Wouldn't it be great if you could somehow spot those SEO opportunities on your site which were low effort and high value? Well this post gives you a few ways you can do that! Sweet.
I'm going to be digging around in the recently released search queries report in Google Webmaster Tools:
Step 1 - Gathering The Fruit
The first thing we need to do is gather all the fruit (aka keyphrases). So within GWT select search queries and select just "web" queries and in this case I've selected "United States" since that's the main target market for SEOmoz. The more we can narrow this down the better data we get, if we leave image search etc in there and leave countries like Serbia in there the less accurate our data will be:
Once we have filtered the data we then want to download the data to Excel:
Step 2 - Identify The Low Hanging Fruit
Once we have the data in Excel we can do some monkeying around to get some meaningful insights. When you download the data you will be presented with a lot of dummy data like this:
So I run a find and replace on the following two items:
- Find "<10" and replace with 1
- Find "-" and replace with 0
Be sure to only run these over columns B,C,D to avoid stripping out anything from your queries column in A!
Now, once we have this data it will depend on the number of impressions and number of keyphrases how exactly you want to slice the data. Analysing the data for SEOmoz I found that selecting all avg positions that were not equal to 1 and impressions over 200 gave me a sample set of 97 keyphrases to look at which wouldn't take very long at all to whiz through and look at. If you have more time or if you have more keyphrases you might want to get a longer or shorter list.
As I mentioned this gives me a list of 97 keyphrases for the SEOmoz site. Let's take a look at what some of those opportunities are!
An Example: SEOmoz's Low Hanging Fruit
In this post I not only wanted to show you how to get the data but also to give you an insight into what kinds of actions you could take and what sorts of keyphrases you might look at so I ran the above process for the SEOmoz site and found the following low hanging fruit. Bear in mind that there weren't that many really easy wins in the data since SEOmoz generally has fairly good SEO (unsurprisingly!). Still, it gives you an idea of the thought process.
Keyphrase: SEO | Ranking: 9.4 | Impressions: 49,500 | Clicks: 590
Ranking URL: https://moz.com
Now, I've shot myself in the foot a little here by picking a keyphrase which isn't really a low hanging fruit, it's actually a highly competitive keyphrase! That said, it's useful information to have. Without having rank tracking set up for the site it instantly tells me that SEOmoz are highly competitive for this term, especially as some of the sites that rank above them are Google and Wikipedia. That said, there's room for improvement. Maybe time for some more badge-bait Rand?!
Keyphrase: Social media marketing | Ranking: 7.9 | Impressions: 8,100 | Clicks: 320
Ranking URL: https://moz.com/article/social-media-marketing-tactics
Again, this is a highly competitive keyphrase but one which I feel SEOmoz could perform better for. The current ranking is working ok for them but could certainly improve dramatically if they could shift the ranking from 7.9 into the top 3. Digging around we see that the page is linked internally from every page in the navigation with good anchor text and it has a total of 255 root domains linking to the page so there's clearly a fair amount of work already done here. That said, I feel like there's an opportunity waiting since SEOmoz links out to lots of other sites from here and most of those blogs will likely link back to the SEOmoz guide if there was a nicely written email. Incidentally, if you're looking for a link from SEOmoz and have a top notch guide for one of the sites which doesn't have any resources attached then get in touch! So long as you link back to the page ;-)
Also, BONUS TIP - while researching backlinks in this space I stumbled across the fact that Amazon authors can get links from Amazon.com check out Darren Rowse on Amazon and then compare to Rand Fishkin on Amazon and you'll see that Rand has missed an opportunity to get blog posts imported and hence get clean followed links from Amazon. Sweet!
Keyphrase: What is seo | Ranking: 3.9 | Impressions: 1,900 | Clicks: 210
Ranking URL: https://guides.seomoz.org/beginners-guide-to-search-engine-optimization (Note here that this URL isn't reported in GWT, it's the old beginner's guide URL which now redirects but the same keyphrase stands).
Here, I see the answer being a little easier than the above keyphrases. The term is less competitive and the title of the page doesn't even mention "what is seo"! My actions would be to reword the title tag to be "What is SEO? The Free Beginner's Guide to SEO from SEOmoz" and to mention "What is SEO" on the page at least once (currently it only mentions "what is search engine optimisation" and although Google knows they're the same phrase I'd still like to see the exact phrase on the page somewhere). Also, there is no navigation link on the site to the beginner's guide so slipping a few links into the next few blog posts with the anchor text "what is seo" will help boost the rankings for that phrase.
Keyphrase: Free seo tools | Ranking: 4.2 | Impressions: 480 | Clicks: 73
Ranking URL: https://moz.com/tools
The term "seo tools" is fiercely competitive but the "free seo tools" term seems like it would be a lot easier to go after, in fact SEOmoz is one of only 2 of the top 10 results which doesn't mention the term free in the title tag of the page. This could be rectified easily and in addition to that the page doesn't even mention "free seo tools" on the page. Personally, since this is something people search for I'd be tempted to re-design the page to add a "Free SEO Tools" sub-header to differentiate between the free and PRO tools. That way it's a good user experience and also gets the phrase on the page.
Keyphrase: Keyword research | Ranking: 19 | Impressions: 110 | Clicks: 12
Ranking URL: https://moz.com/blog/10-steps-to-advanced-keyword-research
Note here that the impression numbers are so low because the page is ranking 2nd page. Not having a page ranking in the top 10 here is a mistake for SEOmoz I think (sorry, I mean opportunity!). The correct page is clearly the page on keyphrase research from the new beginner's guide and the best way to make that page rank is to throw some more internal links to the page. Currently there are basically no internal links to that page except from other beginner's guide pages. Linking to it from elsewhere should be able to get that page ranking. One idea to help get internal links to pages of the beginner's guide would be to automatically link to the keyword research page from any blog post within the keyphrase research category. That way it would essentially get more deep links internally to the individual pages of the beginner's guide.
A Data Warning
A warning here that I'm still not sure how much I trust this impression and click data. I wrote a post over on Distilled about how the reported numbers are way out from analytics numbers. To be honest, if I was doing SEO full-time for SEOmoz I'd like to think I'd have better resources of keyphrase research, ranking data and visitor numbers but using Google's webmaster tools search queries report can be a nice quick way to identify keyphrase opportunities for a site which you're less familiar with or for which you don't have these things set up.
Thank you Tom for this alternative method.
As you says replying to seo-himanshu, this one is not meant to be the only methodology to use to find out missing occasions. But it can be surely a good base, as it offers the opportunity to check out fast what are those opportunities a website is missing. This quality - to detect in a glance the low hanging fruit - is the most important I see in this method.
Then, the results have to be confirmed by other tools like Google Analytics, especially to better evaluate the ROI the optimizations can generate for those keywords.
Finally, what I appreciate is a secondary value of your post: the fast but meaningful insight to your work methodology.
Hi Tom, Sorry - I gave a thumbs down.. reason being the Amazon tip - which you should really add to the "protips" section on seomoz, unless that area isnt being updated anymore? @rand? I do actually agree with a lot of what you've written, however I would add that with the "starring" of queries my preference is to star all "non-brand" keywords, and then run the above on those. As brand based keywords (at least when its your brand) I'm not that interested in, seeing as we're in position number 1 for most of them anyway.. the functionality for downloading just your starred queries I've been told is on the way! Rob
*edit* Thanks to a twitter conversation thats now a solid thumbs up to Tom :)
Thanks Rob - I've added as a protip for those busy pro members like yourself :-)
We should really update the tips section more often...
As for the starred queries - it can be really handy to take a look at that data, though equally you could filter within excel quite easily.
lol amused at the two thumbs down - clearly not pro members ;)
thumbed you UP just for having the guts to post the reasoning behind your thumbs down on the post. I hate when people thumb down without giving any constructive criticism or reason for the thumb down.
I'm thumbing you up for thumbing rob up for thumbing my post down! I totally agree that it's a good thing when people give reasons why they thumb things down - if we don't know why it was thumbed down then we can't make it better in future.
I totally agree with your comment, Tom... aslately there's been a thumbs down fever of that kind
Thumbs up Rob. It would be great to see the Pro tips more active.
Nice tip Tom - been using this one for the last few weeks to push content up from the second page.
Has anyone seen a grease monkey script to pull the full dataset yet? May need to break out my JS toolbox!
I really, really like that idea!
I love how dummy-proof SEOmoz makes SEO. If you come here and read a blog post every day, you too can be a pro SEO. In fact, that's how I got my new job (day two!)!
Awesome, congrats!!
Ohh awesome! Hope day three will be better than days one and two put together.
If 70% of GWT keyphrase data is useless, then what's the point of identifying low hanging fruits through GWT. GA 'top landing pages' and OSE 'top 'pages' are much better in finding such fruits.
Firstly, the 70% figure is pure headline-bait. That's not any kind of accurate analysis that's produced that figure (as stated in the distilled post).
Secondly - most of the inaccurate data in my opinion comes from the clicks (and perhaps because of this the impression data) but the ranking numbers have been pretty spot of whenever I've checked them so this method of identifying existing ranking positions is still valid. I agree that there are slightly more robust ways of getting these kinds of data points but actually most of them require an awful lot more work - I was impressed how quickly and easily I could identify quick wins for a site which I wasn't familiar with.
The aspect of this which top landing pages misses from GA is the ranking position and popularity of the phrase - yes you can add these into the data but as I mention this is quite labour intensive and you start to rely on other unreliable data points such as the google adwords tool. All of these tools have inherent noise and therefore I think there's value in approaching these problems from multiple angles - I'm not saying this should replace using the top pages report or any other method, just that this is another way of getting some interesting data.
You've excelled yourself again, Tom. Ever since the Search Queries feature appeared in GWT I've been trying to utilise it as much as possible.
Fortunately it's quite self explanatory, but this article is a bit of a kick up the arse to make sure that we're utilising it's features correctly.
Cool Post Tom- unlike other Excel-Tricks this one can actually be done within 5 minutes, while reading the article.
Although the data may not be 100% accurate, it surely helps to identify some of the low hanging fruit very efficiently. Definitely helpful for quick wins with limited resources!
Generally speaking I find larger volume sites to be more accurate with WMTs.
I usually copy and paste the list of keywords driving traffic into Adwords, filter off new suggestions and then just copy n paste them into my spreadsheet. Use Vlookups to populate an extra field then calculate market share to spot some more fruit.
Thanks for another "easy to follow along and implement" post Tom. While I'm not an excel novice, I always love excel tips n' tricks kinda posts that walk me through the process.
Interesting that many of your recommendations are internal linking oriented. We blog and try to work on our internal linking all the time, so this post makes me happy and gives me some new ideas.
The Amazon tip IS a gem...now to just go write a book :P
Really good post Tom. I think if SEOmoz create an API for their Keyword External Tool we can merge with your idea (spreedsheet) and then we can get a tip if that keyword is or not difficult to rank.
Again, thanks for sharing this ideia.
After I have filtered (for example > 200 impressions and average position not equal 1) -- What data relationships do you suggest I look for? Could you share the mental rules you use to identify opportunities (especially when you are new to a site)?
Hi Tom,
I agree with your suggestion to differentiate between free seo tools and paid seo tools, as it creates a streamlined user experience. I have seen this work with a gaming site that implemented language to clearly define free online games and subscription based online games. After the implementation, their registrations for free online games increased dramatically - as did their ranking. User experience should always be a focus, but it becomes particularly important once a achieves a high ranking.
Regarding your suggestion for internal linking, you stated, "One idea to help get internal links to pages of the beginner's guide would be to automatically link to the keyword research page from any blog post within the keyphrase research category." my question is, does it matter if the internal links are in-text links, inside of the footer, or a part of the primary navigation bar? I am curious if the location of an internal link from another page on SEOmoz, for example, would have an impact one way or another.
Thanks for this post, it's been of great use. However, I was wondering if you could shed some light on why my Search Queries tell me I'm ranking highly, both for some really desirable keywords and some entirely undesirable that are all very highly competitive. I'd like to say it's because our site rocks but the truth is we just aren't ranking for them.
For example, we're a video production company but for the web. We never mention, discuss or promote TV advertising yet it tells us we're number 3 for "tv". Utterly ridiculous, untrue and irrelevant.
Ryan
My question is whether this data is exclusively SEO data because, Tom, like you I don't trust it. The click data I have just reviewed for a client is too good to be true so I am wondering whether it is incorporating paid search data too....
This post has been very helpful to me in organizing data from GWT Search Queries - I'd also recommend Hugo Guzman's post on using the search query info to find "money keywords". My method is now a synthesis of both articles. I'm writing now (months later) really to comment on the great tip about Amazon author bios. We did it for one of our clients, but maybe after this post they got an influx of new blog feeds and clamped down, because all links from Amazon blog feeds now appear to be nofollow. I'm not saying that's not worth anything (subject to the great debate), but it's just something readers should be aware of.
Best,
Aviva B
Thanks for the tips! It came just in time - today I had decided to do some investigation into GWT Search Queries and how they can best be used for optimization analysis. A search for anything people have written about it - and your post came up first, just posted yesterday. Thanks!
I got a lot out of this. It is helping me see areas for improvement in our site in terms of SEO. Thanks.
Very good information . I am not able to understand the logic for
When i ran find and replace search, excel showed "Search key not found" Why ?
I have been reading SEOmoz for quite a while now and have picked great stuff.
I had the same error; try typing in the keys without the quotes around them and see if that works. Worked for me, hope it does for you too!
Sorry for any confusion, you should be searching for the text inside the quotes, not with the quotes included. Let me know if that doesn't solve the issue.
Nice post Tom. I love the actionable tips!
Is it me or has the export recently started to include ranking data? I played with the data, comparing to KWT volumes a little while ago and if memory serves - the rankings were not included.
https://seogadget.co.uk/playing-around-with-google-webmaster-tools-click-data/
Yeah, I'm pretty sure it has started to just recently.
Yeah - they (The GWMT team) really should let us know about all these little tweaks :-)
This seems like a great way of searching out oppertunities specific for certain pages and keywords...thanks for sharing!
Right, off to log in to webmasters....
Excellent post Tom. Really good examples of how to use the GMT Search Query Reports for SEO!
I like what you show, in the past I was shackled down to going thru the analytics segmentations-nav summary ,,, and all that can get quite numbing at times. We'll give this a try to see its effiencey, Thanks Tom!
ive got my tool SEO Site Tools newest version [3.0] pulling those downloadable csv and parsing them automatically so it would be trivial to add a report to show this... ive also got oAuth working so you can create reports from your google analytics and webmaster tools data (no excel required! even better all data processing happens on your browser no external server to proxy or anything)
oh and i need samples of bad seo for my work on detecting cloaking and hidden text on pages so please help and send it in
bother me on twitter! @cartercole
Just curious: what is your creative mind planning with those examples (I can give you some, even if in spanish).
yes please... one of the things i want to add soon is multiple language support so i will also need help with translations if you were interested in that
drop me a message at [email protected]
Great post! I did the analysis and now I have a question - I think one of the reasons many of our pages are getting rankings and impressions but not clicks is that the pages that rank aren't hugely relevant (they're product pages). Do you think it's worth putting up a totally separate (and highly relevant to the term) optimised landing page for each of those keywords? Will the fact that the site already ranks for that term (but on a not particularly relevant page for that term) make any difference, for the positive or for the negative?
Great article here Tom: we did something very similar a while back before creating a couple of blog posts for our site and ended up ranking pretty well for [viart review] in the UK. We get a reasonably steady, if low, stream of traffic and attract new links from time to time, both of which are pretty handy for a few minutes' extra work.
Wonderful article,
I have just starting using it and it is great, good to see some great data, which we can really use