An interesting and, as far as I know, unannounced change has just taken place with the Google Keyword Research Tool: If you go and check the search volumes for your targeted keywords, you’ll likely find that all of the numbers are far lower now than they were before.
The change affects all search volume numbers. The ones we’re interested in for SEO purposes are generally the exact match volumes, though. In some cases, the drop in search volume is dramatic, with numbers being only fractions of what they were just a few hours or days ago and in other cases the changes are less dramatic. From the quick scan I had through my keywords list, it seemes that all of the search volume numbers dropped by at least 50%.
So, what’s going on, here?
Well, it’s been known for a long time that the search volume numbers presented in the keyword tool tend to be inaccurate. More precisely, they tend to be inaccurate on the side of exaggaration. I don’t think I’ve ever heard of someone getting surprisingly more visitors than expected when hitting the number one spot for a targeted keyword. On the other hand, I’ve often heard of disappointingly low traffic coming to the number one spot of a sought-after keyword.
There’s a thread here on the Warrior Forum and another one on the topic here on Backlinks Forum. From the developing discussion on both of these threads, it seems like the numbers are now more accurate, overall. So, the simple explanation for what happened with the keyword research tool is this: It was inaccurate before and now it’s fixed.
But what are the implications? So far, there hasn’t been an official statement by Google, so we don’t know what was changed or why. Did they simply improve the filtering for automated, non-human queries? If so, does that mean that there are far more non-human queries to Google than there are actual searches done by human users?
And what does this mean for niche marketers and SEO’ers? Will the change in numbers impact what kind of keywords they go after or even change/invalidate entire strategies? Also: Is that part of the reason for the change?
And for some more unpleasant implications: What does the change mean for Google itself? Is it actually getting only a fraction of the traffic we formerly believed? Have they been deceiving advertisers with grossly inflated search numbers?
Don’t get me wrong: I’m not predicting the end of SEO, here (as I’m sure many will…). Traffic volumes themselves haven’t changed, only the search volume estimates have. Still, it will be interesting to see what develops of this. Also: Prepare for some epic moaning and complaining as well as apocaliptic forum threads (particularly the Warrior Forum is a favourite for those).
What’s happened to your keywords? Will you change anything, now that the numbers are different? Let me know in the comments.
Cheers,
Judging from the comments (thanks to all commenters!), we all seem to agree that the new numbers are generally more accurate and that’s a good thing. It’s certainly better to work with accurate numbers when you’re doing keyword research. Now, you should be able to look at the exact match search number for a keyword and assume that a number 1 ranking for that keyword will get you 40% to 50% of that traffic.
On the other hand, we’re all still wondering about what caused the change and why the numbers were so skewed for such a long time, before.
I also still have not seen any official statement from the side of Google.
As was kindly pointed out by Josh in the comments, there’s a semi-official statement about this change here. A Google employee wrote:
“If you use both the previous and updated versions of the Keyword Tool to search for keywords, you may notice differences between the tools for statistics on Global Monthly Searches and Local Monthly Searches. This is because the previous version of the Keyword Tool provides search statistics based on Google.com search traffic and traffic from search partners, while the updated version of the Keyword Tool provides search statistics based on Google.com traffic only. We’ve updated these statistics based on user feedback, and hope you find them helpful for keyword selection.”
Very interesting Shane, More accuracy should help, though I confess to always being slightly baffled by KW research, mostly because every marketer tells you a different way to do it.
Thanks for the “heads up” Shane, I had not heard of this. I think it’s good news, after all faulty intel is worse than no intel in my view. We all knew the numbers were too big anyhow, but it’s interesting to speculate on why big G. has changed what they publish.
:-)
Alex
Hi Shane,
Bloody Hell!!!! Just checked a few of my keywords and 1 in particular ‘tumblestones’ has gone from 3600 monthly in the UK to 590. Thats a drop of 84% !!!!! Now I know why the traffic has not been as good as I thought it would be.
Thanks for the heads up
Jon Lake
Thanks for this update. No I don’t think this will affect SEO. At the end of the day the traffic you are getting is the real life situation whatever the search volume results are by google. It might make it more difficult for new comers trying to choose keywords with good search volume but if you go for building an authority site rather than a mini sniper site then you should be OK.
Howzit Shane,
Seeing this is my first reply, allow me just to share a few sentiments.
Your information and “free resources” are out of this world. You have cleared up so many things w.r.t to the world of IM. The Keyword Research tutorial is really great and finally I have laid that topic to bed…(given the discussion I might have to wake it up again:-)
Thank you for doing such a great job…your integrity and honesty really sets you apart and makes you true blue…a cut above the rest.
Given the discussion…maybe it was long overdue! Frankly, I think and feel it is going to open up a can of worms…for google at least. Questions around true market ownership will be a hot one, PPC Revenue Drop another and one which stands out for me is Yahoo, Bing and MSN etc. It seems they are worth more than the credit we give them.
I guess it is a scenario of digging past the rocks and stones and finding possibly diamond worth opportunities?
What is sad and really it is…What happens to ALL the hard work we put into getting those SEO campaigns and websites up and running, which was all based on those “search volumes”
Carl
Having just bought an exact keyword domain and had 10 articles written for it, I checked Google again today.
My keyword has gone from 5200 monthly local searches for the exact term to 140.
I will carry on and see if it all stacks up.
If the results are now correct, it looks as though micro niche sites are going to have a hard time.
Probably need to build authority sites to get enough traffic.
Now, let me see, where did I put that auto-blogging plugin?
Cheers
Sandy
Not sure why you would think that micro niche sites are “going to have a hard time”… If the numbers are more accurate now, then it should be easier to pinpoint niches that are worth targeting. It’s not like people have all of a sudden stopped searching on google… If the numbers were wrong and have been corrected, that is a good thing.
Well, I did a huge amount of research to find a niche that had over 4000 local searches a month for the exact term; low competition, a domain name and had good commercial intent.
To now find that if I get to No1 in Google I will get about 50 visitors a month isn’t exactly making me do naked cartwheels across the room.
Given the length of time involved, and the degree of difficulty, my concern is just how many micro niches will have enough traffic to make them viable.
Just a thought.
Cheers
Sandy
Thanks for the great info…so glad that you made this post…google keyword tool is changing rapidly and is great to know of all the updates
Thanks for posting the update, good information provided in this article. I also found changes in the search results numbers, so i always use different keyword search methods, some free and some paid which gives me an overall idea
thanks
As I’m following your Backlink Battle Plan, I’m not too worried about the change in Google’s search data – after all, you are so right when you say that it is only a recalculation of the over estimation, not the actual search traffic.
We marketers may end up feeling bad, despondent or even cheated by Google, but what’s new about that? All we need to do is to keep focused on the way forward and spend more time listening to sensble advice from people we trust (You, for example!) and accept that Google will do what Google will do and understand that we just have to deal with it.
I always thought the keyword tool search volumes were suspect and now I guess the truth is out ” they were suspect”.
Let’s hope its now fixed and that we have good information to work from.
thanks Shane
Hi Shane
As always thanks for the information, I had always wondered about Googles calculations because if you use their analytics as a measure of traffic to your sites it doesn’t agree with the numbers of click throughs leaving it
I had never been able to understand that more people left than came in the first place, maybe that’s next for a rethink
This kind of sucks but at the same time will probably present new opportunity as we begin to figure out what this is really all about. Either way this is valuable news I’m sure we all can use. Thanks Shane.
I doubt this will make any difference. The bottom line is that anyone that was into SEO for more than a few month realized that the Google numbers are way off base.
This shouldn’t come as a surprise to anyone. As you said, there is no one that suddenly got more traffic than what he expected and too many that got the number 1 position only to find that the traffic volume is not close to the numbers shown by Google.
The surprise is that Google didn’t make any announcement regarding this change and I wonder if they will.
Actually it is a good thing.
Truth is now displayed.
No longer we have to frustrate ourselves for choosing a “high-hopes” keyword.
Thanks for all the awesome comments!
The general sentiment seems to be that it’s good and hopefully we now finally have accurate serach volume numbers.
Also, that it doesn’t make a difference in practice.
I tend to agree. If the numbers truly are accurate now, it’s great for keyword research!
But I’m also wondering why there wasn’t an official announcement and am looking forward to seeing/hearing one.
G’day Shane and all,
As we all knew, Google’s figures were always pretty rubbery to say the least, and there was even a major difference between Google’s own two keyword tools, so I would say that this is definitely a step in the right direction.
But we still don’t know that these figures are correct, just that they seem to be MORE correct than before. So I would suggest that anyone who uses set guidelines for monthly search and competition numbers just needs to revise their minimum monthly search figures upwards, maybe double or triple what they were using before, and carry on.
For myself, I always suggest doing a ‘reality check’ using a quick PPC campaign to gauge the ACTUAL traffic numbers rather than taking Google’s figures on trust, and I will continue to do that. But at least now, Google’s figures might mirror some sort of reality, so that has to be a good thing for niche and affiliate marketers!
Shane, thanks for bringing this to all our attentions, greatly appreciated.
Terry
Hi Shane,
This news is abit disconcerting. However all that means is that we have to tighten up our SEO strategies. After all, we want accurate keyword stats and if they are now more in line then our endeavors should be successful especially if we keep on using The Backlink Battleplan. Remember, we are still well ahead of the other marketers by using BBP so lets keep up the battle.
I would prefer accurate numbers. It has often been said that unique searches make up about half of all searches.
It could be that Google was “estimating” those into the equation before.
It will make some strategies change. For example when people say target only words that have xxxx number of searches a month.
Either you need to lower the search volume number for your starting point or you need to search harder for good volume low competition keywords.
It will give a more accurate view of what is feasible though.
I don’t think it will impact Google’s share though. I know from my stats that Google is still the lion share of my SE visitors.
It will be interesting to see the threads at the Warrior forum though :) Maybe I will go start one haha
Good point. Now I’m even happier that I always recommended avoiding arbitrary search numbers in your research (typical “at least X searches…”). I don’t have to re-do my keyword research guide. :D
Yeah, I also get far more visitors from Google than from the other SEs.
I can’t figure out how this change can threaten SEO folks like me. I would rather have more accurate search volumes than hyped up data because I am the one who will have to explain to my clients why the traffic is not up to the expectations derived from exaggerated data previously provided by Google.
As we all know, or at least I hope we do :), accuracy is a cornerstone of SEO so this is good news, if this data is more accurate of course.
Thank you for your post and your training which I recommend to anyone who is taking SEO seriously.
Boris
Dear Shane, thanks for the essential information.
IT happened to me as well – my keyword phrase was showing about 50K pe mo and now its down to 8k!
Market Samurai still shows the older higher figures though.
Problem is I bought a premium domain for my keywords based on Googles information and now they changed their tune.
I wonder if I have legal recourse??
I doubt it, to be honest.
8K plus related searches is still good, if there are a few buying keywords you can rank for.
Thanks for this article. I checked a keyword for one of my sites a few days ago and it has gone down by a huge margin, having gone through them now it seems it is correct and something has happened. One of the keywords has gone from 14,000 phrase local searches to 720 ! When I first found out it was 14,000 (about 4 months ago) I couldn’t believe that there were so many people out there searching for this and that there were 450+ searches everyday, it now turns out this is maybe not true and a figure of 24 per day maybe seems a little more realistic?
Still I feel there are some points to be addressed:
Firstly Google needs to publish something official. Why hav’t they already done so? Did they really think no-one would notice? Have they got to come out and admit a mistake, but no doubt it will be made out an improvement to there servcies.
The margins are two large. I.e the keyword mentioned above is about 5% of what I thought it was. Or put another way 13,270 less searches per month or 440 each day!!
I lost confidence in Google adwords a long time ago, simply because with a budget of £50 per day it was not working as it was 2 years ago. As you know the bidding for keywords is based on what people are prepared to pay, so if I thought a keyword was getting 14,000 searches I would be prepared to pay more, so would someone else and so on driving the price up! – when in reality we now know this is rubbish so will the cost per clicks fall when people start not paying for them?
I’m boring myself now but I could keep ranting on! – Google have made a mistake, were they hiding this for some time and had no choice but to change it before getting court out?
Well, for me it was really a surprise, but a pleasant one in as much as many of my keywords had fairly significantly improved global monthly searches and a longer list of related words too.
Thanks for the article Shane, informative.
Interesting! I was thinking that all of the numbers went for a nose-dive.
Hey Shane,
Thanks so much for sharing this extraordinarily important information on the recent changes that Google imposed on their keyword tool data.
Although I absolutely agree with everyone who is pointing out that more accurate numbers will ultimatley benefit every markteter who´s doing SEO to drive traffic to his web-pages I´m still mad at Google as of today.
You see, I´ve put a lot of hours of work to research some local market search numbers and have prepared these numbers to offer local businesses my SEO services – luckily enough I haven´t send out the data yet. The tables have completely turned with the new search volume data as some niches that seemed really promising are now shredded into pieces. I have to consider numbers which dropped by 78 (!) percent, yes 78 percent and that is humongous, that is just plainly crazy and I wonder how long Google knew that their data was so much off the course…they really betrayed thousands of marketers with their obviously pumped-up numbers and it wouldn´t be any surprise to me if some of them will try to sue Google for their completely inaccurate numbers!
Anyway, so glad that I found out about these changes before I started promoting my services with wrong numbers and I guess it´s now time to go back to re-work with the new and hopefully more accurate search data.
Keep on rocking, Shane. Your information is unrivaled!
Mirko
P.S. I just had to write a short blog post about this myself to share it with my community.
Hi Mirko,
I think you describe the problem really well: No one minds more accurate data. The problem is that the difference is so huge and the change from the larger to the smaller numbers is an inconvenience for any running or developing projects.
I still haven’t seen an official statement from Google, which is strange. I kind of expect there to be an explanation along the lines of “before, we counted X and Y for the numbers and now it’s only X” (e.g. network and search or something like that). I wonder how things will develop.
I think Google inflated their number of searches when their AdWords program came out in order to attract as much potential advertisers as possible.
It seems now that they have enough advertisers, they decided to make searches as close to real number as possible.
Hi Boris,
Thanks for your comment!
Don’t think that’s the case, though. If they only waited to get enough advertisers, the adjustment would have happened way sooner…
But no doubt the inflated numbers played in their favour!
To say it more simply than Boris did – Google deliberately lied to their customers (PPC advertisers) in order to get their money. Sounds like criminal fraud to me.
Google has responded to this issue at
http://www.google.com/support/forum/p/AdWords/thread?tid=5241a33035617b73&hl=en
Awesome, thanks! I’ll add a reference to the article right away. :)
Hi Shane,
Great post. I’ve also noticed these changes from when they initially happened.
Whats funny is that I did some KW research in Micro Niche Finder, purchased a domain and when I checked it a few hours later in GAKT, it gave me some really strange numbers.
Anyway, my point is that moving forward, us SEO’s are used to change and have to adapt to whatever is thrown at is e.g. caffeine/mayday algo updates etc.
We can do one of many things, either give up altogether, keep searching for more golden nugget keywords with higher search volumes or really bring our expectations down to reality.
I think the latter is probably more likely as I have also had some good rankings on site sites but not got the traffic that one assumed. Initially I thought that this was a VERY bad thing for SEO but after thinking about it, I think it actually is good since it is realistic and we have a much and more transparent view (I say transparent quite loosely though).
Great blog and love your videos.
By the way, its always good to go direct to GAKT to tally up your numbers if you use tools such as MNF, Market Samurai etc just for due diligence purposes.
Zaheer
Thanks for your input.
You’re right: There can be differences in what KW tools show and what the Google tool shows. Sometimes, the KW tools are also not clear about what match types they are displaying.