Sometimes, the page you're trying to rank - the one that visitors will find relevant and useful to their query - isn't the page the engines have chosen to place first. When this happens, it can be a frustrating experience trying to determine what course of action to take. In this blog post, I'll walk through some of the root causes of this problem, as well as five potential solutions.
When the wrong page from your site appears prominently in the search results, it can spark a maddening conflict of emotion - yes, it's great to be ranking well and capturing that traffic, but it sucks to be delivering a sub-optimal experience to searchers who visit, then leave unfulfilled. The first step should be identifying what's causing this issue and to do that, you'll need a process.
Below, I've listed some of the most common reasons we've seen for search engines to rank a less relevant page above a more relevant one.
- Internal Anchor Text
The most common issue we see when digging into these problems is the case of internal anchor text optimization gone awry. Many sites will have the keyword they're targeting on the intended page linking to another URL (or several) on the site in a way that can mislead search engines. If you want to be sure that the URL yoursite.com/frogs ranks for the keyword "frogs," make sure that anchor text that says "frogs" points to that page. See this post on keyword cannibalization for more on this specific problem.
_ - External Link Bias
The next most common issue we observe is the case of external links preferring a different page than you, the site owner or marketer, might. This often happens when an older page on your site has discussed a topic, but you've more recently produced an updated, more useful version. Unfortunately, links on the web tend to still reference the old URL. The anchor text of these links, the context they're in and the reference to the old page may make it tough for a new page to overcome the prior's rankings.
_ - Link Authority & Importance Metrics
There are times when a page's raw link metrics - high PageRank, large numbers of links and linking root domains - will simply overpower other relevance signals and cause it to rank well despite barely targeting (and sometimes barely mentioning) a keyword phrase. In these situations, it's less about the sources of links, the anchor text or the relevance and more a case of powerful pages winning out through brute force. On Google, this happens less than it once did (at least in our experience), but can still occur in odd cases.
_ - On-Page Optimization
In some cases, a webmaster/marketer may not realize that the on-page optimization of a URL for a particular keyword term/phrase is extremely similar to another. To differentiate and help ensure the right page ranks, it's often wise to de-emphasize the target keyword on the undesirable page and target it more effectively (without venturing into keyword stuffing or spam) on the desired page. This post on keyword targeting can likely be of assistance.
_ - Improper Redirects
We've seen the odd case where an old redirect has pointed a page that heavily targeted a keyword term/phrase (or had earned powerful links around that target) to the wrong URL. These can be very difficult to identify because the content of the 301'ing page no longer exists and it's hard to know (unless you have the history) why the current page might be ranking despite no effort. If you've been through the other scenarios, it's worth looking to see if 301 redirects from other URLs point to the page in question and running a re-pointing test to see if they could be causing the issue.
_ - Topic Modeling / Content Relevance Issues
This is the toughest to identify and to explain, but that won't stop us from trying :-) Essentially, you can think of the search engines doing a number of things to determine the degree of relevancy of a page to a keyword. Determining topic areas and identifying related terms/phrases and concepts is almost certainly among these (we actually hope to have some proof of Google's use of LDA, in particular, in the next few months to share on the blog). Seeing as this is likely the case, the engine may perceive that the page you're trying to rank isn't particularly "on-topic" for the target keyword while another page that appears less "targeted" from a purely SEO/keyphrase usage standpoint is more relevant.
Once you've gone through this list and determined which issues might be affecting your results, you'll need to take action to address the problem. If it's an on-page or content issue, it's typically pretty easy to fix. However, if you run into external linking imbalances, you may need more dramatic action to solve the mismatch and get the right page ranking.
Next, we'll tackle some specific, somewhat advanced, tactics to help get the right page on top:
- The 301 Redirect (or Rel Canonical) & Rebuild
In stubborn cases or those where a newer page is replacing an old page, it may be wise to simply 301 redirect the new page to the old page (or the other way around) and choose the best-converting/performing content for the page that stays. I generally like the strategy of maintaining the older, ranking URL and redirecting the newer one simply because the metrics for that old page may be very powerful and a 301 does cause some loss of link juice (according to the folks at Google). However, if the URL string itself isn't appropriate, it can make sense to instead 301 to the new page instead.
Be aware that if you're planning to use rel=canonical rather than a 301 (which is perfectly acceptable), you should first ensure that the content is exactly the same on both pages. Trying to maintain two different version of a page with one canonicalizing to another isn't specifically against the engines' guidelines, but it's also not entirely white hat (and it may not work, since the engines do some checking to determine content matches before counting rel=canonical sometimes).
_ - The Content Rewrite
If you need to maintain the old page and have a suspicion that content focus, topic modeling or on-page optimization may be to blame, a strategy of re-authoring the page from scratch and focusing on both relevance and user experience may be a wise path. It's relatively easy to test and while it will suck away time from other projects, it may be helpful to give the page more focused, relevant, useful and conversion-inducing material.
_ - The Link Juice Funnel
If you're fairly certain that raw link metrics like PageRank or link quantities are to blame for the issue, you might want to try funneling some additional internal links to the target page (and possibly away from the currently ranking page). You can use a tool like Open Site Explorer to identify the most important/well-linked-to pages on your site and modify/add links to them to help channel juice into the target page and boost its rankings/prominence.
_ - The Content Swap
If you strongly suspect that the content of the pages rather than the link profiles may be responsible and want to test, this is the strategy to use. Just swap the on-page and meta data (titles, meta description, etc) between the two pages and see how/if it impacts rankings for the keyword. Just be prepared to potentially lose traffic during the test period (this nearly always happens, but sometimes is worth it to confirm your hypothesis). If the less-well-ranked page rises with the new content while the better-ranked page falls, you're likely onto something.
_ - The Kill 'Em with External Links
If you can muster a brute force, external link growth strategy, either through widgets/badges, content licensing, a viral campaign to get attention to your page or just a group of friends with websites who want to help you out, go for it. We've often seen this precise strategy lift one page over another and while it can be a lot of work, it's also pretty effective.
While this set of recommendations may not always fix the issue, it can almost always help identify the root cause(s) and give you a framework in which to proceed. If you've got other suggestions, I look forward to hearing about them in the comments!
The best way I find to see what the search engines thinks is the best page from your site for a term is to do a site search with the term you are checking.
E.g.
site:yousite.com + Term
This shows you the pages in order that Googles ranks them. Also its a good way to identify duplicated title tags.
Great post!
Thanks for the advice Andrew ! I just discovered that, for a specific keyword, not all the pages I wanted to be displayed first are not !! Let's go back to work now :D
And thank you Rand for this helpful post!
Where is the thumbs up then? ;)
Just gave you one now.
Thanks for the tip Andrew :)
Just got one from me too Andrew. Thanks for sharing the technique with us.
One thing to remember is that Google want to serve the user both relevant and unique search results. I guess this can be one reason as well. Here goes...
You have three pages:
- "Green car", which also mentions yellow and blue.
- "Yellow car", which also mentions green and blue.
- "Blue car", which also mentions green and yellow.
Your competitors have a page about "Green car" as well.
User search for "Green car". Google likes your competitor pages best, but thinks your "Yellow car"-page is interesting because it's slightly different ("unique") than your competitors pages, and therefore decides to show the Yellow-page instead of the Green, even your "Green page" IS more relevant.
Yeah - good point. It could be part of the QDD algorithm.
Good to know that srm!
Thanks for the advice!
Thanks, Rand. This will be a great resource for Q&A, as some variety of this problem pops up often in questions.
I'd like to add just one aside regarding external link bias. If you have a lot of external links with anchor text favoring the page you don't want to rank 1st, don't always assume everyone else is wrong. Sometimes, we get so caught up in how we think our sites should rank and trying to control the visitor funnel that we miss out on behavior that would actually be more beneficial (and profitable) in the long run. Be willing to consider that Google might be right on occasion.
This is another reason why you should pay attention to your Title and Description, let the user decide which page is the most relevant to their search by providing them with the right kind of information to make that decision.
Most business owners feel that traffic comes through their index page, and after they see some basic stats they're often shocked. Simple stuff like pagination can mean the difference between focal points shown, and missing.
Great post, although I don't think I will risk testing on the corporate site.
It also brought the thought of reputation management to my mind, the following describes influencing external pages position from Search Engine Journal: https://bit.ly/ctKmnC
I'm very glad that you mentioned the internal linking issue. Too many people simply overlook the tweak they can make on their own site that make a big difference.
Oftentimes, we're focused on trying to get more external links to a page when all you need to do is fix your internal site structure and your internal links.
I'm not saying that external links don't help, but first make sure your site is fixed.
What about a large surge of clicks to one page? surly google would change it thinking it more relevant.
How many I don't know and a bit far fetched but just my 2 cents :)
It would also probably only work with much smaller sites or at least only doable due to traffic volumes and how many people you could get to help you.
Say all work colleagues on their home computers?
I watch Alice in Wonderland last night so sorry if this is a bit crazy :)
We've talked previously about usage data and CTR in the SERPs. My opinion is that a tactic like the one you describe wouldn't be effective and that on a granular scale, clicks and click-through rate are too noisy and gameable to be used by the engines. On more macro-levels, I certainly believe they will use them to tune and refine, but I'd be careful about presuming that getting your friends to click will help move the rankings needle.
I dont think this helps too. You must renew the amount of traffic u send. Like news clicks are maybe transitory, in my opinion. So your work may start again and again. More likely you are in best practice by working one time as suggested.
Thanks Rand, from the looks of things it looks like you're still pumping out some of your best work while sleep deprived haha. I think the only thing worse than you being up starting a post at 11pm West Coast time was me being awake and in the office to read you tweeting about it at 7am GMT!
The same here... 8 a.m. (Madrid time)
To read that tweet switched my "MozAlams" on.
Certainly an interesting quandary. If you find yourself in this kind of situation it's certainly worth taking your time and making incremental changes especially on competitive key phrases as alluded to here. It may be a case of just being greatful for what you have! The suggestions seem pretty sound to me although some are fairly dramatic so I'd definately start with the smaller changes just to try and nudge Google in the right direction.
Doing "nothing" (apart from monitoring the situation) is a viable option sometimes, but I think these are good tips if you really feel the need to do something about it.
How about throwing a link with the desired anchor text you want to rank for, from the "first listed page" to the "second listed page" which is the one you want to rank first? I think that this will also nudge Google and you can see the reaction.
this a very good advise! to me it's better to tell Google what is the best landing page for a given kw from the page google bot had previously choosen
I think that the right strategy may depend on the link disparity between the two pages. A very severe disparity would probably require some link juice funnelling or external linking to the second page, since nudging probably won't be enough to overcome a huge disparity.
A great comprehensive take on a very common problem.
"Just swap the on-page and meta data (titles, meta description, etc) between the two pages..." Surprised to see "meta description" in there, since it's not in itself a ranking factor (except perhaps via its influence on CTR, which is a ranking facotr). I would say of all the on-page stuff the title is the most important one here.
But of course the devil is in the links! If we search for "Matt Cutts," for example, we get this:
Note how his glog ranks well above his site's page. It's one of those not very common cases where the PR of the blog's main page (7) is higher than that of the site's home page (5), all because of backlinks and relevance. This appears to be intentional: notice how the main home page lacks a proper title, and the first thing you read on it, in large bold letters, is
You probably want my blog.
with a link to the blog page, which is also linked to from all over the world, it being one of the most popular SEO blogs (but no match for Rand's IMHO!).
So anyway, here it's intentional, but when it's unintentional it happens for exactly similar reasons. I just thought it's a suitable illustration of Rand's point here.
Philip, forgive me if I sound rude, but I do think that Matt Cutts' blog doesn't link to the home page. And since the home page links to only Matt Cutts' blog, It passes the PR at a great value.
Then, mattcutts.com + external links = PR 7.
Please correct me If I am wrong.
interesting post..
typically, it has been my experience that the homepage seems to rank for just about anything and the internal pages struggle because of the links that the homepage has
no amount of on page optimisation seems to help with Google rankings in these cases - all I have been able to do it to increase the quantity and quality of links to the internal pages and wait...
it is a nice feeling when the ranks start to improve and the effort pays off
we also use this approach to help our website ranks better on "shoes" searches here in french. We have some good results and we've seen the PR of internal pages growth
Good post Rand - just a note but I bet this is the kind of post that people would like to see examples for. I'm sure SEOmoz has the wrong page ranking for all kinds of things :-)
I had a bunch of examples from members, but didn't request permission to post them. If I find some SEOmoz ones, I'll be sure to add them in. Good call Tom!
This is a really good clear way of going through the issue. Sometimes it can be hard to work through these things systematically. It's sometimes instinctive to start with the strategy that may not actually get to the root of the problem (ie killing with external links)...
Is this what kept you awake the other night?
On the positive side, you got a double listing!
I've been careful to make sure the pages I want ranked are getting ranked, and haven't had this issue, but I'm glad to have a few ways to correct it if it does happen. Thanks!
This is the issue I am facing with my site. I was so confuse why all my design pages are ranking for Development key phrases. This article is an excellent help for me.
So what happens when you write several hundred blog posts about, say, SEO? How do you stop all of those posts from not only competing against each other, but more importantly competing against the SEO services you offer?
PS I'm not an SEO provider.
This is the kind of things that I'm to glad to learn. Because when you are in a situation like that and you've just read the books that are all over the web, you learn how to grow the page rank of your site, but you don't know how to attack this kind of problems when you find it.I still have to much to learn about, but you make it easer Rand, thanks.
I seem to be stuck with a combination of 2 & 3. An adverse consequence of too many directory submissions. Put simply, they resulted in a large number of inbounds, with targettted keywords as anchor text, but since they were directory submissions they all point to the homepage. We've done all the onsite changes we can but are having real trouble getting the right pages to rank. Taking the keyphrase off the homepage (apart from as an anchor to the "right" page of course) isn't enough to fix this in my case (did that a month ago!).
It doesn't seem to have been mentioned, but the right landing page appears to get much better click throughs, and much lower bounce rates. When users get dropped right on the homepage it's clearly too much work for today's web surfer to hunt around for the right link.
If I had my time over, I'd be much more selective about what anchors linked to the homepage :(
Intersting post and good reminder Rand. But what about the news world? Any difference in it?
Thanks Rand. This is good stuff and I'll be bookmarking it for future use, but for right now, I'd settle for any page to make it to the top of the SERPS for my current clients ;-)
Kill with both internal and external links. In case of internal links, virtual silos can help. If the new relevant page is not ranking well then link it out from the less relevant but stronger page(s) which are ranking above.
Great post Rand! Thanks!
I was wondering, however, how would a 410 status code (GONE) influence the ranking. Is it a alternative for the 301 status code? This question comes from the context where you have a newer page regarding the same content, but you are opting to go for a 301 (old to new) to rank the new page higher than the old one.
I guess the biggest downside would be the loss of historical data on the old page, but I'm curious to your take on that :)
410 means permanently removed - if you do that, just be aware that any link / authority / trust / external metrics applied to that page will evaporate. In cases where a page is actually ranking for a term you're targeting, I'd be much more inclined to save the juice and/or redirect it elsewhere.
Thanks, that was my thought as well (losing all the current data to the old page). However, I couldn't stop thinking about it in relation to this post. It's the easiest solution, but also the most harsh one.
Thanks for that Rand! its one of the most common nightmare's for SEO's. We did successfully manage to get a site resolve this issue however we did not dig as deep as you have listed.
Funny, I've never really thought of which page I'll rather rank over the other. I've seen certain pages ranking for certain things I would rather it doesn't but never thought to investigate it because I believe it all leads to the same place.
I guess when it comes to regular content and blogging this can be very important. So it's something I'll make a effort into fixing.
But that said, I've definitely seen #3 in action! The joys of linking internal pages from juicy external links!!
Thanks for the tips Rand.
I see #1 most frequently too, which is a good thing because it's the easiest of the 6 to correct.
Funny you should post about this rand, I have been racking my brain over this on one of my sites this weekend,
I have a "Notice: Undefined index:..." which relates to a WP plugin on my site ranking or above my main site url... I have been trying to figure out why and how to fix it. I will take some time tomorrow to try some of these methods and see if they help.
Shawn
I was had an issue like this about about a month ago. Our issue was on page optimization. By devaluing the keyword on the page that was ranking and keeping the page we wanted to rank the same we saw a positive change.
Either way it is alway nerve-racking. You would hate to see both page loose the ranking in the serps.
Thanks for the tips. I'm sure Ill be visiting this page again in the future
Great Post, im really agree with you. Jorge Hane