "big widgets" including the term: big widgets - which is the one term you didn't want to show up for.
Sadly this is not possible. OUr company would like this exact thing as well. ( I hope you are not our competition!)
The only way we could achieve this is to input many many iterations as we could think of.
key phrase
-[key phrase]
would produce the result you are after? Must go and check this out again.
red widgets
-[red widgets]
will show for:
big red widgets
widgets red
but not for:
red widgets
I do not believe you can negate an Exact match.
We had turned the test off but put it back on again after your first post about it and have been running it for a few hours. Some interesting results:
When we tried it this morning it worked absolutely as we hoped it would. With the second test we have seen the following:
1. Whenever we search on the key phrase (two random non-existent words) the advert never showed. We have probably racked up around 100 searches on it now.
2. When we searched on the key phrase and another random word it showed some of the time but not all the time (approx 1 in 6 searches). This pattern repeats itself with different random words.
3. Whenever we reverse the order of the words in the key phrase, it shows every time.
We increased the daily budget to £500 to make sure that 2. was not a result of a budget constraint.
It therefore feels like the adgroup has not reached a high enough confidence level for adwords to broad match it all the time. This may mean that if you get relevant clicks, when it shows, that the amount of broad matching will gradually increase? Any other explanations welcome!
I think your first test result was due to the updating intermittancy and you just 'got lucky' with your test hits.
I see no evidence for the hypothesis that one can minus an exact match. (I sure wish it were possible)
It therefore feels like the adgroup has not reached a high enough confidence level for adwords to broad match it all the time.
This is not my understanding of how Expanded Broadmatching works (maybe I am wrong). It has been my experience that new BM'd terms WILL BM immeadiately (aside from server update time) and LATER over TIME 'drop out' of BM'ing if the 'secret' CTR is not met.
[edited by: nyet at 4:33 pm (utc) on June 21, 2004]
Also did you manage to get the 100% no show when you did the exact match?
Do you think there is any chance that the "1 in 6" delivery will improve as broad matching gets greater confidence?
see my edit to my post above.
Also did you manage to get the 100% no show when you did the exact match?seems so. But if you ignore the brackets [ ] this is proper behaviour. I think the 1 in 6 is the anomoly.
It has been my experience that new BM'd terms WILL BM immeadiately (aside from server update time) and LATER over TIME 'drop out' of BM'ing if the 'secret' CTR is not met.
I expect you are right. It is not something we have investigated in any depth yet. However we have lost some of the BM phrases pretty quickly in some cases (within a few hours anyway)
I expect you are right. It is not something we have investigated in any depth yet. However we have lost some of the BM phrases pretty quickly in some cases (within a few hours anyway)
My theory about this is that the (automated) turning on and off of the BM'ing for a word is fluid. 'Off' in one search and 'on' in the next depending on the 'added/modified' EBM words.
I suspect this is why G does not reveal the EBM status of the words to the advertister.
HOWEVER, we certainly have terms which stopped alltogether. (last we checked) And I think that not knowing this really stinks!
It seems in the last few months the 'secret quotient' has been relaxed somewhat...... ( I think?)
css
-[css]
because I was the 1st in organic results for css. It worked exactly how I expected -- my ad showed for anything containing css, but css alone.
