|Results - Test with no www in display url|
adwords: www.example.com vs example.com
| 11:37 am on Mar 20, 2012 (gmt 0)|
Everywhere I read, they say to remove the www. from the adwords display url, so you have more room to add a descriptive term to the end.
I found that if you already have a well written ad that is , have a , and a ; you will get worse results by removing the www to make the display url more descriptive.
I was a little shocked by the results.
I was sure that:
would perform better than
but at a 99% confidence interval, there is a significant differnce on ctr and conversion between the two.
The only thing I can think of, is that if the ad is already well written, then cluttering up the display url is of no benefit.
Also my company url, is suggested of a benefit in its name, which may be a factor.
Also, most ads I see include the www prefix.
Unfortunately my url is a long name, so i am unable to test www.example.com/pink_widgets.
Has anybody seem similiar results with and without the www in Adwords A/B testing?
| 2:31 pm on Mar 20, 2012 (gmt 0)|
AdWords adds the www whether you want it or not.
| 2:33 pm on Mar 20, 2012 (gmt 0)|
"Everywhere I read, they say to remove the www."
'They' are always talking about the general case, not your specific one. Test it (as you did).
| 3:03 pm on Mar 20, 2012 (gmt 0)|
Used to be that you had control over how your display URL was shown, less so today. As Netmeg said, the www is added automatically when displaying ads. But before they did, it did seem to make a positive difference in click rate when the www was NOT there. They also now lower-case the domain name part.
However, I noticed that it seems to be the reverse today. Small test but two identical ads except for the display URL: one with the www, one without, words in URL capitalized, just the domain name. Doesn't make sense but the ad with the www has a higher CTR, and by a good margin too that it can't just be by chance.
The only thing I can think of is that search partners may respect the capitalization and structure and this affects the data. Haven't checked it out but will soon.
Never did comprehensive tests when adding keywords after the domain but good point that it may be cluttering with no benefit. Although my instinct is to make it look more like a natural link.
| 3:29 pm on Mar 21, 2012 (gmt 0)|
as we're trying to improve CTR, so is G... does get confusing, and makes impossible blanket statements about which answer is always right.
| 4:47 pm on Mar 22, 2012 (gmt 0)|
Yeah, can you image in any other publishing industry, where the publisher can get away with changing your ad copy without your permission.
I wish Google would either just accept or reject the adcopy, or suggest changes; but never change how the ad is displayed based on the phase of the moon, or the alignment of the planets.
Or at least have a profesional advertiser button, to click so you can manage the presention of your own ads. The unwashed masses would leave the button blank, and let Google manage the ads based on general trends.
| 6:06 pm on Mar 26, 2012 (gmt 0)|
I agree that testing is the best way to go but even at that you may not get accurate split tested results because Google now will take domain names and add them to the title line of ads in top positions above search results.
There seems to be no rhyme or reason to when they display results with the URL in the title line so I think it may be difficult to do a clean A/B test.
Personally I like to add some extra messaging to the display URL but have found that when Google moves the display URL to the title line it can look strange (or messy) when the extra messaging is included.
| 9:47 am on Apr 2, 2012 (gmt 0)|
lgn1, I've found much the same thing, with extra description after the domain the CTR tended to be worse, not a a lot worse, but slightly so.
I think it is because the extra sub directory/description can look a bit spammy, and some people might see it as a comparison shopping site or similar.
The ads were well optimised previously, with the domain being an exact match for the search term.