For the past decade, most of us in the field of search have relied on Google's AdWords data (either in the public tool, the API or the tools inside AdWords accounts). It's the best source we've got, but many marketers may not realize that sadly, the numbers and queries may not always match up to what's actually happening on Google's search engine. I'll illustrate with an example.
An SEOmoz blog post ranks in the top 2-3 results for many keywords around the phrase "blog traffic." Here's a screenshot of some of those rankings:
I went into our Google Analytics account and pulled the related keywords along with how much traffic they've sent in the past 30 days:
Then I went to Google's AdWords Tool and searched for "blog traffic" to compare the suggestions:
Here I got confused, because many of the terms that we receive traffic for are NOT shown above in the list... Is Google hiding them? Do they not know about them?
To be sure, I typed them into Google's AdWords Tool manually, performing [exact match] searches only:
Holy cow... There they are. So, AdWords does have volume for these, and will display it, but only if you enter them exactly (or rather, "more exactly" - you can find them if you do sets of imprecise, but closer queries, too). I made the chart below to illustrate which terms were available from the broad reserach:
As you can see, there's ~50% of the terms not shown in the suggestion list, which is fairly substantive and could lead to some serious missed targeting opportunities.
THE IMPORTANT LESSON: Running discovery-focused searches in AdWords may not show you all the valuable/high-volume keyword phrases connected to a word/phrase.
There are a few ways to address this challenge:
- If you have the budget, my top recommendation is to buy a few, very broad keywords in AdWords, send them to a relevant landing page on your site, but realize you probably will lose money on the campaign. The goal isn't conversions, but rather to learn by watching the keyword terms/phrases for which you get impressions. This is also great conversion-testing if you have the budget to invest, but even a week or two of data can be highly valuable for future keyword targeting.
- When searching in AdWords, start broad, and then enter narrower queries and note the new phrases that come up. Make sure to use exact match, and be diligent in testing variations. Google only lies through omission.
- The relative numbers of searches aren't perfect (as you can see above), but they are relatively decent. In fact, I'd say they've improved in what they show vs. the actuals you'll see compared to prior years. However,
- Use your own analytics as a guide to find new terms/phrases you might be imperfectly targeting. And if you see keyword variations that have a unique or different intent, it might even pay to create a more targeted page for that query, and you often need less work to rank, since Google uses the "indented results" system to drop a second URL from the same domain directly underneath the first one on a given page.
Now I'd love to hear from you - what are your experiences around keyword research in AdWords? Are you seeing the same thing we are? You can share your thoughts in the comments.
BTW - Given that 30%+ of our referrals from Google searches are keyword (not provided), I'd venture to guess that all of the numbers from our analytics are underreporting by about that same percent. Keep that in mind when comparing the data from AdWords vs. our analytics above.
Hi Rand,
I believe that there is not a single tool out there that can provide the most accurate information. Besides wanted to ask you, did you select "only show ideas closely related to my terms"?
Let’s take a simple example here -
Search for a keyword like Bail Bonds and do not select "only show ideas closely related to my terms" and then search for Bail Bond Agent, you will see it right there. However if you select "only show ideas closely related to my terms" and then search for the term "Bail Bond Agent" you will not find the keyword.
So one care that needs to be taken is that you DO NOT select "only show ideas closely related to my terms" when you are looking for a broader range of keywords.
Also there is a considerable difference in the number of keyword ideas that the tool gives when you are logged in using your adwords account.
For example, for the keyword "bail bonds" i got 100 suggestions as compared to 800 when i was logged in using my adwords account.
There are some amazing tools within Google Adwords like Placement Tool, contextual Targeting tool etc which are some of the most underrated tools as far as keyword analysis is concerned.
Besides there are others like SEM rush (analyze your competitors data), ubersuggest, Soovle which can be used for keyword analysis.
Of course then there is Google Webmaster Tools impression data and Google Analytics keyword data.
There are cases where I have noticed that a particular keyword is showing 0 searches in exact match in the keyword tool whereas in Google Analytics I am getting over 1000 visits via that exact same keyword.
Yes there are a few flaws but then which tool does not have flaws. Please remember people that its a FREE tool. Although it’s not the most perfect tool in the world but i believe that if we take certain precautions it can be used in the most optimum way. It will remain the number 1 keyword analysis tool for me. So I will definitely give it a Thumbs Up.
- Sajeet
Fully agree with Sajeet. Why didn't you just log in to AdWords (you don't even need to have any campaigns running and it's free), I made a similar search for [blog traffic] and got 554 keyword ideas.
For sure, it's not really fair of Google to hide functionality this way, making more people sign up for AdWords, but it's not right either to throw mud on one of the most profound keyword tools of all SEO times!
Yeah - you can see that's what was used above. However, I also tried this inside SEOmoz's AdWords account, and saw some similar issues - they suggested a lot more stuff, but using broad queries wouldn't always show the exact phrases in order of popularity. You have to get specific, which is why refinement is key.
Gianluca's post below offers some good advice on this, too.
Thanks for the critique Sajeet - certainly a fair one.
Great comment Sajeet. 18 thumbs up and counting.
When I use the AdWords tool for keyword research, I always make sure to put in as many random and crazy keywords as I can think of to get a wide variety of return suggestions from google.
I think I should call this day a Keyword Research Day. Since morning I have been reading posts on Keyword Research just read https://searchengineland.com/the-keyword-research-rabbit-hole-110489 and what a coincidence that SEOmoz also has a post focused on the same topic.
In fact I think you cannot rely on any one Keyword Research tool as such because in the long run the search behavior keeps on changing and also varies from country to country . Keeping all options open for a global audience and targeting for the maximum possible keyword phrases should be the goal for the keyword research phase .
Instead on focusing on only the keywords suggested by keyword research tools and Adwords also focus on the possible psychologies of people from various geographic areas unless you are focusing only on a very specific region where only the lingo and jargon of that place matters for the search of that site. People from different places , of different age groups and of different educational backgrounds think differently and use different terminology and jargon.
Instead of focusing on only certain fixed key terms we should try to keep the possibilities of a wider set of permutations and combinations of keyphrases open for the site. The analytics data in fact can give a good idea of what searches are actually taking place country wise. A detailed discussion with the website owner who has a vast experience regarding his industry also helps.
Hence we usually ask for a list of possible terms he thinks his website should come forward when someone is searching on the web from the website owner, then we can add our research results and come up with the widest possible keyword list which can be targeted as per the content available on the site.
A little off tangent but I like the Term Extractor tool of SEOmoz which gives quite a good idea about the focus and direction of the site.
Do you not think this is would be a mine field and like shooting in the dark?
Would like to kow of the results when attempting to asert which keywords a searcher would use and trying to address this from a psycological perspective.
No, its putting everything together. The research done via Keyword research tools, discussions with the client to understand the jargon and terminology of the industry and also keep in mind the geographic target in mind where the suggest data is very useful.
For example if it is an edible oil refinery manufacturers and exporters site based in India but would like to cater to the global market also then Keywords like
If such a wide range of possibility is kept open for the site to rank then the potential of the site to cater to a wider audience is possible .
I think this is an incredibly insightful suggestion! To use your own example:
I (being from the South/Southwest United States) never use the term "edible oil" to describe the products mentioned. Therefore, I would NEVER search for that term in a "customer/buying" mode. I might search for it while in a learning mode (if I saw the term used somewhere, and wondered what it meant), but my queries would NOT lead to "convertable action." That means, once I researched enough to satisfy my curiousity on the denotation of the term, I would not take any action on landing pages, or sign up for newsletters, etc. I just wanted to know what somebody meant when they used that term.
HOWEVER, if I wanted to find info on "cooking oil" or "frying oil" for my "fry daddy" or turkey fryer at the deer lease, then I'm going to be searching in a "customer/buying" mode.
This is a wonderful point that the object of the search (term) may be the same, but lingo, and more importantly connotation-based terminology, will be very different. If you're trying to capture global traffic then you need to know what the object/term is "called" in various areas/regions to capture search intent!
Exactly, the basic thing is that the user does not see our keyword tag and search for our site he keys in what comes to his mind and the search engine bot takes that string of query and tries to hunt for the relevant site from the search engine index. So, making your website being available for a wider combination of queries is more beneficial. Like in the above case we want the website to rank if the person writes cooking oil, edible oil or vegetable oil whatever is the lingo and jargon used by the user.
Thanks for turning my day into a Keyword Research Day - that SEL article is also great! Thanks!
I think the idea of creating personas is a very powerful thing when it comes to keyword research. Not only are we discovering more variations of a keyword, but we're pinning it with the "intent" or the search.
"Homes for sale in Edmonton" = intent is likely to be that the person is shopping, and has some savings to put an investment down on a house
"Buying a home" = intent is likely that they are looking for information about buying a home (i.e. buy vs rent, or differences between buying a house vs condo)
The AdWords tool I find does a great job, but limited in its ability to suggest "variations" of a keyword. A brainstorming session before heading to the keyword tool has always helped me to identify what I already know, before asking the tool for ideas.
As also Sajeet wrote above, when I do keyword search I tend to use the logged into Adwords research tool.
Not only it makes faster the search (800 kw listed vs, 100), but - with a wise use of location and languages and other options - it offers a quite correct amount of keywords to study.
Personally I tend to start with with the "phrase" match option, as it is more reliable than broad match and give a more realistic idea of the volume of searches for the "query family" my keyword is part of; that way I can understand if a keyword is part of a group of larger keyword variations.
Then, I too use SEMrush to analyze my competitors' keywords, but, especially for a new site, I use it with the purpose to detect what low hanging fruits they are not targeting well but my site could, hence to start conquering those search queries.
Finally, Ubersuggest and Merge Words (+ a deep analysis of the keyword and queries shown In GWT and GA) are definitely a great help in refining the keyword search.
Let's say, keyword search and forecasting is painful, long and needs a lot of tweaks, therefore to rely only on one tool is not surely enough.
I love to combine tools, especially Google tools.
AdWords keyword tool + GA + GWT + ?? = GOLDMINE :-)
Oh and let's not forget a tool called Excel! I love it :-)
A long, painful, time-consuming process to be sure, but you've outlined a great methodology. My frustration is just that Google, in AdWords' public and private tools doesn't just give the full rundown of the most related keyword phrases in order of search volume without requiring this refinement. It also makes a lot of early entrants to the SEO field (and many, many bloggers/writers who rely on the public tool) choose the wrong phrasing/targeting, which sucks.
That sucks indeed. I can imagine that's annoying but web data is imperfect and will be imperfect for the next couple of years. The people who have the skills to understand and deal with this imperfect data will probably have more success than the ones who don't.
I like your post cause it can skill people to use multiple tools to base their decisions on.
Which is interesting, since they've been talking about "leveling the playing field" for people who aren't SEO experts. You'd think they'd be interested in making the easily accessible tools more effective, instead of giving the advantage to people who have the time and expertise to use these complex methods.
I wrote a YOUmoz a few months ago saying that you also have to be careful about word limits. Put in a 1-word phrase and you'll only be returned phrases up to 3-words long. So if you put in "seo," you won't get "seo search engine optimisation" (4 words), but if you put in "search engine optimisation," you will.
2-word phrases go up to 6 words, so "car insurance" will miss out "cheap car insurance for young female drivers," a potentially lucrative keyword here in the UK (especially if others have not noticed it!)
I'm in charge of keyword research at the agency where I work. Here's my usual tactic:
You might come across a lot of the same keywords more than once, but they can always be trimmed out. Better to find a keyword twice than not at all...
That usually covers most of it. If I'm unsure, I'll keep adding in terms until I'm happy I've found 99-100% of the keywords. But it's a job that should be done by someone who appreciates taking their time and going into detail - give it someone who's a keen 'rusher' and they might miss out huge chunks. And if you're an agency and the client picks up on any gaps, it could be hugely embarrassing.
Hi Rand,
Thanks for your post.
I can confirm exactly what you say: hidden terms in broad researches, but data are still there if you know exactly what to search for.
Probably the tool was heavily used by SEOs all over the world and this way their computing load has diminished sensibly.
Anyway, why use exact match queries? We all know that these numbers are not significant in themselves, but only related one to another. So, why limit the search to the exact queries instead of having an idea of all the potential of a combination of keywords including the searched one?
Ciaoo.
Gianpaolo
Very eye opening, now we need a blog post comparing all the other keyword research tools :)
I think this is a great example of a lower range keyword Rand, the fact of the matter is if you are only using the keyword tool for KW research you are limiting yourself you need to use various sources of data a did a post about it on SEOmoz a while ago - https://www.seomoz.org/ugc/advanced-seo-keyword-research-tips-and-ideas-14216
If you run the keyword tool on broad initially, it will give you a much broader range of terms which you can then rerun through on exact match to return traffic estimates.
Running on exact will limit the number of keywords suggested.
I don't think that's true actually. Running a few test searches right now, switching from broad to exact actually shows the same or a few more keyword ideas.
One thing to keep in mind is Google's built this tool for AdWords advertisers, not SEOs, and ultimately they have plenty of incentive to skew the results they're giving back to be more commercial and to emphasize more competitive terms that make them more money (they had a product update that did this pretty blatently a while back, then rolled it back, but the intent was there and it still makes sense for them to lean towards more competitive/commercial suggestions in general within the AdWords tool).
That said I like using suggest data - I think this is likely purer because the suggestions they're making are aimed at getting searchers to the right result as quickly as possible, so they're somewhat more likely to use the most popular terms based on actual search volume here (though it could make sense for them to show competitive suggestions with aggressive AdWords bids here first as well, obviously). So either just manually grabbing suggested and related searches from google.com or using a tool like ScrapeBox to automate grabbing lots of different suggestions and related terms, then running those through the AdWords tool to get a sense of relative volume tends to be a good way to get a much deeper list of keywords, in my experience.
AdWords keyword tool is great to kick off a campaign or group with. After starting a campaign the real keyword research gold can SEO keyword analysis & Search Query mining.
Like Tom I use suggest data too. When creating a post title, URL and meta title; if I bounce back and forth between Adwords and a tool like Ubersuggest I can usually create posts that are effective at bringing in long tail traffic pretty quickly.
Very informative article. Brings some excellent points to light
Hey Rand,
Thanks for posting on this. I changed my approach to queries with the Adwords tool late last year after poking around for a while trying to figure out why the terms in analytics didn't seem to match up with the terms exposed in original keyword research.
The new polling tool is kinda cool, but could do with a little more explanation...Initially found it a little confusing and had to stop and think carefully about my response. I'm still not 100% sure whether my thumb down is saying what you think it is :-)
So, to clarify, I thumbed down because I figured that was giving a "no" answer to the question you posed. Hope that was the right choice.
Sha
Hey Sha,
Like you I too get confused with the polling tool. I hope thumbing down means we agree that adwords keyword tool doesn't provide the exact match up terms in analytics.
-Hyderali
Hey hyderalis,
thumbs crossed ;)
Sha
Yeah - pretty sure that's right. Quipol is interesting, but not as robust as I was hoping for. Maybe they'll increase functionality to make for more complex/nuanced polls and additional features in the future. It's a pretty product to be sure.
Sha and Hyderalis, your votes are perfect as far as I can tell.
Sorry about the confusion, all!
I'm the guy who built Quipol, so I appreciate the feedback.
Sha and Hyderalis, you're voting correctly as far as I can tell :)
Rand, I'm working on it. Thank you for your patience and support!
If anyone has questions, you can reach me at max [at] quipol dot com.
Hey Max,
Welcome to the SEOmoz community!
Thanks for responding to my query too - much appreciated.
Congratulations also on coming up with a very cool tool :)
There's no doubt that having it tested here is a great opportunity to see how it is received and with more than 450 votes already it seems to be doing pretty well.
Sha
Thanks so much, Sha! Appreciate all of your kind words. It's always great getting feedback from a candid, thoughtful community.
So there is this problem in the 'KEYWORD IDEAS' portion of the adword keyword tool? right. So if someone is for example using the same tool to get the data around a keyword only rather than getting the keyword ideas from the tool, dont you think its safer for them?I would prefer to select keywords randomly on my general reserch rather than relying completely on the tool as it often misses important keywords with good traffic.
So the basic point is "NOT TO RELY ON THE 'KEYWORD IDEAS' PORTION OF GOOGLE KEYWORD TOOL."
Great article.
I still find it increadible the quantity of long tail keywords the sites I work with get.
The solution: since now we know that we have only 50% keywords for AdWords, we should cover the other 50% with help of other keyword suggestion tools. I would also suggest scraping the keywords your Top 10 competitors use, and expand that list. In general, the process should look like this: suggest keywords, remove irrelevant, suggest more, remove irrelevant, .... and so on and so forth until you are satisfied with the list you got.
Google is losing a lot of money by providing incomplete data. Exact match bidders are not getting these terms when using the tool.
Having LONG distrusted the Keyword Tool (to the point where I rarely even look at it anymore) I usually try to budget for broad campaigns in order to do real keyword discovery. And then every few weeks, I pull them into Excel and fish out the juicy gobbets. This becomes even more important now that (not provided) has taken over my analytics.
Unfortunately, I don't think the suggestion tool has anything to do with traffic or searches, but rather their semantics software.
I really like their kw tool, but I always crosscheck with my own operational searches and other tools, like spyfu.
One observation, since they have stopped showing kw search data, I think there has been a slight shift to where they are actually worried (why hide that data) and are on the defensive rather than innovative (offense) side, which is wall street speak for squeeze profits out before we cash out.
As an online marketing practitioner I am hedging resources towards "non-google" aspects of user and online traffic focus.
It's not good business for Google to give us the perfect set of keywords. Stop relying and be creative.
I've always relied empirically on the campaign traffic (like you suggest) rather than the keyword research tool which has always been paradoxically in various states of cripplement over the years. And I've been using Search Query reports since they were introduced in May of 2007 to reveal true user search terms rather than guess at them.
Google has more keyword knowledge than any other firm on the planet yet still doesn't make our life as easy as it could be in research (think "speculation" here, which G hates).
You can get more up to date hints I think just by letting the autocomplete feature suggest phrases to you, since these are then revealed as being in Google's lexicon. But of course that's slow and more manual with no volume hints (I'm sure there must be a tool out there to help).
But I have long held the belief that this is deliberate on Google's part. Why?
Look at the workflow they publicly show when you're getting started. It's this (note the order very carefully):
1. Open an account
2. Write an ad
3. Choose keywords
See that keyword selection comes last?
It's because you have to already have a landing page, an offer, and have described it (however well or not as the case may be) in your ads.
So you have to prove yourself to be a real life person or business with something to sell or promote. That's the most important thing to them. They then go and crawl your page from the ad url and many things are instantly revealed like
1. is your site in their index (no? alarm! microsite!)
2. is this page indexed (why not?)
3. are there any other links to this page (no? alarm! squeeze page!)
4. many others I'm sure ...
Related and appropriate keywords will come to us from users - and the real measure of the usefullness of a keyword (in combination with its ad) is whether it will convert (at profit) for us.
I write exhaustively about conversion on my blog at www.davidnrothwell.com
Thanks for listening!
Interesting read Rand. In all honesty when doing keyword research for myself and clients, luckily enough I never relied solely on Adwords. I always took into account analytics data too. I'm quite glad I've been doing this already as I never knew about this and may have missed some vital keyword data.
Cheers
I think Experian Hitwise is a pretty good tool for keyword research it gives keywords at ISP level not seach engine level.
Completely agree, and if run alongside a PPC campaign (so you can extrapolate %'s into absolutes) I find it gives more accurate 'visit' predictions than Adwords.
Also, be logged-in to any account, you get much richer, complete data.
Also, always use exact. Always use Local volume, not Global (for US sites).
The tool is only as good as the one who uses it.
Try to sort by relevant, not Global searches.
also master the finer nuances of the tool.
go quickly down the list and "star" (click the star icon) each term that is truly relevant and has high enough volume for you to care about. After 10/20/30, you'll see those keywords gather at the bottom right in the "Starred" rollup area. Now, go up above your keywords and click "More like these" + "Starred" and you'll get a better list. Keep "starring", keep "More Like this-ing" for a few cycles. Then and only then, sort by volume. Now you'll get a proper research list.
Use UberSuggest.org to get all of the related queries in Google Suggest and run those through the Adwords tool. ;)
The cynic in me finds it funny that, either way, to get the best experience, you need to:
a) have an AdWords account - it's commonly known that you only get a limited view in the keyword tool if you're not signed in.
b) have an AdWords campaign - as Rand points out, the big problem is the blind spot between the keyword data AdWords is showing and the data in your GA (that you're already getting traffic for), where running a broad, and costly, PPC campaign is the only solution.
Two 'conversions' for Google, eh?
Hello Rand,
Yes certainly you want to take it with a grain of salt at times and that is why using actual AdWords campaign data works out better. But I think your data is wrong, you are comparing it seems you are comparing your global web analytics data so it has to be against adwords global monthly searches. The figure always seems to be rounded up.
If you use the CTR data you should be getting around 18.2% of CTR at first posistion, so track that back would be around 1016 visitors a month for "blog traffic". This sits about halfway between Google (US) data and global data.
If you use the same CTR data you should be getting around 7.22% of CTR at first posistion, so track that back would be around 3573 visitors a month for "increase blog traffic". This shows you are underestimating this traffic potential for this keyword based on AdWords data.
Other points to note "improve blog traffic" and "increase blog traffic" share the same top 3 results and only one or two results seem to swap posistions. More times I think this has an impact on the data sets as Google seems to understand similar terms are close enough. This might explain why the traffic for that keyword based on CTR is twice what AdWords estimated.
One final point that I've seen that sqews that data is when there are no AdWords shown, as in these terms, I've found that matching my estimates back to real data with some slight adjustments/refinements can be around 5-10% off in most cases.
I don't think the title suits the article well. You're talking specifically about the AdWords keyword tool, which is known for (at least to me) not giving exact numbers or all keywords. You can use it for a rough estimation and for some ideas, but indeed don't focus blindly on the tool (at least not as the only source).
However, running an AdWords campaign to discover the search volumes and real keywords usage is ofcourse the best method to define keywords. Remember that buying broad keywords and checking the real keywords will only show keywords which lead to clicks. If you want to know exact search volumes, make sure you also buy in exact keywords (short/ long tail). However, also these numbers may vary, as this depends on certain factors. Keep an eye on the impression share for this, to get a clue if your ads are always being shown.
Unfortunately for us, there is no single tool which gives us all the data we need while doing keywords analysis. A good keyword analysis can only be achieved by combining the usage of different tools. However, that being said, I have been observing differences in the actual numbers in GA with respect to numbers shown by the AdWords Keyword Tool.
Inorder to do justice to your keyword analysis, it makes sense to refer to a combination of tools to ensure that we don't miss out on important keywords for the campaign. One important thing we should do is ensure that we do not miss out on Google Suggestions when typing for the keyword because those are the best and most relevant search terms. Also, WMT is a treasure chest of information which should not be missed out on. A couple of other tools include Soovle, SEMRush etc.
Great Post Rand. I actually use your Mozbar to help me with organic keyword research. I find that to be a very accurate way to access keyword difficulty for organic rankings. The serp overlay feature is great!
Hey Ajay Yadav i also face the same issue but thanks to Rand Fishkin for this super post...
I'm glad i found this beofre the weekend. I was going to use some of my GA data and Adwords to key in on some long tail queires. I prefer to use the exact match selection when looking for keywords as opposed to broad matches.
Really nice post. Didn't know such things could happen .Keep it up. :)
Great post that - obviously, when I look at the comments - articulates what many of us have been thinking about the AdWords tool.
I've always found it interesting that, the longer I run an AdWords ad group and keep adding more phrases, AdWords itself keeps suggesting even more (in the suggestions box to the right). "Where were you phrases when I really wanted you, at the beginning of this campaign?" I've wondered many times. I don't know why Google parses them out to us in little bites.
Like others, I'm constantly amazed at the variety of long-tail phrases that show up in site analytics; I add those in phrase- and exact-match to my AdWords lists.
Despite Google AdWords is not showing all the keywords, it's still the main tool of many webdesigners and SEO specialists. I'm using it a lot and don't see any alternative at the moment.
Shame that Market Samurai is giving all the troubles now.
Nice, yep, nothing beats sending traffic to your site at modified broad match, and having a gander at what the cat drags in. It's just one more example of how we have to buy our demographic research, more and more. Thanks for the post Rand.
Have you tried the search while you're signed in? I see the message on the first image "Sign in with your Adwords login information to see ful set of ideas for this search?" Does this make any difference?
May be lucky with clients sites I get to work on, however for most sites I find exact match search volumes in adwords keyword tool a very useful guide for eventual demand once hit Top 5 - 30 results.
For some keywords with low demand, 10 - 100 exact searches, even find clients where number of actual visits from each keyword exceeds googles figures :-)
Still a great rough guide to potential level of visits once ranked well, just remember it is a rough guide initially and tell clients will become clearer once ranks well.
Greetings All,
I am not new to SEO, but I am new to this commuinity. So please exuse me while I get the lay of the land.
Recently while doing research for content relating to Virginity, I did a keyword search for Virgin. There "0" results for that word.
Not sure what's up with that, so I did search for related terms and phrase:
virgin
i am a virgin
to be a vigin
i was a virgin
like a virgin
what is a virgin
when i was a virgin
Same results "0" for all the terms above.
Can anyone shed some light on these results?
If you are using Google Adwords Tools in determining what keywords you are using, you must choose "exact" search and not "broa"d and "phrase search". "Exact" search is the actual keyword enquiry by the people who is looking for the keyword.
Great points to remember for a PPC starter like me. Choosing proper keywords for a website is one the most important factor and this post will help a lot in making selection of good keywords for organic and paid marketing.
reall informative sharing, keep it up
Yeah in my job – in house SEO/copywriter for a company that sells products into the construction trades, typically a very traditional industry and slow to adopt technology! – I use AdWords as a very general springboard. A loose guide to start with.
Call me a renegade, but for the industry sectors at hand, I will occasionally completely ignore the fact that a keyword shows ‘0’ search volume. For a couple of reasons:
So we’re taking a bit of a gamble – my management as my ‘client’ are happy with it – on the gut-driven possibility that search volume data will grow for these terms as the kids in construction grow up and become bosses or purchasing officers, and there we’ll be, in a couple of years, sitting on first page with very little competition, and being found to be relevant and enticing enough to buy from.
I hope I’m right!
Anyone know how to search spanish keywords in Google Adwords? It usually says 0 searches but that is def not true. Thanks in advance to whoever knows.
Make sure you have your location and language settings right. Spanish keywords in the US can be underrepresented in the tool.
Also, if you're in a niche market (in terms of vertical, not geolocation), try a number of different locations to get local dialects and variations. I used to run sites in Mexico and Spain and in my vertical, some keywords were hugely different.
Thanks Rand, very informative post. I'll definately keep it in mind when researching my keywords.
Very spot-on analysis Rand - there's a definite pay-off for deeper research. And collecting/correlating KW data from your GA account (and Webmaster tools also) will help you develop clarity and lead to more aha! moments. I think there is a tendency to forget that the GKWT is intended for PPC - furthermore, I thinks it's in Google own self-interest to show primarily KWs that garner the most clicks (and higher CPCs) - not necessarily the most searches.
Another aspect in developing your KW list is assessing the competition - try out your list in a ranking program for your compeitiors - see where they are in the SERPs and calculate the cost/difficulty in ranking with them - their IBLs and domain trust (and once you're there with them how you'll get the click over them - page title/call to action/compelling meta description).
I too have noticed this. I would search for a head term, expecting to see keyword suggestions that are highly prevalent in the industry I work in, and they wouldn't show up. At first I thought I was simply missing them, thinking that I got lost in 800 results, but after a few times of this happening, I came to the conclusion that they were missing from the suggestions.
So now I try to do several searches so that I don't miss anything.
Pretty interesting stuff, thanks Rand. Now that I know this is going on, I actually wonder why I didn't think of it myself as it makes complete commercial sense for Google.
They want PPC customers to use broad match campaigns as discovery tools. If you think there is extra traffic to be had around "blog traffic" queries and you have some spare cash laying about, you are bound to have blog traffic on broad match to get fresh exact match ideas out of the search query report. This way, Google get paid for helping you with your research. So of course they are not going to show all the keywords in the keyword tool and kill this (rather substantial) revenue stream.
Hi Rand! Great post.
Imagine, we have typed exact phrase "blog traffic" in Google and get result page with the following message: "About 56,000,000 results (0.14 seconds)"
The same phrase "blog traffic" we searched in Google's AdWords Tool also performing [exact match] and got 1,500 Global Monthly Searches.
I would test my chances by calculating old fashion number KEI (Keyword Effectiveness Index):
1,500 ^ 2 / 56,000,000 = 0.04.
So, number KEI for exact phrase "blog traffic" is 0.04. Sorry, but I would not start optimization for this phrase - the competition is too high. In this situation AdWords Tool can be used as a generator of possible variations, even if they look doubtful. We can check it out in Google.
I am egerly waiting for comment on this
This one is really nice. Exactly what I searched for... and guess what? - "adwords keyword tool" ... lol
I can't even get any exact match or phrase match terms to show anything in my Google Adwords keyword tool suggestions. How did you do that, Rand?
Good tips!
I have always seen the difference in "possible traffic numbers" and the actual that top ranking sites are able to achieve.
My hunch has been that Google want people to desire more of the traffic and therefore turn to Adwords to increase their relative market share.
Other than than cynical hunch, Google probably just wants to provide some and not all the data that they have for free eg both keywords and accurate traffic numbers.
Andy
oh
you shock me with what u r saying
i bought many doman names depending on goodle adwords information
oh my goodness
how to be sure now?
for exampel i just bought this domain name egypt last minute cz of google ad word
is it worth buying?
Hi Rand,
I usually diversify when I am looking for the right keywords. I use all popular tools and also, I do some of my tests to assess and get the best options. Right now, with all the changes in the search algorithms, I’m continuously doing a lot of variations. Google made a lot of changes in a short period of time so I expect there would be more changes.
Great job, that is an excellent piece of work written for all who desire to know better
While I will still use the adwords keyword tool for basic keyword research, I also use 2 different tools that I purchased some time ago. Between all of these tools, I can find good data for various niches, but I am still perplexed as to the lag between actual results that I see with Google analytics, my own onsite analysis information and the adwords keyword tool.
I do agree however from the article that you should start general and work your way to a more specific key phrase. The time that one spends on keyword or key phrase research the more time that will be saved down the road, but that only works if you have a bit of an idea about how to conduct this research in the first place and are able to uncover those "nuggets" out there that many are already seeking.
Thank you so much for the post!
OK, i added broad keyword in AdWords, but I see only 1 keyword in Google Analytics which shows thousands of impressions and some clicks. Where do i see keyword terms/phrases which i get impressions for?
As always, when I am looking for expert experience in the world of SEO, SEOMoz is right there. I literally ran into this exact situation where I couldn't quite understand why some really obvious keyword phrases weren't automatically showing up in the keyword ideas. I love the idea of using GA to add additional value and had thought of it. Because it would require going back to my client to get access I hesitated to do so for fear it would make me look incompetent. So glad I'm not the only one who had this issue! Great post and so appreciated!
I also realized that some of the keywords didn't match with the Google Adwords Keyword Research Tool results. My blog is on the first page on Google for the keyword "Disable USB Ports" which search query size is 18000 times per month. When I checked my Google Analytics it shows me just few visits. I can't explain that. My blog posts are always reviewed by English Native speaker that eliminate the possibility of grammatical mistakes.
Reading your article brings up some questions in my mind regarding actual AdWords result / value. I have heard that a #1 search result in Google yields a max of 42% clicks for total search traffic for a keyword. That obviously leaves around 58% of the total pie floating around.
My main question is where can we find some general rules of where this 58% splits out? For example, will the #1 AdWords slot end up getting something like 30 or 40% of that remaining 58%? Will an AdWords slot #2 get something like 10%? What will right column AdWords results get in traffic? And finally, what kind of traffic can the #2+ search results likely see when they're above the fold?
Thanks again for getting me thinking.
This post has just made my day thank you so so much i will be relaunching my site https://www.greenlandng.com and this should help me to improve my search result and the hunt for long tail keyword search. Thanks so so much.
If my keyword is “reading books” and you would like to ban the ad from people looking for downloadable books, how would keyowrd be submitted to AdWords?
I can't believe i may have been missing out of great opportunities for relying solely on Google keyword tool. This is indeed an awesome post as i cant wait to search for all missed opportunities.
thanks for the post
Can you explain this part?
Use your own analytics as a guide to find new terms/phrases you might be imperfectly targeting.
OK got it
And if you see keyword variations that have a unique or different intent, it might even pay to create a more targeted page for that query,
OK got it
and you often need less work to rank, since Google uses the "indented results" system to drop a second URL from the same domain directly underneath the first one on a given page.
could you explain this part in more detail?I don't understand
thank you!
That is exactly what I was wondering the other day when setting up a client's campaign. The client had his list of keywords that he thought would be good for his business and when searching through the Keyword Tool, I didn't see them. I entered them as exact terms and there they were. You hit this one on the head! Great post!
i like point number one in this article. "realize your going to lose money" but there is still a lot you can learn from adwords
especially now since there is (not provided)
Great article with some very insightful info. I plan to adjust my use of the PPC tool accordingly.
I agree and great analytic guy, just simple reason to answer this.
1. In our analytic, we can't see specific what keyword that bring visitor to our site, this is google secret and they keep that. 2. Almost people, blogger and internet marketer in the word use this tool and this trick. 3. The logic is google sell ads and more click by visitor they get money from advertiser, of course google dont want to show this keyword in serp but only in ads.
Hello Rand, and all...
Thank you for sharing this post Rand. I noticed much of the same as you have mentioned as well. The Google Adwords Keyword tool can produce many varying results if you know how to tweak the outcome. Using the filter feature can help or hinder (and you have to remember to turn it off between searches), using the "Only show ideas closely related to my search terms" can also help or hinder -- depending on the initial search.
These are just a couple of examples too. There are a handful of filters available to us and we just don't see the power of everything we need if we don't know how to use them.
That said, one of the absolute most POWERFUL methods to the madness that I have found invaluable to every keyword search I do - is to add the ~ (tilde) symbol modifier before every search phrase I look up. This seems to pull up the most related keywords possible to a phrase. *Even moreso than using the "Only show ideas closely related to my search terms" feature.
This is just one of my many keyword research strategies but it's by far and wide the most used on my part as well.
For those NOT using an AdWords account, if you add a wildcard operator between words in your phrase, you can get some more diverse results sometimes. For example, if you search for "blog * traffic" you get a few more of the results that weren't showing up in Rand's example. However, if you are logged into AdWords, it doesn't seem to work the same way. You actually end up getting fewer results (351 vs. 600) when you add the operator. Interesting.
Great post. I see this often as well. As a solution I love researching in Ubber Suggest, taking that keyword list and pasting it back into the AdWords tool. Google will give you search volume for these terms, though Google originally said these terms didn't exist.
I have also noticed that the AdWords tool often limits brand keyword research, specifically for larger brand names. Rather than giving us everything under the sun, the initial list it gives back is high-level and general in nature (missing hundreds of keywords with moderate search volume).
new lesson with me. i will attention that
I usually use keyword research tools like "keyword snatcher" and "keyword optimizer pro" to collect all possible keyword ideas for a particular keyword and then I will put them on google keyword tool to check their search volume datas...
This is the very powerful way for me to do keyword research. what you say guys?
Helpful tip what do you recommend for keyword research besides adwords
I use adwords to get a sense of the overall amount of searches...dont take it as a precise amount though...but I like to use the SERP overlay feature on the Mozbar to compare competitors and measure difficulty for organic rankings
Interesting observations. Did you notice that all the keywords that Google had hidden were all high competition keywords. Any relation to PPC revenue?
great post
Great article. I have noticed this quite a few times, however just thought I wasn't being thorough enough, however realistically it should work how it is expected so that we can focus on the words that matter.
thanks Rand.
Hey there! I love this blog and all of SEOmoz's social channels. Just wanted to share that there is a small typo in the article: "reserach."
As a PPC agency, one of our first new client moves is to request Analytics access. In addition to improving our keyword exploration, the landing pages and conversion rates help in determining hierarchy and bidding schemes. Analytics isn't the only source to use, but definitely should be among the prime ones used.
Excellent post, Rand. We've also noticed that searches with local emphasis show no traffic, whereas Google Analytics DOES show traffic for the same keywords. That seems related to your findings. What do you think?
Thanks for the help!
Scott
Rand, thanks for the case study, I have been doing this for years and not knowing it, let me explain. I thought why spend so many hours of keyword research when I can do a little and let my audience tell me what is best. Again you will spend more at the begining but I always stay on budget and it takes about a month tops to fine tune.
I set up campaigns and adgroups and take 4-6 good keywords for each adgroup with 1 broad keyword for that group that is as broad and relevant as you can have. That one keyword I refer to as my breader keyword the rest are all set to exact which will cost less for my add to show then the broad match version shich we all know. This breader phrase will do three things for me.
( I run query reports twice a week for each adgroup to find the following and trim spendind quicker)
1. New relevant keywords that I add as exact match.
2. New Neg-Keywords for the adgroup level
3. New adgroups I can add with more specific landing pages.
The process takes about 4 weeks to fine tune where I have impr. shares of 95% + and a lower CPC my quality scores rise to no less than a 7 most being 9 and 10 again saving money. at the end of 4 weeks I will leave the broad match term but also add it as an exact. The broad match I like to leave for a year to see if there is also seasonal spikes for my phrases.
Again I never knew of the issue you have shown above but good to know my method of campaign set up is handling it.
thanks again
Kevin
Thanks for the post,
Actually I've experienced this problem by myself during my last kws research. They're hiding results in the suggestions and to avoid the problem I've tryed to get as many kws as I could by editing different kws and then get the new ones and only at the end check the exact match traffic in order to make a selection of them.
pretty interesting, I always just use Adwords as a "general guide" it's hard to know how accurate it is sometimes
Nice post - I've been noticing this for years and compensating for it by being more specific - it would be interesting if we could work out how Google was determining what results to show in the keyword tool.
I think adding specific keywords comes in handy once you've reached the point of grouping your broad keywords together for content mapping on your site. It is definitely time consuming, but at least there is a solution to uncover those very niche keywords/phrases.
Interestingly, there is a pattern in the missing keywords in your example, Rand. It looks as though the algortithm doesn't recognise the phrases 'to your' & 'to my' as simliar, and hasn't suggested any variations of these. There is a case here for a bigger investigation into phrase patterns that are/aren't matched by the AdWords keyword tool.
Perhaps the key takeaway here is to 'ensure that you have researched a few modified versions of the phrase'
Another fantastic post with great documentation.
Perhaps this is an area where SEOmoz can use your crawl data to build a better tool! Us pro account users would love you for it!!!
I was always a little suspect of adwords keyword tool. Mainly due to my theory of Google wanting to make $$$ (as any company does) with ad campaigns. To do this I always thought they provide misleading data to potential advertisers (not horribly misleading but I think you get where I am going)..
Does anyone have some quick thoughts on WordTracker accuracy? I love the new beta version but the only downfall at times is being unable to find data for more specific keywords/phrases.
Anyhow thus far I would count on WordTracker vs adwords but do still resort to adwords as needed but do take the information as not 100% accurate as Rand has mentioned.
Thanks for the post Rand! I kinda noticed that with Adwords some time ago :)
Hey Rand,
Nice Post.
What I do, aside of using other tools, is to put my primary keyword(s) in Adwords Keyword Tool, Select a few of desired keywords, download these and put all keywords back in Adwords Tool and check "include closely related to my search terms". It gives me a better set of keywords than the original ideas.
One can also use spy tools for organic search terms of competitors.
For smaller search markets (google.ie) the adwords exact keyword numbers can be way off, in both directions, saying there are far too much, or far too little compare to actual numbers.
Interesting. I have thought for a long while that the search numbers adwords is reporting is more less inaccurate. Just think about it.... if they suggest every keyword out there. They will not make as much per click. They make the money by companies bidding more on clicks for the keywords they suggest. Get a bunch of compaines that want the the same word. Jackpot.
Hi Rand,
I have also been experiencing this with the "new" keyword research tool; however, I want to note that I've only noticed it as a problem since Google updated the tool about a year ago (or so). I agree, you have to manually enter these keywords to actually yield the data below. Is something you also noticed as being connected to the "new" keyword tool, or do you suspect this was always the case?
Thanks,
Justin
a very good post
I am curious what data we could pull using Google's BigQuery service (https://developers.google.com/bigquery/) or by using Python also provided by Google (https://googlewebmastercentral.blogspot.com/2011/12/download-search-queries-data-using.html)
anyone tried using these services?
unfortunately I haven't yet
Oh yes, and get ready for a ton of frustration. There is lots of data there but the hoops to jump through are not nice. I have two developers working on it and they're frustrated!!! Let us know if you figure it out.
Yes, unfortunately this has been happening for a long time. And there are times when even if you enter the keyphrase manually, they won't show you any search volume data, but then after a few days it comes back. Logged in or not.
Thanks for the post Rand
It was short, but not to sweet, Very informative and really an eye opener, I have however wondered why what adwords and analytics show me and whats on my site doesnt always match up. Thanks for the 4 tips we can follow to get this problem sorted.
Thumbs up -.-
Thank you for poisting this explenation, however I'm missing a very good (IMO) suggestion about that same issue, what about going into one of your Adwords campaigns and using the "See search terms" option? There it gives you many more ideas for phrases that people actually searched for, and not only an estimation. Also - nothing is hidden there, which is nice.
P.s - I'd love to see the answer to @Sajeet above. :)
It does give many more ideas, but often has the same fundamental problem that a broad query won't always show the closely related/matching exact phrases in order of volume (or even at all) - you have to refine and get more specific to see everything.
It’s not from a day or two but its been for a while that Google Adward tool does not show the complete data and I tried matching this up few months back with my analytics and I noticed that suggestions of keywords are very limited.
Using multiple keyword research tools can help you verify your data and allows you to suggest more possible keywords that can help you with conversions... It’s also important not to stuck up to some fixed keywords and use the verity of keywords and combinations to increase the list of suggested keywords and see what is important for your keeping your target and audience in mind.
Over all a great article by Rand! *as usual*
This has definitely been my experience with the AdWords tool as well.
A different sort of illustration of how inexact/aggregated AdWords queries can be seen if you search for certain singular/plural pairs on broad match. Typically, broad match will allow a singular or plural to match its plural or singular counterpart. As a result, these pairs typically have the same or very similar search volumes. If you are trying to determine whether the singular or the plural gets searched the most, don't do this:
https://cache.gyazo.com/7985685eee23279e3c76e3e36412d227.png
I agree regarding the Suggest data.
I too give importance to the suggest data.
Hello Rand Sir,
Very true, but I have a doubt in my mind that you are showing us only 1 site visits of the keyword of "blog traffic" but I think google keyword tool show the data of all sites have this keyword "blog traffic" like- google 1st page SERP....
If its not true please explain me, thanks....
I'm not sure exactly what you're asking, but yes, I'm only showing visits to SEOmoz's page, which ranks highly, but certainly isn't getting all of the clicks for these searches. In fact, I suspect that in many of the cases shown above, there's more search volume overall than what G's reporting.
I want to say, you are showing the visits to SEOmoz's page that has no. 1 rank on Google SERP but another 9 Google SERP pages also got some visits.....
Like-
Keyword Rank Site Name Visits
1 www.seomoz.org/blog/........... 185
2 www.blogtrafficexchange.com/ .....
3 www.increaseblogtraffic.net/ .....
Number 2 and 3 would also got some visits, so all sites till Number 10 would have been getting some visits. If will count over all visit of first page of google it would be approx 500 to 700 (not sure exactly)
I think PPC is the best way to find the clicks of any keyword..
Sir, i want to know there is any difference between visits and clicks if yes than explain me
A good point, you've brought up here. The adwords keyword-tool wasn't made for SEO-people, but to get people advertising via Adwords. In order to get a complete overview of your relevant keyphrases, you'll need to throw in some logic and knowledge about the market. Which keywords do the sales-people come up with, etc.
And of course, there are more tools available on the internet, as is mentioned in a number of posts here. If it were too simple, a lot of us SEOs would be out of work;-)
Hi Rand, I have found this to be the case too. In fact, I had researched keywords so much and for so long that when I started using the Google Adwords Tool (about 6 years ago) I immediately noticed the difference. Now don’t quote me on this, maybe I started using it 8 years ago, but either way, prior to that I was using Wordtracker. I found Wordtracker to have a much more complete list with much better and more accurate suggestions. Again, I haven’t used Wordtracker in a while, but when I was it was pulling its results from Overture (Now Yahoo Sponsored), 7Search, and a few other sources. The Wordtracker Keyword Research Tool and the Overture Keyword Suggestion Tool were certainly the best and most accurate. 7Search still has a Keyword Suggestion Tool it is located here >> www.7search.com/KST/. I have already performed the search for “Blog Traffic” and “SEO Moz” and received very little traffic results or none. The search for “Blog Traffic” produced results for:
increase blog traffic – yes you had this,
blog traffic - yes you had this,
how to increase traffic to your blog - yes you had this,
comedy traffic central directing cop blog – NA,
traffic to your blog – NEW You didn’t have this,
weather wsb mellish traffic blog news – NA,
blog traffic marketing blogs adsense- NEW you didn’t have this,
Now with the “Blog Traffic” search results you have to keep in mind that the projected traffic numbers are going to be lower because of the reduced traffic that 7Search receives vs. Google Adwords. Now I imagine you can extrapolate that out if you want to take the time, but the important take away here is that they had a few suggestions that Google didn’t. And the even funnier thing is when I used their exact match tool and typed in some of the words on your list they actually had them in their database too.
Sorry, the search for SEO Moz produced no results :-(, so it is certainly not the best reference, but doesn’t hurt for addition al suggestions.
Also, I remember a long time ago Google stating that this was an incomplete list of terms and essentially merely a sample. I noticed that you access the Google Keyword Suggestion Tool from outside NOT being logged into your account. Google states on that page in the Q&A:
Why do search traffic statistics vary in the Keyword Tool?
Search Traffic Variations
The Keyword Tool shows search data from Google search. Data from the tool may vary due to the ad group you've selected, or whether you've accessed the tool from outside of your account.
I performed your search for “Blog Traffic” when logged in and received 552 results vs the 100 results you receive when performing the search logged out. When logged in I saw many of the words you displayed, but there were still a few I didn’t find. But I bet they rotate through different sets of keywords.
Word Tracker is surely a good tool... the shame is that it is (almost) totally unuseless for any language but English.
I've always assumed that adwords won't necessarily give up all the good, high traffic keywords related to you initial terms. Starting broad and becoming increasingly more specific works for me.
However, I don't think you can just use the phrases you're given if you're going to achieve the best results.
You've got to bring something else into the mix.
The beauty of mixing in a nice dose of instinct, creativity and external (customer) research is that may you'll find those high relevancy, good traffic keywords that your competition aren't targeting.
If adwords is suggesting the same list of related keywords to everyone trying to research your root term then my suspicion is that the competition will be looking at those too.
As others have mentioned, looking at your GA, GWT and adwords data, looking for patterns will get you started. Running ads can help supplement this, especially if you don't have the data from GWT/GA!
I also try to think about who I'm trying to target, what are their goals/intent etc. After all, traffic is one thing, traffic that converts is another!
I use competitor keywords as well to get a wide wariety in Google adwords. Sometimes simply by looking at the keywords main competitors are optimizing for is a good start. Of course doing your own research is a must.
hey smart people,
I second you all and it’s not from today. I am experiencing it for many last months but what’s the catch? Add "probability" in keyword selection. Good enough, Google "Suggested" the keyword with all the best means Google can but why don't we think if Google showed us their cards then what’s the bet? Google is for Google not for us :)
SEO is getting more and more versatile as we now taking care of customer behaviors and what not and what not. So we should add the element of Probability for the keyword selection.
Interesting post. I 've just written a post on Google SERPS showing Questions without any answers whatsoever ranking 1-3 on Google but not sure if i'll publish it. Would Google penalize me if i bag them?
I also have been suspicious of the adwords tool for a while. While it does help some, it is pretty frustrating that there is so much mystery there. I guess Google doesn't want to make anything too easy for anyone. (Except maybe searchers).
Nice post Rand. I usually start off in Adwords then take those suggested terms and plug them into ubersuggest, then go back to Adwords with that list, which works very well for most campaigns.
Great point to make. I think the Google keyword tool should just be used to come up with keyword modifiers. Then, we can use concatenation to make up our own keywords. The results from concatenation can then either be plugged back into the keyword tool to repeat the process (as many times as necessary) or utilized as the target keyword list list.
I wrote a script for generating the phrase, broad and exact match as well as optionally adding in all possible broad match modifier combinations for a set of key words/phrases:
https://gist.github.com/1840016
I added instructions to use it as a comment on the gist (because single line breaks don't work in this comment box)
Thanks Rand for this alarm like post. Seriously I got more shocked when I saw for my own set of Hotel Keywords. I searched for those keywords in Google Adwords tool and then in My GA account, and the suggestions are much more lesser compared to the versatile set of keywords shown in GA for their Traffic! Instead Google adwords is showing some other keywords assosciated with my Hotel related keywords(that's not the issue) but as you had said , it's not showing the major ones approx. 20% keywords are not shown in adwords tool compared to that of GA report! I wish I would get some sort of solution about it from SEOmoz. Thanks!!
Hi Ajay, my advice is to not only rely on Keyword Tool data use various sources, I made a post about this on the SEOmoz blog a while ago - https://www.seomoz.org/ugc/advanced-seo-keyword-research-tips-and-ideas-14216
Hope it gives you some additional ideas for the hotel keywords mate.
Thanks James.. for your feedback.. By the way.. I have been using your suggested tools since the day I communicated with you for your post https://www.seomoz.org/ugc/advanced-seo-keyword-research-tips-and-ideas-14216 and believe me It is much useful for me even at this time. the only thing for which I regret is that Even Google's own keyword tool would show different results than that of Google Analytics(Never thought Google's own two products will generate different yields for the same topic) ! James your Post is truly valuable for me and everyone!!! You had said in your last Post that you have more information about it on your table and I am curiously waiting for it!
this is a perfect post you made James.... I love it! great job!
The "new improved" adword tool rolled out about a year ago is MUCH poorer at suggesting related terms in my experience. I did not see any talk about it at the time and thought maybe it was just me. My question is, Did Google decrease the usefulness of the suggestion feature on purpose or was it a bad implementation? Does anyone know?
Hi Rand,
This is the first time im commenting on your blog, Your post really got me to think & research. Thumbs up from me as Google Keyword tool was never meant to be precise when it comes to the Keyword Tool https://support.google.com/adwords/bin/answer.py?hl=en-GB&answer=25148 However this tool only gives us a sort of basic idea on where to get started. We neet to put a bit more effort to find more precise data.
About your above experiment which I also had a try and found the below.
The keywords which were not suggested by Google to you was suggested to me when I used keyword "blog traffic" and used https://www.seomoz.org/blog/21-tactics-to-increase-blog-traffic-2012 on the website.
how to get traffic to your blog
how to increase traffic to your blog
how to get more traffic to your blog
how to get traffic to my blog
Inorder to get precise Keyword ideas we could do the following
Competitors URL + keywords = More precise Keyword suggestions
Thanks! i think this is an amzing tactic :-)
Thanks! Adriaanb
Hi Rand,
Another thing one can do is use the relates search terms at the bottom of a Google search and then copy and paste these into Adwords keyword tool and select exact.
I am sure they do this to prevent SEOs from getting all the correct data from the start and then blasting these keywords into their copywriting schedules.
Anyways nice post I also did same test and u are right spot on.
This has been an irritation of mine for quite some time, and I've been using some of the tips above - spot on, Rand!
Gavin
Thanks a lot Rand for such a wonderful information that really helps to enhance our knowledge & know more briefly about the google adwords for keyword analysis. I totally agree with your article because we know that keywords plays a most crucial role & without the proper selection of keywords we can't achieve our goal.Keywords should be taken in such a way that they clearly define or covers the selected area from the niche that they are associated. The target keywords should be specific according to our requirement.
Tools that can not exactly be, we should only refer, thank the author. meaningful posts
What tool do you suggest using as an alternative?
Assumption:
- How to get traffic to your blog = 31 characters- Will not fit ad copy title. right?
Maybe that's the reason why Google's not suggesting this "long tail" keyword in the AdWords Keyword Tool... Right?
Thanks Rand. This is great advice. I'm going to give it a test drive and see what omissions I can correct. I find keywording complicated but with SEOMOZ I'm getting my head around the best way to beat it. Thanks heaps.
I've noticed some weird issues with Google Adwords keyword tool traffic estimator as well. For example - I was doing a search a few weeks ago for a heavily competitive term, and Google showed traffic for one term, but showed absolutely no traffic for another very similarly-related term with the exact same competition level. I think Google's hiding a bit of data from us that they don't want us to completely know about - especially when it comes to the high converting traffic driving, real money keywords. It's only a hunch, as I have no concrete proof of this, but it got me thinking along those lines seeing that kind of disparaging data on keywords that were otherwise 100% identical in every respect (competition level, etc.) The only thing different was that Google wasn't showing the estimated traffic for one, but was showing the estimated traffic for the other, even though their competition level of the keywords were totally identical. Something featuring these keywords in such a saturated market is bound to have very similar traffic levels, since there are many many other people in the market going after the same exact keyword variations. Something to think about.
Thanks for bringing your expertise to this topic.
With the literature citing the Adwords as essential to keyword research I was starting to wonder if I was missing something. So I analyzed pre- and post-campaign keyword and traffic data on a bunch of sites and found way to much variability; then when I analyzed traffic and SERPs for other keywords (random and similar language based words) I really got suspicious but, where to go with it??
Your tips are perfect and timely as is your "important lesson". Aha, this perspective and direction is what we need.
Reading the comments, I now get the "why" too!
Sounds like many of us are struggling with the same issues. Maybe, the "targeted opportunities" as you cite can come from the actual language of the users used for searching/communicate. Our analyses supports the comments by people above referring to the audience factors, semantic language or psychologies. Based on your post and the discussion, I am going to keep the Adwords, but look at it differently and add as much understanding of language as we can from other research.
Great discussion and gives me some ideas on more things to try. Better get back to the spreadsheet :)
Thanks sir. I was confused on this always. Your post made me some clear. Yeah, it happens.
I have observed another point; i am confused on it.
I also use to take the suggested keywords which comes at the time of typing on search box(the instant suggestions).
I believe that when a visitor types the querry, he must click the suggested keywords immediatly rather than what he had come for. But i wonder when i check those kws to find out the search volume from adword tool, there is no data or very less data for those.
So is it wise to take the suggested keyword which is shown at the time of typing?
yep, take adwords keyword research tool with a grain of salt—this is where the "art" in online marketing comes in.
Also, if you look at the semantics of the phrases you desire, the ones that don't show initially (yes/no) all have something in common - "get" or "drive" in the phrase. That's because Google starts with "relevant" matches to your query, and you have it sorted by Global volume. Initially, it doesn't consider "get" and"drive" phrase as relevant.
That's why you need to grow you list (usng Stars and More Like This). You'll eventually pull in those "get" and "drive" phrases.
Anyone who only uses the 1st list presented will naturally be dissatisfied and go looking for answers elsewhere. But in the larger scheme, properly knowing and using the tool will yield all you need to know and save on time in the long run.
Nice post. I've suspected this for a while as once you've had a running a campaign for some time, you will find that the "search terms" report offers better kw suggestions than the actual keyword tool does.
Plus, Google can't give everything away or we'd figure it all out and everybody would be gaming the system.
This is probably one of the best and most informative posts I've read in while regarding keyword research. I for one have suspected this from the AdWords tool, but never have tested my concerns.
Thanks for sharing this Rand.
Great article Rand. I still use an old school brainstorming list for keywords and the drop them in adwords tool. After making some choices I follow analytics to see where I'm getting traction and adjust accordingly.
There is no doubt that the tool use to be much better than it is now for suggesting keywords.
Thanks for the post. I do agree that Keyword tools doesn't give you everything but it still a great tool. But in my personnal experience the best keywords aren't alway the most competive ones ( the one everyone looks in Keyword tools ) so I do use Google Keyword tools, but I also do a lot of brainstorming to find those who aren't suggested, related one, and everytime I do find some good one by thinking : What would I search to find this...
I also do Adwords Campaigns to built data and learn from it by looking a which exact queries brought the good visits and refine my campaigns by adding those queries has exact match and limit the bad one by adding Negative Keywords to block my ads from those.
Keywords research is Art & Science at same time.
Yeah, having a dummy adwords campaign is a great way of testing out CTR also for the keywords you rank on organically as well.
I like how you mentioned "Keywords research is Art & Science at same time" as my company is called Art Science Web :)
Hi Justin,
What a coincidence ! I like you're company name then, it's a great description about what our work is all about.
Sure adwords isn't exact nor does it offer you every single option for your keywords but its numbers are close enough to the amount of traffic you should expect. As for not getting great keyword recommendations, I find that to be a pretty good thing in my opinion because your competitors won't be able to find your keywords using adwords if you're the only one that knows of them.
there are many things that I don't like about Google Adwords, but I have learned to use its keyword tool quite effectively for keyword expansion for building traffic and profitibility (conversions).
My way of doing things has always been tracking how keywords are performing in terms of opting in or just pure traffic combine with time on site - its a benchmark that I always shoot for.
I guess what I'm getting at is, there are some mis leading numbers when you're looking up search volumes in adwords.
But, if you are anal about the keywords that you're going after and tracking it. Then in my mind you're taking a pulse of your current market and with the knowledge of what keywords are working for your site (traffic + conversions) - then these kw are worthy to go after on the organic side.
We noticed both the drop out for (not provided) and unlisted terms a little while ago, but realised the way to get an understanding of volume from possible new terms what to run a comparitive search on a term with a known volume at the same time. This then gives a quick comparable reference to the possible volume for each new term.
i can luangh on all of u...
this is something is well know to all but no one express it as everybidy thought google is god...
Rand really u did great job...now actually seo executive start hating google ...
nice post.. keep sharing
This tells me one thing. Google is evil.
Well, I guess, if you had to peform the task of finding needle in a haystack of nearly 20TB index, 200 million times a day, even you would be like this, its hard to focus on everything and make everything perfect. but thanks for Rand for pointing out the issue.
I think it was better before but it doesn't give perfect result any more.
I had to wonder how many people just read "thumbs down for Google Adwords" and followed instructions. And then I laughed.
Anyway, I don't think AdWrods has ever given perfect results. It's a good place to start. Here's one reason you might have seen the effectiveness decline: everyone uses it now, and everyone looks at what everyone else is doing. The result of everyone using the same tool in the same way is that no one gains any advantage in terms of keyword targeting.
Then what's the problem the with the first comment of this blog https://www.seomoz.org/blog/the-inside-scoop-to-finding-linkbuilding-opportunities-with-free-alerts ?