| 4:15 am on Jul 9, 2008 (gmt 0)|
Sweet - thanks for the information. With the public Overture tool now officially out of the picture, this is extremely welcome.
| 5:01 am on Jul 9, 2008 (gmt 0)|
I presume these are "broad match"
No 1 [ exact match ] terms we hold have nothing like the volumes claimed [ not even close ] , even allowing for variances on regional G filters.
Oops .... i stand corrected , you sort them by broad , exact , phrase and negative
... but still looks way to high
[edited by: Whitey at 5:09 am (utc) on July 9, 2008]
| 5:10 am on Jul 9, 2008 (gmt 0)|
It's broad match when it first loads, but there's a dropdown box to switch the Match Type. Also these numbers are Google's total search volume, whether clicked on or not. That's bound to be a whole lot more than any one site's click volume on a given word, even if you hold both organic positions 1 and 2.
Note, there's a thread about this in the Adwords Forum [webmasterworld.com] for discussions related to advertising - so we can focus on organic search discussions here.
| 5:23 am on Jul 9, 2008 (gmt 0)|
Ahh, news of the year. The paid keyword tools will now have to come out with some very good USP, else perish. An Analytics of Keyword Tool market.
[edited by: McMohan at 5:25 am (utc) on July 9, 2008]
| 5:47 am on Jul 9, 2008 (gmt 0)|
Great tool indeed but doing some little mathematics by multiplying avg monthly search X CTR of 30% for KWS I appear organic #1,2 the figures I see does not matach much accurately with my original stats.
| 5:49 am on Jul 9, 2008 (gmt 0)|
|the figures I see does not match much accurately with my original stats. |
What about if you apply CTR %'s e.g. No1 position = 20% of searches approx .
Does it come close for you?
| 6:04 am on Jul 9, 2008 (gmt 0)|
thnx, great tool, it shows that how much advertisers believe each keywords and how these keywords imp for them also for users
| 6:30 am on Jul 9, 2008 (gmt 0)|
Google data with numbers (even if rounded off)... great news!
Now, if only the data download links would work I'd be a happy camper. I can't get the frame to print either. It is possible to get a space-delimited text file by highlighting all of the framed phrase content, but it would be a major pain to try to use.
What's really good to see, though, even if I have to use ball-point pen to write them down, are numbers for infrequently searched but very important phrases in B2B markets like big-ticket, large company hi-tech widgets, where the buying cycle might be a year or two or three. The data simply falls through the cracks on other engines and tools.
It's amazing how much time this info would have saved in organic targeting (and yes, the companies buy AdWords too, so Google wouldn't be undermining itself by giving away such data).
I'm seeing some glitches in the tool right now. Sometimes the green bars show up... sometimes they don't. I'm looking forward to getting some downloaded info to compare with other data I have.
| 7:41 am on Jul 9, 2008 (gmt 0)|
Great news, it is highly welcome. All keywords tools will be out of market, Google is killing many :) .. Will you need wordtracker now , naaa naaa.
For organic search the separation of exact match and broad match will help a lot.
| 8:10 am on Jul 9, 2008 (gmt 0)|
On a set of phrases in a keyword space I'm fairly familiar with, I'm seeing some major anomalies between the Google numbers and the Trellian tool... and they're very hard to reconcile. It's leading me to believe that right now, at least, the Google tool is completely dropping some phrases big time.
| 9:03 am on Jul 9, 2008 (gmt 0)|
If you can trust the numbers this is the best news I've had all year. I hope it remains in place. Imagine the pain that must be going on in the keyword research industry today. This is pure gold. Once you start looking at the numbers you quickly work out how off base some of these tools are.. So who is telling the truth? And what does google stand to gain by sharing this data?
| 9:30 am on Jul 9, 2008 (gmt 0)|
|If you can trust the numbers |
Early days yet, and I need to do more checking, but I don't trust it - AT ALL.
Biggest margin of error so far :
3 word term estimated at 40,500 / month (exact match)
We have 1st organic place, plus top AdWords spot.
Total impressions according to AdWords is 138
Total paid and non-paid clicks 91.
That's over 290 times real demand - about 10 times worse than Overture ever was.
| 10:10 am on Jul 9, 2008 (gmt 0)|
On some searches the [exact phrase] shows a number, on others the same [exact phrase] shows insufficient data, so the grains of salt can start to be added.
| 10:25 am on Jul 9, 2008 (gmt 0)|
|Now, if only the data download links would work I'd be a happy camper |
| 12:26 pm on Jul 9, 2008 (gmt 0)|
This is great news for those frustrated with some of the less than reliable KW tools out there. Even if these results are skewed somewhat, you cannot argue with the reliability of the source or the fact that this immense data is FREE
Not sure the geo targeting is quite right - there are some Americanisms thrown into my samples which cant be correct.
| 1:54 pm on Jul 9, 2008 (gmt 0)|
My experience with the keyword tool since they introduced it is that it's probably a good indication of trends, but I wouldn't want to bet the farm on its pinpoint accuracy. I suspect that will still be true, even with numbers attached.
I'm wondering if it's not still beta though - I don't get the new tool every time I go there, sometimes I get the old one. Easy to tell at first glance; they added a captcha to the new tool, and there wasn't one on the old.
| 2:34 pm on Jul 9, 2008 (gmt 0)|
What puzzles me is how the numbers can be wrong in any way. The AOL data leak shows that (for at least some users) they do record every search.
If there are inaccuracies I can only think of two reasons
(a) their data set is too small and (like Wordtracker) they are multiplying the numbers they have to scale for the whole of the web as they see it. Depending on the demographic of users we could get inaccuracies because they don't have a wide enough sample.
(b) they are deliberately making it vague or even over-estimating to encourage bidding wars.
| 2:54 pm on Jul 9, 2008 (gmt 0)|
Looks like the numbers are here for real - the Adwords Blog posted an article about it [adwords.blogspot.com].
When it comes to funky data, this may be at least a partial clue:
|If you access the Keyword Tool from within an ad group, the search traffic statistics will factor in your campaign's country and language targeting (if you target a region or city, only the country will be reflected). If you use the standalone or external Keyword Tools, your country and language selections will influence these statistics. |
| 6:53 pm on Jul 9, 2008 (gmt 0)|
I really, really want to congratulate Google on providing helpful information BUT I can not. The information is not close to being accurate.
I follow several industries and I know the keywords and traffic levels by heart. When I use this tool I can not recreate the data that I have personally seen across my network of sites for years. Even accounting for the differences between broad match and exact match, the numbers are just off.
The really weird thing is that depending if I use Google Trends or Google Adwords it tells me different keywords are the highest traffic keywords.
The paranoid, skeptic in me is thinking google is purposely diluting the accuracy of the data. The more grounded side of me is thinking that they are just not using the right data sets.
| 8:17 pm on Jul 9, 2008 (gmt 0)|
For certain KWS even if I assume CTR of 10% for my top organic positions the data seems to be quite inaccurate while for some KWS its close to perfection.
| 10:27 am on Jul 10, 2008 (gmt 0)|
I've crawled over the keywords in my niche and found a couple of volume terms. My site is strong so it only took an hour to build a quality page of content and we hit the results quickly at #2. In 24 hours i'm going to know how close this search term is but already it looks good. It's quite an education if the numbers are real.
| 10:57 am on Jul 10, 2008 (gmt 0)|
Hmm, sometimes I see a little blue horizontal relevance bar in the chart against each keyword selection, and at other times I see a stylised 12-month summary brown bar graph of search volumes. Both come with numbers.
Not sure if I am using a different URL for one of them or something.
| 2:16 pm on Jul 10, 2008 (gmt 0)|
I get different things at different times too, g1.
| 10:24 pm on Jul 10, 2008 (gmt 0)|
Yeah, and there is more if you change the end number on the parameter to 1 or 2 or 3.
| 5:12 am on Jul 11, 2008 (gmt 0)|
Does anyone care to take a shot at the "why"? Google has done ths for their own benefit not ours. So how does it help them? Does it help them sell more adwords perhaps? I think this is what they are banking on.
| 5:44 am on Jul 11, 2008 (gmt 0)|
Sure, if potential advertisers can see numbers for keywords that they haven't tried, that can encourage Adwords spends, and even more, better ROI for the advertiser and therefore fewer discontinued campaigns.
But I also think that Google really does want to help the average webmaster. They feel that improving the Internet as a whole will benefit everyone involved. Call that "enlightened self-interest" if you will, but I'm not suspicious of everything that happens in Mountain View, and I sure don't see Google as my enemey.
I commented on this a few years back. The whole playing field that Google and webmasters generate is a fascinating paradox. First, we do need each other. Second, our self-interests are sometimes in conflict. This is what game theorists call a "competitive-cooperative" environment. It's a good idea to stay conscious of both sides of the paradox. If you get too far in either direction, you end up making foolish moves or missing out on good things.
| 11:19 am on Jul 11, 2008 (gmt 0)|
Even with the benefit of search numbers and having the top rankings there are so many variables in play these days it's only ever going to be a rough guide, but I will take it.
| 12:58 pm on Jul 11, 2008 (gmt 0)|
The site I am helping out at the moment has found five related keywords and phrases that they had never considered before, from just half an hour playing around with this. As a charity, they would never have paid for that information.
| 2:31 pm on Jul 11, 2008 (gmt 0)|
As someone that has had beta access to this feature for over a year now my tests show that the numbers are relative at best. I tested this by comparing a few hundred keywords that we always bid 1st in adwords to the number of searches that the tool shows. For terms that we averaged a position of 1.1 the impressions were about an average of 25% of what the Google tool was reporting for the total number of searches. After escalating this issue to our Google Adwords team at G they responded that the impression numbers where the single best way to determine search volume (if you bid an average of 1.5 or lower).
| This 41 message thread spans 2 pages: 41 (  2 ) > > |