Hey there folks! Today's post is a hands-on walkthrough of some of the decision making used when choosing the keyphrases to target. Producing a list of the most important terms in an industry is nice, but actually choosing the right keyphrases is essential. The post was largely created in response to a question submitted by Kien in the comments of my last post.
What to Expect
This post should provide you with real-life examples of the keyword decision making process and help you make sense of the output from the revamped Keyword Difficulty Tool. If you're already a hardcore keyword research (and keyword difficulty) guru, this is more of a refresher, but should provide valuable insights to journeymen and perhaps a bit more transparency into how to choose the right keyphrases for your site (plus a bit of a tour for those of you who haven't used the Keyword Difficulty Tool for a while).
While you may get more value from this on smaller sites, and those that are newly launched, large sites with specific keyword targets may benefit, too.
Two Different Camps
There are really two separate and distinct camps on keyphrase research implementation: those who always go for the highest volume search terms that are moderately relevant to a page and those that also give consideration to the competitive landscape for a certain term. Regardless of which camp you fall into, this tool can be immensely useful. If you find yourself in the "competitive" camp and always go after the highest volume term no matter what, the Keyword Difficulty Tool could and should still be used to track your linkbuilding efforts vs. the competition and to have some understanding of if/when you might be able to rank for the term in question. There's no point in setting goals (even if they are lofty) without having some idea of how to reach them.
Image: Ebaums World
However, if you are in the camp that likes quick results and has a bit more time to give to the keyphrase research process I strongly recommend going the extra mile to indentify realistic targets for the short-term and keep those lofty ambitions in the back of your mind to be addressed as a site grows, gains new links and hopefully acheives higher authority and trust metrics.
But what if my site is brand new?
As a general rule any newer sites should probably be aiming for the "less-moderately competitive" types of keyphrases for the most part. The one exception to this rule would be for exact match domain names.
If you can afford to look at these metrics in this light you can scale up the competitiveness for your terms as you improve the quality and competitiveness of your website.
Image from: Rand's Post on Multiple vs. Singular Keywords
Thinking about it logically (with the above graphic in mind): if you could rank first for a term with 30,000 global monthly searches or somewhere on the fourth or fifth page for a similar term with 300,000 global monthly searches which would you choose? The optimist might choose the second option, but the truly intelligent will pick the first for now and aim for the second option down the road. It doesn't take superior math skills to figure out that a small fraction of 1.2% of clicks from the 300,000 searches will not amount to anywhere near 42.1% of 30,000 searches.
I'm not saying it's always better to play it safe, I'm just saying that realistic goals will help you achieve more in the short term, and the opportunity to re-evaluate these goals will help you in the long term.
Does this keyword require a new page? How many terms can I target on a page?
This depends on a couple of factors. It depends, again, on the competitiveness of the terms but also largely depends on the strength of the overall site or page on which the keyword is being targeted.
As covered in Rand's post in March on this issue, a highly competitive term deserves "single page targeting". This is true in most instances and particularly good advice for smaller, newer sites. The way this works is somewhat backwards, but experience suggests stronger sites can target multiple terms on a page yet also can afford to rank for a larger number of long-tail keyphrases. This may seem a bit unfair, but it is what anecdotal evidence has shown me.
The long and short of it? If your site is big, unless the keyphrase is highly competitive it can probably be targeted on a page targeting other similar terms as well. However, if the phrase is extremely competitive it deserves it's own page.
In the case of smaller, newer sites there are forces working against the best approach. On the one hand, smaller sites will have fewer pages indexed and will not have a great deal of authority to rely on and spread throught the site. On the other hand, this also means they cannot drop the term "breast augmentation" on the same inner page as "breast enhancement" and expect to rank for both terms.
How do I know if my Site is Strong Enough to Rank for that Term?
The Difficulty Tool pulls in some nice metrics from the rankings that allow you to see a fair bit of information about the other sites ranking for a particular search term or phrase. It won't always be a simple case of "my page/domain is stronger than theirs, thus I will rank". There will always be other factors: is the keyphrase an exact match for the Top Level Domain [TLD]? Are the other sites targeting multiple terms on the page or just the one? How many inbound links does the other page have? How relevant is this specific keyphrase to your term?
I think you get the point here, it won't always be a simple fix but let's look at an example to try to get a clearer idea of how to work through this.
Let's look at an example from a recent client project I did for a plastic surgeon (honest). In doing research for terms around plastic surgery for a brand new website I tried to get my head around some of the inner nuances of some of the terminology and procedures to try to better understand search behaviour. As you can see from the above research the broad match search volume for breast augmentation is considerably larger than the others [insert corny joke here], however the local search volume is quite comparable for the top two terms and they are about equally competitive.
This particular area of research can be quite complicated because you also have to look at the intent of the searcher and weigh that with the product offered. The term "boob job" whilst funny, is probably not likely to lead to serious searchers who are considering having a selective surgery so that coupled with the lower search volume means we can probably get rid of that one for now (though it might be worth bearing in mind for future link bait).
As an aside: by doing a bit further research into the types of pages that rank for the two terms and looking into search behaviours a bit it actually seems that people searching for "breast implants" are also not likely to be the highest converting traffic and there are certain social stigmas associated with the various terminology. So, in this case we can actually probably rule out breast implants (in terms of the main target of the top level page) because it is not likely to lead to highly converting traffic. This does not, however mean that we won't want to target the term at all, just that it may move down the priority list a touch.
So, the very first step has helped us eliminate two of the terms for the time being- so we're making some progress! The next step is to compare the terms that seem as though they might both be realistic targets for the page but are also relatively similar in terms of competitiveness. Although the scores assigned by the Keyword Difficulty Tool can be very helpful when comparing a term that is ranked as a "10" versus one that is ranked as a "95" these "difficulty scores" do not provide enough information alone when comparing two simialrly competitive terms. Thankfully, the tool gives us a lot more data to work with.
As you can see above, although "breast augmentation" seems to be a slightly "less competitive" term based on the diffculty metric there is a clear outlier within this chart (which we can go ahead and guess is going to be Wikipedia without even having the rest of the data) that looks like it will be extremely difficult to outrank even if the top spot seems slightly weaker than for "breast enhancement."
Similarly, the overall landscape for "breast enhancement" actually seems a bit more realistic as a target for a new site. Thus, in this case (based solely on the likelihood of ranking) we would actually choose to target "enhancement" rather than "augmentation." And try to work our way up to the more difficut term by building links to the site as a whole and specifically trying to target this page before shifting our approach on the term targeting. But before we make this a final decision, let's have a slightly closer look at what the competition really looks like.
As you can we were right in assuming the 5th spot for the "augmentation" comparison has been taken by Wikipedia, though not on a page directly targeted towards "breast augmentation" (hence why it's probably riding as low as it is in the rankings).
Meanwhile, setting aside the Wikipedia page it looks like the top spot for "augmentation" is actually being held by a rather weak site that happens to have great anchor text in the domain. This is a perfect indicator of just how much benefit having a strong TLD with exact match anchor text can be, but unfortunately this sort of tactic won't help much when you're trying to land the client who wants a "tummy tuck" instead.
So, what do we do?
In an ideal world, the client would be a great big site with loads of authority and without much sense. They will have been targeting "boob job" and have 302'ed all of their old links so we can make some quick changes and win. In this case, we go after all of the terms, do a bit of linkbuilding and we'll probably turn out just fine.
Meanwhle, in the real world situation we would recommend going for "augmentation" as our targeted term for a couple of reasons.
First, it is probably the "best" of the keyphrases in question. It targets the right kind of customer/searcher (we know this based on existing data and background insights on behaviour) and it has the highest search volume at the broader level.
Second, this keyphrase actually makes the most sense for the page we set out to build. As a top-level page it gives us the opportunity to (over time) target some of these other terms on the page (with the exception, maybe, of "boob job") . Augmentation is the most generic term and will allow us to discuss "implants, reduction, enhancement, etc."
Third, after having looked more closely at the sites/pages currently ranking for these terms it actually seems like it will be easier to rank for this particular keyphrase (please note that Wikipedia is not even directly targeting this keyphrase in this case).
"But what about all the other keyphrases? I don't want to waste them!"
This is where the post comes full circle. If you're building the small/new site the most sensible option (in the short term) is to create a page that is optimised for as many of these pages can be justified and for which you have research. As we mentioned ealier on, you can't just go after every single keyphrase in the industry on individual pages from the get-go because they won't all be indexed.
Try just to use some common sense: create the augmentation page high-up within the information architecture, construct a page for "reduction", "implants" and "enhancements" and forget about "boob job" for now. This term may get some traffic but if it doesn't fit with the theme/tone of the site then save it for the linkbait and build strong links to these inner pages now.
This technique creates much more work. But with a brand new site this is to be expected. Try to structure things in a manner that you can get rid of some of the smaller pages targeting extremely similar terms without impacting usability. This is essential and will make your life much easier in the future.
"What if my client is a massive site with great links?"
We should all be so lucky. This is obviously a different ball game we're talking about here. But, if you are fortunate enough to have a Domain Authority that is considerably higher than your competitors for the keyphrase(s) in question you aren't going to struggle too much and you only need one page to rank for a number of terms.
Image via: 3 Meeses
If this is your starting point, I would advise creating one hub/landing page for all "augmentation" related terms. If your site is strong enough the Wikipedia example quite clearly illustrates that some of these other pages may be superfluous.
There's no need to jam all these keyphrases in the title-tag either. If there are enough inbound links and the site is trusted enough you can probably just go for the highest search volume terms so long as the term is related to the service offered (never forget usability!), if you are in this position kick back, relax and just wait for the little guys to catch-up!
Actually, I WOULD go for "boob job" (no laughing please). I'm no expert on searching for cosmetic enhancements, or the people who do.
I do, on the other hand suspect a relatively lower IQ amongst those that were looking for such things. So they are more likely to use phrases that are easy to spell and quick to type. Going on the same logic, a "boob job" searcher with their lower IQ translates into a visitor that's easier to convert to perform an action. The one searching for "breast augmentation" will keep looking around.
Totally disagree. If we use your own logic that someone who searches for "boob job" will have a lower IQ, then we can assume - using this flawed train of thought - that they are also likely to have a lower income and therefore unlikely to convert into a paying customer. It is unlikely we want to class an action as signing up for a newsletter... it's hardly a topic people want monthly updates on.
What sort of conversion did you have in mind for this person of low IQ that would be useful for the plastic surgeon?
I'm thinking footballers wives in a nutshell, or those aspiring to be such (in that case, as it's an investment, they probably deem it worth taking out credit to get it done). In any case, it's good we all think differently - we'll all be chasing the same keywords otherwise.
I have to say, I felt similarly when I read the suggestion that "boob jobs" was out. I think you just explained the rationale for this feeling in a way that ruffled feathers!
The problem when you do keyword research is what Chip and Dan Heath (Made to Stick) call the "curse of knowledge." We all have our own set of highly developed filters that guide our thinking and language.
When doing keyword research it's important to remember you are a sample of n=1, and think about the market in the widest sense.
My gut feeling is that "boob job" would be a very productive keyword that would make it into my list, and the dismissal of it simply reflects an untested social bias on the part of Sam? :)
Jeremy.
I'll make this quick as I've inadvertantly derailed the topic of Sam's excellent post. I'd like to publicly apologise for the reference to IQ's. On reflection, I should have used "ordering vocabulary".
I'd be hesitant before making assertions about someone's IQ based on their search queries. Even if we conducted an extensive study and found that to be the case, the correlation between IQ and income is ~0.40. Even if we found the same correlation between IQ and search queries, it's likely that we could only be able to explain very little of the variance in income using search queries. The truth is that IQ isn't all that great of a metric. It gets tossed around all the time as some holy grail of personal worth, but in my humble opinion, it's extremely overrated.
I sort of agree with this, but not because of an automatic assumption that certain types of women tend to have a lower IQ. In this specific instance however, I have read stories that one of the most popular "sweet 16" gifts has become boob jobs (scary). If this is the case, then it is not likely the 16 year old will be making the actual purchase, but is likely that they are doing the research themselves and using terms like "boob job" could capture their interests.
As a side note... I haven't been able to use this tool for days. I sent a note to support with some questions about it and after hitting "send" the tool immediately stopped working all together. I haven't received a response back yet. Is it only me experiencing this issue?
It's working great for us - just tried again. If you haven't heard back, please shoot me a personal email - rand at seomoz.org.
Hey Andy - while I don't totally agree with the precise sentiment of segmentation by IQ based on keyword searches, I have to applaud (and give you thumbs up) for thinking so creatively.
I believe there's still a long way to go on the "psychology" of keyword-based intent and that some smart marketers are likely doing remarkable things on this front. There's probably a lot of opportunity here, too, and lower competition to boot.
Thanks for making me think a bit deeper about this topic!
I appreciate the step by step approach. This really shows how to break down the KW difficulty report and make some insightful decisions.
I think KW research and selection is the foundation for a successful campaign and I'll be sure to use this guide as a reminder when I'm explaining to clients how we have come to select our keywords.
Thanks,
Thanks for this great post. I really like those posts on how to get the most of the SEOMoz PRO Tools.
Yeah. What mathieuv said. The more posts that dig into the intricacies of the PRO tools the better. Bring 'em on I say.
It is interesting how many SEO companies do not use the keyword difficulty tool when determining their keyword strategy. I have used it to incorporate both the more difficult terms and less difficult terms so that clients can begin to see immediate imporvements in rankings. Granted...I am always straight forward about the level of diffuculty and clients seem to prefer to see moderate improve right away, over waiting months to start ranking for the more difficult terms. Clients also feel confident when you can provide them an idea of when they will see improvements and with what words, which the difficulty tool can assist with.
Once more great post and perfect add on by Gianluca. Thx for that! While all that is said makes pretty much sense to me, it is still crucial to explain the chosen strategy to your customer. and that's not always that easy.
'...but I WANT TO rank for all terms...'
Amen. That part is always the most difficult! Unfortunately I haven't got any great insights on how to handle that :)
Sorry, you "don't have any great insights on that"? With respect, managing customer expectations and having the ability to explain what is and isn't possible should be one of the first things you learn in SEO agency life...
Yep. That's what one learns pretty fast. And obviously that's pretty important to. BUT it is not always easy. If you only have customers who are like 'yeah man, you're the SEO guy, I trust in you.." then congrats!
A lot of customers are, especially when talking about smaller ones, are more like 'i give xx$ so come on and bring me to first place". Of course you have to be able to explain what this 'SEO thing' is about. But you can't assume that everyone will be insightful.
However that's what makes consulting fun. Time after time. You never know who's going to sit with you on the table :-)
Sorry for any confusion... what I meant by "don't have any great insights" was that I didn't have any insights that would work in all cases. Obviously this is something I deal with on a day-to-day basis what I meant by this was that I didn't have any broad-sweeping fixes for it!
That is to say, it takes a creative answer that should be decided on a case-by-case basis- because, as Sayfun so rightly points out unfortunately "because I know best" is not always the best approach to take with a client.
Ok, that makes more sense. Still, it might come back to the fact you sometimes have to turn down business if you don't want an unworkable set of demands from a client to come back and bite you at a later date.
I'm aware that, especially in this economic climate, that sweeping statement will have many people spluttering into their coffee.
That's spot-on Bludge.
There are definitely instances in which it is really not worth the headache for any of the involved parties. If a client can't be worked with and are unwilling to do what is necessary to be successful sometimes you have to throw in the towel and part ways.
Thanks for the great post. I think that applying your basic idea of what keyword phrases convert can also be applied to traffic analysis. Not only to determine why a visitor bounced on a certain page, but also to determine what key phrases tend to be the stickiest.
To take it a step further I have a client that tells me everytime his manufacturing company gets a RFQ so it would be easy to track down that visitor in the logs. From there it would be interesting to see if a pattern of keywords pops up. -> On a maintenence plan so it's not a total waste if nothing is learned. & if I do learn something... :)
Thanks again! You've definitely given me a few things to think about :)
Great post, especially for a new SEO with a plastic surgeon as a client. One word of advice though, Our "boob job" PPC adgroup IS converting!
Thanks very much for the insider tip! I've already started crafting a list of terms that might do well for PPC and linkbait/bloggy content (boob job, moobs, etc.).
I'm not sure I quite understand all of this, but something that stood out to my newbie SEO mind was this:
"If your site is big, unless the keyphrase is highly competitive it can probably be targeted on a page targeting other similar terms as well. However, if the phrase is extremely competitive it deserves it's own page."
We have a large national site that ranks right up there with the other big guys and is highly competitive regarding keywords. Are you meaning that in order to rank well for our highly competitive keywords we should target these keywords one at a time on individual pages of our site?
What about a navigation bar with tons of keywords we're trying to rank for that appears on every page? Wouldn't this be counter-productive, essentially cancelling each other out?
Maybe I'm just super confused. No, wait--I am super confused. Help?
Take my reply to the first part of your question with about 3 1/2 kilos of salt Summer.
If you've got an big authoritative site and you aren't getting the ranking for the keywords that are important, then consider breaking some of them out and sticking them in their own "silo". At the same time, as Sam pointed out, don't let your usability suffer. It's a balancing act.
Regarding your second question and your navigation bar, I too am looking forward to seeing what others have to say about it.
Hi Summer!
I hope not to have confused you too much :)
Usability is the key here. The weird thing is with big/authoritative sites you can generally get more individual pages indexed/ranking AND target more keyphrases on a single page. Seems a bit unfair to the little guys but that seems to be the way the old cookie crumbles.
Bearing this in mind, for the super competitive terms on your site (or those that are absolutely central to your brand/product offering) I would advise targeting the high volume/relevant terms on their own page. But definitely don't do this at the expense of usability.
Having a navigation menu with tons and tons of links will likely not look very nice from a usability standpoint- and it's also worth considering that as a general rule Google tends only to count the first 100 links on a page so make them count!
Hope this helps,
Sam
Thanks Guys. The whole navigation bar thing is hard to deal with because, as we're an e-retail site, it goes directly to the usability issue...we can't really get rid of it or pare it down any because then our visitors won't have any way to navigate the site in a variety of different ways such as by product category, by name brand, by vehicle year, make and model, etc.
When I say "ton of links" I only mean around 50 or 60, the major keywords we try to target, which are the product categories and they already each have their own pages. I plan to go back and reread this post a couple more times. :O
:)
Thx for answering my question Sam! Very insightful with a real client example. As for "What if my client is a massive site with great links?" I've been fortunate enough to come across a couple of these and after a few minor on-site nip and tucks, let's just say the clients were very happy with the results.
great post!
Really interesting read thanks.
I haven't actually used this before but will have a look into it now, I usually just use the google keyword tool and after some research into the company pick out the top results, when doing personal websites or freelance in the past.
Thankfully the team of SEO consultants we have at my current company are able to 'convince' the customers of which keywords to have instead of the customer telling us what they want.
Anyway, will look at this tool and work on my SEO a bit I think, coming from purely a design background my SEO isn't great but i'm getting there!
hey Sam, thanks for the useful insights on this post, it's great reading... I really enjoy the KW difficulty tool since it's been upgraded.
I know exactly what you mean about being able to confidently target the highest volume keywords if you are working for a big, trusted site, as I find myself doing exactly that for the British Council. However when SEOing the smaller sized sites I own and manage, I have to step down and choose my keywords a lot more carefully and taking into account many more considerations. Thanks again for sharing this.
Hi Sam,
really liked your post... and is one of those you have to read with calm to "catch it all".
I think you touched a sensitive note, at least in my case, because I have to struggle with small business websites as Bed and Breakfast and Boutiques Hotels... and you all know how the Tourism market on the web is hugely competitive... and how it is possible that hotel's profiles at Booking.com or Tripadvisor can rank higher than their sites themselves.
In these cases, apart from detecting the right keyphrase (and make it accept to the client), it's always useful:
Ciao!
Thanks a lot for these added tips! These are great and I think it's always nice to hear what people are doing with some of the "little guys" as it were. People tend to get caught up in "SEO is just for big sites" mentality and I think it's always great to hear what people are doing for the little guys. Thanks a lot for your added input and value to the post!
Sam, fantastic post. I own a consulting company and work with many small businesses and as we know SEO for small sites is MUCH different than large sites. I appreciate the differentiation in your post and am hoping to read nuggets of great information like this in the future.
Cheers!
I really like your posts Sam. They are well diagrammed and easy to digest so big thumbs up.
Although keyword research was one of the very first things I learned in my SEO walk, I can never get enough info about the subject. And tying them into the SEOmoz toolset is an added bonus.
Great post, but it's a bit like watching the rest of the class out at recess since I'm given no budget to have access to the specific PRO tools... Still, good ideas I can take into continuing my keyword overhauls. Appreciate the humor in the image usage too!
Great post, as one who talks to small business owners every day I now have more confidence in my own thoughts on keyword selection.
The comments on boob jobs as an indicator of intent is thought provoking too - mmhh a play with the commercial intention tool at Microsofts Adcentre does indeed suggest its a no-hoper!
https://adlab.microsoft.com/Online-Commercial-Intention/Default.aspx
Does that click through rate chart apply to longer tail keywords, such as 3 to 5 words?
Also, the Keyword Difficulty Tool is only available to members. Will this ever be open to the public?
Is there any way to look at our competitors authority? an online tool or Excel spreadsheet to count?
Help me
Thanks for giving enough infromation to understand and knowing how to apply the right key phrases in every smaller sites is a big relief.
Great Post Sam, I love posts like these, they're not revolutionary or going to rock my SEO world, but they help me focus and clean up one particular part of my SEO game. Which is all I can ask for.
Thanks for the tips.
As a beginner in the industry it is extremely useful and I think the keyword tool is great for determining which keywords should take priority. The temptation being to use what we think is the most commonly used words, when in fact I suppose it`s always worth looking at alternatives. Convincing your customer is part of the consultancy package anyway and so it is your job to explain the reasoning behind using such words. Pulling the information though from Google is all well and good but there are never any guarantees that the trend will stick, especially during the course of the SEO campaign.
That is a very interesting approach to sculpting your site to be relevant to various keyphrases. However as a very crude test of importance , when building our site from scratch and then optimising where possible, it wasnt until we really focused on PageRank / linking that we made serious inroads in search pages. All areas are important but relevancy seems very tightly linked to number of people who care!
https://www.kennelmaster-software.com
Cheers for the info, good post.
Thanks for the practical tip on a Monday. Cheers!
Hey Sam,
Very nice tool it really helpfull for small size business.
Good and practical post,
Did not get to use the following tool yet to such an extend. Now will have more confidence to use it in future.
Thanks,
Really great post Sam, big thanks!
I will try and see how your description falls in to our working method, because it really seems interesting!
Great post. Why does top10herbal ranks #3 despite having way more links than the first site? I can't really figure why that could be the case.
Hi Jest!
I can't be entirely certain, though it does look like the dMR for the site ranking first for augmentation is considerably higher (5.51 vs. 4.36). If the little Google PageRank toobar is anything to go by (which I realise in general it is not) the same is true of the homepage's ranking (5 vs. 3).
I would say it's safe to assume that the old "quality vs. quantity" is playing a major role here but I couldn't be certain without digging a bit deeper.
Hope this helps!
This is a good demonstration of the Keyword Difficulty Tool in action.
Are there any video tutorials for the SEOMoz tools? I think that would be super helpful.
There aren't, but I'm actually connecting with someone today on this front (for screencasts of tools). Glad to hear you think it will be valuable. We'll certainly put KW Difficulty at the top of the list.
fantastic post Sam !
Cracking Post Mr. Crocker. Very good to see some tactics for the smaller volume sites out there. This is going to be particularly useful for us at the moment. Thanks Again Charlie
(edit from Rand - removed the link, even though we do love the Link Juice app. If you'd like to link to something, just please make sure it's relevant! Signature links aren't quite our speed here).