Results:
Broad - the most imporessions (obviously)
Exact - as many clicks as broad (surprising), much better CTR (obviously)
PHRASE - MUCH fewer clicks than broad/exact (surprising) lower CTR than broad (surprising)
Can anyone give reasonable explanation theories for my "surprises"?
Note: this was for one specific adgroup I tested. Most did not come out this way, but this one did. Just curious for any theories behind stats
As broad and exact match can be shown by Google more often than phrase match, they tend to build up a good CTR quicker than phrase match, and then will dominate, which would fit with your findings.
But it varies from word to word. Never be too surprised by what AdWords throws up!
For the most part, I use phrase match; the broad match (even with mass quantities of negatives) would just bring too many impressions and clicks, but few conversions. There were very few instances where exact match worked as well for phrase match. But as always, YMMV.
and you dont use exact?
I've had success with broad, but it needs to be cared for. I've had a lot of success with exact, that goes without saying.
But only going with phrase? For a limited budget, that works. But why not use broad?
I am talking about those matches for same kw
Phrase and Exact are the same thing? Now I'm confused...
Can you explain how there is no difference b/w [] and ""?
But only going with phrase? For a limited budget, that works. But why not use broad?
In a very limited number of circumstances, I use broad - mostly for phrases where the words in the phrase are fairly specific to the industry and wouldn't likely to be used in any kind of a generic sense.
I do this because I'm not looking to gain traffic, I'm looking to gain conversions (which, for the most part, are sales)
For example, several clients have products or services that could be summarized in different ways, using various combinations of common everyday words, but if you rearrange those combinations (using the exactly same words) you get just a ton of traffic from people looking for types of get rich schemes, or looking for job opportunities. That's not traffic we want, and the negatives only work to a certain extent - after all, it's the same words, just arranged differently. In some cases, we're trying to market exclusive B2B, and we don't want keywords that attract B2C shoppers.
Way back before I figured this out, I burnt through a whole lot of client money just acquiring as many clicks/traffic as possible. It's just as important with a big budget as a small one (specially if I want to keep my clients!) And oddly enough, the hardest sell for doing this was with some of my clients, who couldn't help but believe the more eyeballs we brought to the site, the more they would sell. I'm trying to save them money by taking a less scattershot approach, and they just want to blast with both barrels. I don't really want to bring in even more non-converting traffic (that we have to pay for) than we get now naturally.
Googles answer on this is the following:
Exact match and broad match get all queries... Phrase isnt necessary. Use only for book- or filmtitles...I am talking about those matches for same kw
That has not been my experience. May I ask where they answered this?
Broad:
Lots of impressions, but low CTR and lead quality / conversions are terrible. I still use broad matches on a very limited basis, only to catch keywords my phrase match keywords might otherwise miss.
Exact:
Insufficient impressions, even at high CPC bids, to keep the keywords active. Gave up on exact matches for that reason. Might be great for some verticals and national advertisers, but virtually useless for me.
Phrase (3-5 individual words per keyword phrase, 2 of them being the locale indicators):
Very reasonable CPC bid rates, best CTR, best lead quality and conversions. My locale specific (95% phrase match, 5% broad match keywords) Ad groups average 11% CTR month to month, some as high as 40% every month, and conversion rates range from 5% to 25%.
I also have a huge list of campaign level negative keywords, plus a handful of ad group specific negative keywords, and add more every week as I examine my site logs. So impressions, even for broad match, should be very relevant. So, I am seeing less than .5% of AdWords referrals from irrelevant searches (1/200), and that is going down every month.
couldnt you run broad ads and avoid this "bum" traffic by focusing your creative?
My creatives are EXTREMELY focused. But if I use broad match, I still get the too much of wrong traffic. For the most part (and after a ton of testing), Phrase works very well for me, with a nice balance of impressions, clicks and conversions.
That has not been my experience. May I ask where they answered this?
And if you ran the three concurrently, depending on your bid and CTR, you'll get some weird data. If you start one, with other 2 paused and rotate, you'll get data that's easier to intrepret. If concurrent, they each competed with each other and if you didn't set things up to favor specificity, you can get broad matches triggered where you think it should have matched exact and so on.
If concurrent, they each competed with each other and if you didn't set things up to favor specificity, you can get broad matches triggered where you think it should have matched exact and so on.
I currently have 3 adgroups per term - broad, phrase, exact...
