Are you guilty of living in the past? Using methods that were once tried-and-true can be alluring, but it can also prove dangerous to your search strategy. In today's Whiteboard Friday, Rand spells out eight old school SEO practices that you should ditch in favor of more effective and modern alternatives.
Click on the whiteboard image above to open a high resolution version in a new tab!
Video Transcription
Howdy, Moz fans, and welcome to another edition of Whiteboard Friday. This week we're going to chat about some old school SEO practices that just don't work anymore and things with which we should replace them.
Let's start with the first one — keywords before clicks.
Look, I get the appeal here. The idea is that we've done a bunch of keyword research, now we're doing keyword targeting, and we can see that it might be important to target multiple keywords on the same page. So FYI, "pipe smoking," "tobacco smoking," "very dangerous for your health," not recommended by me or by Moz, but I thought it was a funny throwback keyword and so there you go. I do enjoy little implements even if I never use them.
So pipes, tobacco pipes, pipe smoking, wooden pipes, this is not going to draw anyone's click. You might think, "But it's good SEO, Rand. It's good to have all my keywords in my title element. I know that's an important part of SEO." Not anymore. It really is not anymore an important . . . well, let's put it this way. It's an important part of SEO, which is subsumed by wanting to draw the clicks. The user is searching, they're looking at the page, and what are they going to think when they see pipes tobacco, pipes, pipe smoking, wooden pipes? They have associations with that — spammy, sketchy, I don't want to click it — and we know, as SEOs, that Google is using click signals to help documents rank over time and to help websites rank over time.
So if they're judging this, you're going to fall in the rankings, versus a title like "Art of Piping: Stunning Wooden Pipes for Every Price Range." Now, you're not just playing off the, "Yes, I am including some keywords in there. I have 'wooden' and 'pipes.' I have 'art of piping,' which is maybe my brand name." But I'm worried more about drawing the click, which is why I'm making this part of my message of "for every price range." I'm using the word "stunning" to draw people in. I'm saying, "Our collection is not the largest but the hand-selected best. You'll find unique pipes available nowhere else and always free, fast shipping."
I'm essentially trying to create a message, like I would for an AdWords ad, that is less focused on just having the raw keywords in there and more focused on drawing the click. This is a far more effective approach that we've seen over the last few years. It's probably been a good six or seven years that this has been vastly superior to this other approach.
Second one, heavy use of anchor text on internal links.
This used to be a practice that could have positive impacts on rankings. But what we've seen lately, especially the last few years, is that Google has discounted this and has actually even punished it where they feel like it's inappropriate or spammy, manipulative, overdone. We talked about this a little in our internal and external linking Whiteboard Friday a couple of weeks back.
In this case, my suggestion would be if the internal link is in the navigation, if it's in the footer, if it's in a sidebar, if it's inside content, and it is relevant and well-written and it flows well, has high usability, you're pretty safe. However, if it has low usability, if it looks sketchy or funny, if you're making the font small so as to hide it because it's really for search engines and not for searchers and users, now you're in a sketchy place. You might count on being discounted, penalized, or hurt at some point by Google.
Number three, pages for every keyword variant.
This is an SEO tactic that many folks are still pursuing today and that had been effective for a very long time. So the idea was basically if I have any variation of a keyword, I want a single page to target that because keyword targeting is such a precise art and technical science that I want to have the maximum capacity to target each keyword individually, even if it's only slightly different from another one. This still worked even up to four or five years ago, and in some cases, people were sacrificing usability because they saw it still worked.
Nowadays, Google has gotten so smart with upgrades like Hummingbird, obviously with RankBrain last year, that they've taken to a much more intent- and topic-matching model. So we don't want to do something like have four different pages, like unique hand-carved pipes, hand-carved pipes, hand-carved tobacco pipes, and hand-carved tobacco smoking pipes. By the way, these are all real searches that you'll find in Google Suggest or AdWords. But rather than taking all of these and having a separate page for each, I want one page targeting all of them. I might try and fit these keywords intelligently into the content, the headline, maybe the title, the meta description, those kinds of things. I'm sure I can find a good combination of these. But the intent for each of these searchers is the same, so I only want one page targeting them.
Number four — directories, paid links, etc.
Every single one of these link building, link acquisition techniques that I'm about to mention has either been directly penalized by Google or penalized as part of an update, or we've seen sites get hit hard for doing it. This is dangerous stuff, and you want to stay away from all of these at this point.
Directories, well, generic directories and SEO directories for sure. Article links, especially article blasts where you can push an article in and there's no editorial review. Guest content, depending on the editorial practices, the board might be a little different. Press releases, Google you saw penalized some press release websites. Well, it didn't penalize the press release website. Google said, "You know what? Your links don't count anymore, or we're going to discount them. We're not going to treat them the same."
Comment links, for obvious reasons, reciprocal link pages, those got penalized many years ago. Article spinners. Private link networks. You see private and network, or you see network, you should just generally run away. Private blog networks. Paid link networks. Fiverr or forum link buys.
You see advertised on all sorts of SEO forums especially the more aggressive, sketchy ones that a lot of folks are like, "Hey, for $99, we have this amazing package, and I'll show you all the people whose rankings it's increased, and they come from PageRank six," never mind that Page Rank is totally defunct. Or worse, they use Moz. They'll say like, "Domain authority 60-plus websites." You know what, Moz is not perfect. Domain authority is not a perfect representation of the value you're going to get from these things. Anyone who's selling you links on a forum, you should be super skeptical. That's somewhat like someone going up to your house and being like, "Hey, I got this Ferrari in the yard here. You want to buy this?" That's my Jersey coming out.
Social link buys, anything like this, just say no people.
Number five, multiple microsites, separate domains, or separate domains with the same audience or topic target.
So this again used to be a very common SEO practice, where folks would say, "Hey, I'm going to split these up because I can get very micro targeted with my individual websites." They were often keyword-rich domain names like woodenpipes.com, and I've got handmadepipes.net, and I've got pipesofmexico.co versus I just have artofpiping.com, not that "piping" is necessarily the right word. Then it includes all of the content from all of these. The benefit here is that this is going to gain domain authority much faster and much better, and in a far greater fashion than any of these will.
Let's say that it was possible that there is no bias against the exact match domain names folks. We're happy to link to them, and you had just as much success branding each of these and earning links to each of these, and doing content marketing on each of these as you did on this one. But you split up your efforts a third, a third, a third. Guess what would happen? These would rank about a third as well as all the content would on here, which means the content on handmadepipes.net is not benefitting from the links and content on woodenpipes.com, and that sucks. You want to combine your efforts into one domain if you possibly can. This is one of the reasons we also recommend against subdomains and microsites, because putting all of your efforts into one place has the best shot at earning you the most rankings for all of the content you create.
Number six, exact and partial keyword match domain names in general.
It's the case like if I'm a consumer and I'm looking at domain names like woodenpipes.com, handmadepipes.net, uniquepipes.shop, hand-carved-pipes.co, the problem is that over time, over the last 15, 20 years of the Web, those types of domain names that don't sound like real brands, that are not in our memories and don't have positive associations with them, they're going to draw clicks away from you and towards your competitors who sound more credible, more competent, and more branded. For that reason alone, you should avoid them.
It's also that case that we've seen that these types of domains do much more poorly with link earning, with content marketing, with being able to have guest content accepted. People don't trust it. The same is true for public relations and getting press mentions. The press doesn't trust sites like these.
For those reasons, it's just a barrier. Even if you thought, "Hey, there's still keyword benefits to these," which there is a little bit because the anchor text that comes with them, that points to the site always includes the words and phrases you're going after. So there's a little bit of benefit, but it's far overwhelmed by the really frustrating speed bumps and roadblocks that you face when you have a domain like this.
Number seven — Using CPC or Adwords' "Competition" to determine the difficulty of ranking in organic or non-paid results
A lot of folks, when they're doing keyword research, for some reason still have this idea that using cost per click or AdWords as competition scores can help determine the difficulty of ranking in organic, non-paid results. This is totally wrong.
So see right here, I've got "hand-carved pipes" and "unique wooden pipes," and they have an AdWords CPC respectively of $3.80 and $5.50, and they have AdWords competition of medium and medium. That is in no way correlated necessarily with how difficult they'll be to rank for in the organic results. I could find, for example, that "unique wooden pipes" is actually easier or harder than "hand-carved pipes" to rank for in the organic SEO results. This really depends on: Who's in the competition set? What types of links do they have and social mentions do they have? How robust is their content? How much are they exciting visitors and drawing them in and serving them well? That sort of stuff is really hard to calculate here.
I like the keyword difficulty score that Moz uses. Some other tools have their own versions. Doctor Pete, I think, did a wonderful job of putting together a keyword difficulty score that's relatively comprehensive and well-thought through, uses a lot of the metrics about the domain and the page authority scores, and it compensates for a lot of other things, to look at a set of search results and say, "This is probably about how hard it's going to be," and whether it's harder or easier than some other keyword.
Check out your keyword difficulty now
Number eight — Unfocused, non-strategic "linkbait"
Last one, some folks are still engaging in this, I think because content strategy, content marketing, and content as a whole has become a very hot topic and a point of investment. Many SEOs still invest in what I call "nonstrategic and unfocused link bait." The idea being if I can draw links to my website, it doesn't really matter if the content doesn't make people very happy or if it doesn't match and gel well with what's on my site. So you see a lot of these types of practices on sites that have nothing to do with it. Like, "Here are seven actors who one time wore too little clothing." That's an extreme example, but you get the idea if you ever look at the bottom ads for a lot of content stuff. It feels like pretty much all of them say that.
Versus on topic link bait or what I'd call high quality content that is likely to draw in links and attention, and create a positive branding association like, "Here's the popularity of pipes, cigarettes, electronic cigarettes, and cigars in the U.S. from 1950 to today." We've got the data over time and we've mapped that out. This is likely to earn a lot of links, press attention. People would check it out. They'd go, "Oh, when was it that electronic cigarettes started getting popular? Have pipes really fallen off? It feels like no one uses them anymore. I don't see them in public. When was that? Why was that? Can I go over time and see that dataset?" It's fundamentally interesting, and data journalism is, obviously, very hot right now.
So with these eight, hopefully you'll be able to switch from some old school SEO techniques that don't work so well to some new ways of thinking that will take your SEO results to a great place. And with that, we'll see you again next week for another edition of Whiteboard Friday. Take care.
One thing related about "Using CPC or Adwords' "Competition" to determine the difficulty of ranking in organic or non-paid results".
I totally agree that that metric is useless in SEO.
However, there's another "metric", which can be really useful also for understanding trends, and that is the "average cost per click".
In fact, many times I was able to discover new trends in different markets (eg.: travel) where keywords with apparently small clicks numbers start to rise in their CPC, and then result in becoming new mainstream and demanded products, services and, in the case of travel, destinations.
Therefore, regularly seeing the trends of CPC for your organic keywords can be a useful way for understanding what is going to be searched, hence - if we are able to detect the rising trend in time - a way for creating content or even specific campaigns before your competitors, obtaining so a quite evident competitive advantage.
Interesting metric - do you use the AdWords API to track down new trends systematically?
Nicely put Gianluca, totally agree with your point about the CPC metric to build the upcoming content crafted before the others do it.
On the money with the CPC trends I think many can get the wrong idea from the other parts of the data that can come from the tool because of it dealing with the money aspect of the advertising, but when you can see a trendy word that people are latching on then you can make that drive organically. I personally love to use Google Trends for this but after seeing your post I think your approach would be more of a soild source for finding high trending words.
Helpful tip!
Regarding the CPC competition, I honeslty don't know how people started using CPC competition to judge SEO competition. It's pretty ridiculous to think of using this for SEO. The funny thing is that this is pretty widespread.
This is useful. When it comes to keyword targeting for content (blogs) I tend to pull in the traffic data from AdWords Keyword tool and then jump over to MOZ's keyword difficulty tool to assess the 'competition'. I then traffic light these two columns in Excel and if both come up green then that is a keyword to potentially target.
I haven't used ACPC in that method, but will try it out as a third metric for research. How long do you test ACPC for before determining that a keyword is a trend, Gianluca?
Whilst I would agree with a lot of, if not most of, what you've said, to suggest that PBNs are no longer are effective is simply not true. It may be black hat, it may be "unethical" (if one believes in SEO ethics), it may be a huge risk to any legit company out there, it may be a lot harder than it used to be, but even with all these caveats it cannot possibly be stated they are no longer effective if carried out properly.
Just because something is harder than it was, or is black hat and is definitely not recommended for long term sustainable growth, doesn't mean it isn't effective.
Fair enough. I suppose it depends on your definition of "effective." In my view, a tactic that sometimes, if all the stars align and you guess right and aren't suckered into a scam or the wrong path, works for a few weeks or a few months (and yes, sometimes, even a year or two), but then gets your site (or all your sites) blacklisted in Google is not an "effective" tactic.
I won't argue, though, that some private networks have shown themselves to improve rankings in the short term, and there's even the extremely rare few that have maintained that for longer. But the folks watching Whiteboard Friday tend to be working with mainstream brands and sites that cannot afford to invest in tactics that have a high percentage chance of harming their long-term potential.
Also notable - we've seen PBNs and their ilk be more effective in developing markets vs. developed markets, so if you're considering it, that's probably worth factoring into your calculus. Generally speaking it'll be easier in, for example, Bulgaria than the US to find examples of manipulative links that still seem to work.
PBNs are efficient in less competitive parts of Google, just like you said. By the way, I know a some very good SEO specialists from Bulgaria, Russia, Ukraine and Eastern Europe in general, if anyone is interested in comparing what tactics work/don't work in "western"/"eastern" SEO hit me up, I'd love to see what this exchange of experience can uncover for both worlds.
We appreciate the whole content. You have mentioned the seo techniques those are not any more applicable, It would be perfect if you can share the the right techniques those are applicable.
so your basically saying this isnt really about SEO but about white hat linking stragtegies that only big brands can really afford ? which is another way of saying moz is really that relevant in the SEO world?
I did a manual analysis of the top 100 most visited websites a while ago and only 3 of them had keywords in their domain name at all, so I see your point, however for certain businesses (e.g. a van man in New York or locksmith in London) where there's really only one or two keywords that you're targeting having a keyword in the domain name is a very good idea and the first page results of Google are full of keyword rich domain names.
This is because there's still a small SEO benefit, it means organic links are more likely to include the keyword and because most of your customers will only use your services once, so your domain name should say what it does on the tin, so to speak, as that's likely to increase click through rate.
If you have a big business and are looking for long-term traffic and customer loyalty, it's a lot harder to gain that by having some-keyword-rich-domain-name.com as your domain vs brand.com, as the second is a lot more memorable and big businesses tend to target multiple keywords anyway, so it's unlikely that you'll be targeting a single keyword.
If you plan on expanding in the future, it's also important not to choose a keyword rich domain name (if Amazon was called books.com, they'd have a lot more problems selling the vast range of products they sell today, for example).
A question like 'Which domain name should I choose' has to factor in more than just SEO.
I agree with you, Martin. On the one hand it's good to have a domain name that includes / refers to your business name so that people would know it's you and would click when they see, but on the other hand I saw a rise in generic keywords domain names that "steal" our traffic and clicks simply because they contain these keywords in their domain name. Not to mention the long titles (more than 100 characters) and spammy long descriptions, many of which a human can clearly see they are generated for each product.
I wonder how much of this is left-over (i.e.: not cleaned up according to what Google is attempting), and/or the systems just not being smart enough to implement what Google would like (theory vs reality)? While I'm a relative newbie at SEO considerations, I seem to run across fairly bad sites, with poor content, ranking highly with keyword domains, and/or relatively meaningless (in terms of quality) links to their site (especially in smaller local markets).
My conclusion, so far, is that the algorithms and AI aren't what they are cracked up to be. Though, I totally agree that it's best to do it right, and prepare for the future, when they might get that fixed.
Along with PBN's, multiple (ie, more than one) EMD or partial match domains can still be incredibly effective tools, especially when positioned as review sites. Now, it requires actually taking the time to have independent 3rd party reviews, but having a number of these have proven effective for the long term for several very competitive niches I work with. Not just ranking, but sending great referral traffic and serving as a fantastic link source that competitors can never have. This may be gray hat, but it is effective. Great article by the way! I am certainly weary of most PBN's that are "auto pilot" programs or for sale as programs...
I won't argue that some EMDs sometimes work, but on the whole, as Mozcast has measured them in our datasets, their presence has dropped from over 6% a couple years ago to less than 3.3% today (almost cut in half). I'd say that's pretty clear evidence that the last few years have been very unkind to EMDs, and from a branding perspective, it's a nightmare to try to earn editorial links or press coverage to them (as many poor agency and consultant SEOs know).
exactly true but not really who moz target to , ie real marketers working with the billion or so small to medium size companies
I somewhat agree with what Martin is saying on the PBN topic. If you believe what you are doing is border line sketchy, then it is probably black hat. On the other hand, if you have a "Network" of sites that are all branded, properly written, educational, compelling and offer their own unique set of benefits, how would this be Black Hat?
This may be considered black hat if all of these sites had home page anchor rich blacklinks to one site, absolutely. What if, these sites only link out ONLY when absolutely relevant (to internal pages for example) and provide users with a benefit from that link.
Im not saying that this is the BEST tactic, but I somewhat agree with Martin, that it is still a tactic, but on what level 5 sites, 10 sites, 50 sities? And maybe, I am completely wrong in my thought process here, but I don't think there is anything wrong with (4-5) sites that provide there own set of benefits that may happen to link both internally and externally when natural and beneficial.
I think that would fall under the "you could have spent the time you spent getting 4-5 websites (or 40-50 god forbid) to prominence making 1 site truly amazing" argument. I get that sometimes link building is hard, but so hard you'd rather try to build a whole new site just to link to your other one, and then you have to build up that site's branding and link equity and ranking signals and content? Just feels like, if not a total waste, at least a vastly inefficient way to accomplish the goal.
This video is basically every argument I have with my managers and clients who have "read up on SEO". Gonna just start sending them the link to this video. Shared!
I think exact-match domains still works since branding isn't entirely exclusive to fancy unique words or names. And it's effective to the general users who doesn't have much idea about brands and still consider keywords as way to know that the website really sell what they look for. Say, someone who is new to internet who is looking for a used pipe will probably click on the website that says sellingusedpipes.com than a branded website like oldchaps.com
To be on safer side, one should use exact match domain only if the brand name associated with the website is a unique. I had two different experiences with exact match domains. In first, I was able to rank a website webdesigncityname.com (without any brand name) for almost all the web design and development searches related to that city. In the second, where an educational institution website was ranking in top 5 for a particular course, created a new website with exact match domain around that course, but this new website (even after a lot of work and good content) never came on first page for same keywords.
There is no universal rule, but It's best to keep the domain name same as brand name, if everything else is equal, a brand name url will always rank better than the exact match domain.
One factor that cannot be ignored with exact match domains is that they do seem to get higher than average CTR than (unknown) branded names for searches that are a match, or close to it. They then often convert that traffic at high rates too, from what I've seen. Eyeballs do seem to like them if they aren't searching for any brand in particular.
Huh. We've observed the opposite. Do you have any data at scale or reports that have shared that? All I've seen lately are the reports showing how brand influences CTR, conversion, and retention positively. e.g. http://searchengineland.com/microsoft-research-sea.... It's also a nightmare trying to get a trademark with a generic domain name https://www.techdirt.com/articles/20100805/1810141... and the search engines have been pretty public about this topic too: http://searchengineland.com/bing-its-a-myth-that-k...
Hi Rand, no meaningful data to back up my observations. Plenty of debate on this topic it seems, but no really convincing data to confirm one way or the other. What data do you have? The Microsoft research you cited seems to be more focused on relevance than brand, which correlates with my own observations. The inability to secure Trade Mark on generic's is a limitation I agree, but then it rather depends on how you plan to utilise a domain. If you view it as simply an address, then generic's provide a good location to do business, plus they're unique. Provided you secure a TM on your main brand/product, as you should, then surely the domain you use becomes less relevant. I always saw the boost that generic's received in the SERPS as artificial, useful, but liable to change and did. It was the high relevance to the related search terms that always felt like the real, inherent, benefit of generics. I don't see generic addresses as necessarily the best solution, but certainly still a very viable one.
Thanks for the resources! I'm researching the impact of including keywords in a domain name for both PPC and SEO. Have you seen any studies/evidence to suggest that users are behaving similarly (selecting non-exact or partial match domains) for paid search results in addition to organic SERPs?
I read the The Microsoft study (http://ilpubs.stanford.edu:8090/1021/1/wsdm12-doma...) which "demonstrates that users have become reliant on domains in assessing the relevance of search results," which focuses on organic results. Can you think of any resources examining this issue from a paid search perspective?
I think you can have the best of both worlds. If the brand is "Smoke Town": smoketownpipes.com. You get your most valuable keyword in there, and you still have a trustworthy branded domain that won't turn off customers, linkers, and reporters. Plus, it tells people that Smoke Town is the place to go for pipes.
EMD can be still effective. I created and managed as editor in chief a website / online magazine that had generic keyword in domain name. It wasn't me who have chosen this domain - it was my job to make it work. It was "mobile marketing" with not that small competitien due to category pages on blogs, other magazines, offers/services/pricing pages etc. It's exactly markeitngmobilny.pl ("marketing mobilny" = "mobile marketing" in Polish). I've run link building tactics like media patronage, guest blogging, expert's opinions and stuff. Links were based on image (logo/banner), domain name (marketingmobilny.pl) and keyword "marketing mobilny". For over a year it's incessantly (as far as I know; I'm not working with that project anymore now) no.1 in Polish Google SERP's for that keyword. Wikipedia is no. 2.
According to me articles and directories submission in SEO will be no longer effective, cause nowadays people are focusing on Blogs. I have noticed that whether we submit good and unique content or not when any search occur on search engine all result come from blogs never from article site. So what is your thinking on it...
Hey Rand, as always wonderful WBF.
In case of strategic link bait, evergreen content like how to, controversies always help but this also depends on which industry you belong to.
I had seen that a lot of folks are still discussing the benefits of link building techniques like directories, article, etc. I have a confusion/question Rand regarding quality directories. In my according making your site listed in directories is a proper way and also sharing your knowledge via different article sites to help others as much as can is also a valid factor then why Google is treating it as spam or making it penalized. So my question to you Rand is what your opinion is regarding: we should fully stop submitting articles and directories.
BOOM - You nailed it Again - Fantastic Fantastic Fantastic.
I would love to meet you , sit down, and talk all this through - and really SMASH these BIG issues: exact match domain, link building, micro sites - I am sure it would go on for ever.
I love how you ALWAYS bring up such pertinent points that are probably on all our minds EVER DAY - I agree with your do-it-the-right-way ideas, but not necessarily the practicality of it from an agency or business perspective.
Here is my take on it all:
1) Keywords Before Clicks - Yes - but people do click on keywords as that is how they are searching - I would be careful about this. Google even tells us to keyword load Google Ads for Quality Score.
2) Anchor Text - agree but the web/Wikipedia is made up of anchor text keywords, and from our data Google is absolutely lapping up internal links as it helps them find out what the site is about.
3) Page for every keyword - again you want to give people what they are looking for to keep them engaged and keep conversions up - I feel people these days are really looking for long-tail keywords and pages/products relevant to these searches. Also vital for CTR and Quality Score on Adwords.
4) Link Building Techniques - Not going there really, but I am sure you could come up with good examples for pretty much all these. Directories for Local Search/G+ and press releases actually running the whole news industry (pre-Google).
5) Totally Agree
6) Exact Match Domains - quite frankly I don't agree on this and Matt Cutts "turned" down the power on these a few years back and then had to "turn it back up" because obviously people had keywords in their domain naturally not just for SEO. e.g. London Taxis
7) Google Adwords Data is pretty weird anyway - how come many of the keywords have the same volume which doesn't even match the natural figures we get from analytics. Also as the paid listings are now dominating the search space - so any competition data on this id obviously pretty relevant.
8) Linkbait - Totally Agree, and I never even knew that went on - pretty sad - any Linkbait is pretty interesting.
Anyway I am looking forward to a heated discussion about this in our office in 15 minutes.
Ciao
1) Sure - I'm not suggesting we no longer use keywords at all, just that we shift the focus from "jam every keyword in as much as possible" to "use keywords intelligently and try to draw in the click more than game the engines' keyword preferences."
2) More detail in this Whiteboard Friday, but yes, generally we agree: https://moz.com/blog/linking-internally-externally...
3) I'd be real careful here. I agree there's value in targeting long tail keywords, but disagree that every variation or potential unique search needs its own landing page. Google is far better today at ranking pages that serve the searcher's intent even when the keywords aren't a perfect match, and the strategy you describe is often responsible for folks getting hit by Panda for having lots of similar or low quality content or having the wrong page(s) rank for the terms/phrases they're targeting. You also don't get the benefit of combining link signals and other ranking signals all on the same page, which can be huge.
6) See my answers above. Basically, EMDs have gone down by almost 50% in Google's first page of results the last couple years. No matter what Matt said in 2012/13, it's a different story in 2016.
7) Check out - https://moz.com/blog/google-keyword-planner-dirty-... - that's why :-)
Cheers for getting back to me here Rand; totally missed it at the time.
Really interesting the WBF the other week - https://moz.com/blog/weird-crazy-myths-about-link-...
and how that compared to our old conversation. Looks like the mild-SEO-scare-tactics of early 2016 are being relaxed. I suppose the idea was to prevent the ream spammers - black-hatters. At Relevance we just want publicity, visibility and promotion for our client's - usually that helps their SEO at the same time.
Are you still sure about your data on EDMs (-50%) - not x-x-x keyword-stuffed EDMs - but actual relevant company names that just happen to be EDMs EDMs.
Would love to have a coffee with you one day...buzz me if you are in London, Monaco, or New York.
Thanks for the updates Rand. Day by day, Google release new updates for unnatural link building and I'm ok with that since time spending for unnatural link building like you mentioned (comments, directories, paid links etc.) is waste of time. Sometimes, we continue to do that as a habbit, but it become more dangerous by the time (thanks God).
I know the comments for this WBF would be good and insightful.
Missing from the list: keyword density, text to html ratio, and obsessing about pagespeed.
Effective in select markets/verticals where you may not get caught (but risky): PBN
This has been really insightful.
I am still not convinced by the CTR as a metric especially concerning titles. It seems that in some of the Google Webmaster Hangouts, the CTR did have a temp effect but then the system made a correction. Should I put on my PPC hat and rewrite my titles for interest/clicks or keep them really informative?
Feel free to test, but in my/our experience, titles that balance keyword use and driving the searcher to want to engage have the best results.
To put some effort and time into the title tags is really worth it. I would even suggest to test it with colleges what title alternatives are most appealing. Or even make surveys. Especially when it comes to the title tag of the main page.
Thanks for the tips on what to avoid. As the owner of my business, and it's SEO as well, some of these are low hanging fruit a that have appeal. But just as in life, the effort is commensurate with the reward
I expect paying $100 for a link bundle or whatever has a negative impact on my rankings for sure!
Concerningly the practice of paid article links is becoming more and more prominent with more authoritative publishers, not just bloggers. The industry calls this 'Native Advertising' or 'branded content' but those with an SEO eye see it as the exact same method of obtaining paid follow-links but with a bigger budget. I've worked on a few 'native' pieces with a $20k budget and for that price, you'd want a follow-link to be included in the package.
Rand, what does Google think of 'Native Advertising' when it comes to brands obtaining paid links from authoritative publishers? If Netflix pays Wired $50k to do a native piece and includes a bunch of follow links to Netflix's site - is this seen by Google as paid link manipulation from an authority site? It seems like it would, but I haven't seen any evidence of a big publisher being hit with a penalty for 'native' content and links. Thoughts?
Hey Rand,
Your piece does an exceptional job highlighting ineffective SEO strategies. As you’ve mentioned, trying old techniques can cost a domain owner’s SERP and search efforts.
I also liked how you gave a detailed narrative on the true value of keyword selection in today’s SEO world. Something I noticed, however, is that keyword research is still a main driver of effective SEO--it allows domain owners to differentiate competitive terms from those that drive desired traffic.
I actually did some research on this topic and other effective SEO methods--which can be found here: http://www.insegment.com/blog/5-ways-create-seo-juggernaut/
Where should we go and what should we follow, read this http://searchengineland.com/improve-internal-linki...
Conflict these these tow ideas.
I don't see a conflict between this piece and Paul's. He's talking about flowing link equity internally on a site, but not about abusing internal anchor text links. What I would say is that internal PageRank is a relatively small influencer of rankings, but can have a positive impact on large websites where indexation isn't great.
Interesting article and information on linking structure. More or less that article is an entire over on internal site architecture and how your links should be placed and where they should link to. This post, on the other hand is more or less and overview on what used to what and whats working now. Or did i miss the real question?
It's a great article overall, but I personally think that the title is a bit misleading. Perhaps a better title will be "8 Old School SEO Practices That Are No Longer RECOMMENDED". If you say it's "not effective", it is down right inaccurate. They are still very effective for most - Check Out This Article For The Real Test Results...
For #3 - Pages for each keyword variant. I have a local sites with 100+ pages of duplicate contents, the only different is slight variation on keyword "locksmith [different area names within same city]", more than half of all pages rank top #3 position, almost all on page 1, and been there for at least 2 years now, still kicking ass.
For #4 - Paid link. I paid most of the links, including using my own dedicated PBN. There's no way Google algorithm can differentiate paid or organic backlinks, only if your site get flagged, and gone through the manual review, or the paid networks are busted. I did have a few PBN sites get deindexed & banned, but the linking money sites are not penalised and still kicking ass right now. So paid links is EFFECTIVE (that's why Google don't want you to buy them), but a RISKY business, I should put it that way.
For #5 - Multiple sites targeting same audience & keywords. In one of my niche, I am currently dominating Google first page with 6 websites, 1 website ranks for 2 pages (on position #1 & #2), that's mean only 3 of my competitors are on first page of Google. So this is super effective and kick ass, because Google can't tell they are owned by same person.
Above are 3 points that you mentioned as "not effective" is indeed very effective for me, at least for now. I considered what I am doing is a risky business, but it does kicking ass, more importantly, save me lots of money and time, compare to other methods that people call it "white hat" SEO.
To follow Google's propaganda of SEO, it is super time consuming (although it is long term I agree), and will cost me a bomb that only BIG company can afford. You spent trucks load of money for good contents to bait for backlinks, another fellow can still beat your ranking easily with some cheap paid backlinks. This is the fact, and I choose to be a second fellow.
So, Keywords in the title with Pipe symbol is dead So, now we have to consider High CTR titles right?
Link Building old school is dead. But, still most of the people are using PBN's to get high rankings.
Now, my question is that if these are the things which are not working. Then what are the things which are working in the seo right now according to you. Is there any white board on that so that it become clear on what to do and what not to do.
Very good advice as always on WBF.
If you want to positioning a web on international market, what do you think is a better tactic? one website with different subdomains (one per each language) or one different web with its own domain (.co.uk, .fr, de, etc) and interconnect these website ( a link for each url to the other urls in the others websites)
Keep shilling for Google, trying to scare people into believing that PBNs dont work. Meanwhile, sites are killing it on the backs of links from PBNs.
Tesla - my guess is that you sell links on PBNs (which is fine -- nothing illegal or unethical about it IMO as long as you're upfront with buyers about the risks). But if it was the case that you were simply profiting and ranking from them, logic would dictate you'd be A) thrilled to have sources like Moz claiming they don't work, as it would keep away your competition and drive down demand which would lower prices and B) very reticent to challenge their effectiveness publicly since every message against them works in your favor.
Maybe I'm wrong and you just want folks to know they're effective? If that's the case, I'd love any links to case studies or examples, or even some sites/pages and queries that folks could go investigate to see for themselves?
I'm not saying, BTW, that link networks never work -- just that they are frequently penalized, highly risky, and it's often very hard for marketers inexperienced in their operations to judge the ones that will hurt them vs. ones that might provide a temporary boost (before the eventual penalty). I'd also point out that most folks reading Moz don't operate in the churn+burn world, and need strategies that will work long term and don't expose their employers/clients' websites to severe risk.
Thanks for responding Rand. I appreciate it.
Funny how there are always two assumptions with this:
The reality is that the most successful PBNs don't monetize their networks by selling links. That is why the "P" stands for - "Private", and the "N" stands for "Network". They are also not frequently penalized. Only those stupid enough to not stay under the radar get caught. Those would also be the ones that buy and sell links. If you are selling links, your PBN is no longer a secret. For those fools, the "P" stands for "Public".
My issue with your perspective on this is that you also assume, or infer, that PBNs are short term solutions. They are not. The true private networks rarely get caught because Google's algo can't catch them. These are not like Blogvertise back in the day with a booth at PubCon or SES hanging a banner that says "Buy Links Here".
Im not going to publicly out anyone here, but you have my email. Send me a quick email and I will be happy to give you an example of a small PBN that powers a network of sites that kills it, and has for 4-5 years. For the ones like this that really kill it, the "P" stands for "Personal".
My intention here is not to educate the masses on how to build a private link network that works, but to keep poking Google in the eye over this crap. They scare the hell out of innocent mommy bloggers over nofollowing links by making examples of a few, and indefinitely punish others with the flustercluck known as Penguin, while you and I can build a real PBN and NEVER get caught.
Sounds like we're certainly on the same side of the issue. I hate seeing folks get hurt by supposedly "private" networks (which I get that you're saying are different from the ones that are truly *private* and operated only by a sole individual or company). That said, I've seen plenty of those *private* networks take hits, too. A couple years back, there was a fairly infamous shutdown of whole loads that were using patterns Google seemed to recognize (centering on fake/manipulative blogrolls if I recall), and even today, the more aggressive discussion forums of the SEO, affiliate, and internet marketing worlds have plenty of threads claiming similar on a semi-regular basis.
If you ever wanted to write, anonymously or pseudonymously, and give Google a harder time about this with some examples, I think that would be pretty sweet :-)
p.s. one area we disagree is Penguin. I think Google's done a pretty good job of closing off many of the manipulative/spammy link avenues that previously carried no real penalty and now have serious consequences.
I believe we have more common ground than not, but your perspective is more optimistic than mine.
Blogrolls and footer links are bush league tactics in the same category as recip links. I do not put them in the same league as a PBN. The other thing is PBNs done right don't scale well without increasing risk. And the smart folks know this. Because it is so much more difficult now to get links within Google's fluid guidelines, many have effectively given up. SEO may not be dead, but for many, it's been abondoned as a viable option to drive traffic. That means fewer links can now win the war. Hence a PBN done right is like printing money. I also dont believe for a minute that Google has identified a single PBN via an algo. Those are all high profile manual actions meant to send a message.
As for Penguin, its a cluster because there has been no recovery for those hit. Its been an indefinite penalty - a Gitmo for webmasters - with no hope or end in sight. I dont think that is what Google intended, but clearly they havent figured out how to release the detainees.
Drop me an email. I will send you a case study.
When I read the words "get caught", it infers that you know what you are suggesting goes against what it is we are meant to do. When "getting caught" is an option, I'd put my money on Google, eventually at least, catching.
Keep thinking that. Your more sophisticated competitors thank you.
Absolute rubbish about the keyword rich domain names,i still use them and have quite a few sites on page 1 some for pretty big search terms,and they out rank big authority sites and some i have recently used which have also gone straight in at page 1 or 2 so to say they have very little benefit isn't true. I always use unique informative well written content for the sites and make good sales to go along with it. I suppose if your goal is to have a large recongnised brand then it may work for you,but like 90% of the people doing any kind of website marketing they want to earn money.So do your own thing dont be a sheep like most are, and don't believe half of what these so called gurus tell you.. it still works.
Rand,
That was an excellent point about EMDs giving off a spammy and less trustworthy perception. I notice that I skip over those types of domains in the SERPS, clicking on the branded domains more often. The spammy association is particularly the case of domains that have hyphens.
Regarding keyword rich domains, you mentioned there was possibly some benefit due to the anchor text pointing to the site.
However, couldn't that aspect also hurt them in regards to having an over-optimized backlink profile? Where the keyworded domain is causing them to have a lot of anchor text of "cheap red widgets" if their domain is cheapredwidgets.com. Or do you think that would that be considered a branded anchor text?
Thats a pretty good question Jeff,
"having an over-optimized backlink profile? Where the keyworded domain is causing them to have a lot of anchor text of "cheap red widgets" if their domain is cheapredwidgets.com. Or do you think that would that be considered a branded anchor text?"
I would say that it would be considered a white hat profile because of the associated domain name, but again, very good question and would love to hear some other thoughts on this.
Great stuff Rand. Not only did you discuss tactics that USED to work but more importantly you offer an opinion on what marketers should be doing instead!
Maybe next week you could do '8 New SEO practices that ARE effective'? I don't practice all of the above but I do believe in 'blogger outreach' (paid article links) if done in a non-biased way with sacrificial links... However if I'm wrong then please let me know and I'll start re-writing my CV :-)
Sacrificial links? Like... You outreach to the blogger, offer to pay, but the link is pointing to a domain/page you can sacrifice if they spam report you? Seems like risky and time-consuming work vs. potentially investing in more long-term editorial link earning practices, but I suppose depends on your niche, risk tolerance, and focus.
I do like that WB Friday idea! Thanks :-)
Hello, I have a few questions.
First, regarding the Heavy Use of Anchor Text on Internal Links. I have a help subdomain on our site and I use anchor text on it quite a bit to link over to other relevant pages in our help site. How much of this is too much? And what would be a good alternative to this? (Should I instead list the full links at the bottom?)
Next, subdomains. Again... how many are too many? I have subdomains for the above-mentioned help and for our blog. (Both of those are WordPress, the rest of the site is not.)
And finally, about the directories/press release links you mentioned... I'm not sure I understood that part. Are you saying that having links to directories is bad? Or that having press releases at all is bad? We have press releases on our site, and most, if not all of them, have been submitted to PR newswire for larger release, however we don't link to anything external. We simply publish the content of the release. (And admittedly, we've not been good about keeping up with getting new releases out.) Is this hurting us?
Re: how much - see this video for more depth on that: https://moz.com/blog/linking-internally-externally...
Having links from many types of "SEO-focused" directories or from press release sites (especially the more sketchy ones, but even some of the best known ones) can be high risk. More here: http://searchengineland.com/google-links-in-a-pres..., here http://searchengineland.com/pr-newswires-answer-go..., and here http://www.seerinteractive.com/blog/press-release-....
Hey Rand: Seems like a lot of what you state is based on people doing it incorrectly. How about a challenge that disproves all the old SEO theories and shows that what you are stating above actually works better?
This is great content but where is the data? Where is the data that shows what you are stating is true and that the old methods when done correctly are not working?
I gave Rand proof that private networks work when done right.
Such a great whiteboard rand. But I always see that most of SEO folks use these all old techniques. but a have a question that if i have 1 page static website with above 10 keywords then how should i manage in title?. because title size is limited.
Hey Rand, I think the list might be missing keyword density, what do you think? :)
If we're going to list things that are missing, it's probably a mile long! I don't think I could be comprehensive in less than a few hours (maybe days) if I tried to cover all the defunct tactics. That said, certainly agree that keyword density died out many years ago.
PBNs still work up to a point - but why would you want to risk it? As soon as that footprint is recognised it's going to be taken out and all the benefits will be lost and worse you'll end up with a long lasting penalty
If only that were true. But its not. So keep drinking the kool aid while your competitors kill it.
Again on the topic of exact match domains. I made a quick site called DragonMedical.ca, with the keywords dragon + medical, selling a product called Dragon Medical Practice 2. The site rose in rank to first page in a few weeks. I'm thinking because I used .ca instead of .com, Google is seeing me as the official site for the product in Canada, and in this manner I'm not competing with the US sites. The site in a couple of months generated over $200,000 in sales, driving me crazy that I can not duplicate the results in the US. Warning, once you get a successful site that makes it look easy, it is like a SEO drug, that will keep you up all night :)
In conclusion, I would recommend doing a country specific domain. I don't have the data to back this up, only that it worked for me on this project. Also it has to be a niche product.
Congrats! Watch out for trademark infringement though. Sometimes, companies will come after sites like that and, if they win the judgement, get ownership of the domain.
I might also point out that in this instance, you're also benefitting from the brand preference and Google's just assuming you're the brand vs. benefitting from an EMD for a generic/unbranded keyword phrase.
I was also worried about trademark infringement, but I contacted the company in question and downloaded their rules of engagement and also had a chat with their marketing team. (just to make sure everything was above board ) And yes, I'm riding on the edge of what I can do and what I can't with this page, concerning trademarks :)
Great article, unfortunately most of these actions are still being used for a lot of persons who named himself "SEO professionals"
I think it is funny how many 'big SEO agencies' are still relying on many of these outdated tactics...specifically the commenting/profile building.
However, I still own a ton of old websites that are EMDs and PMDs that rank on Page 1 going on 2-3 years now. But I haven't touched them either.
Maybe Google won't mess with older sites unless you're still playing with them...haha.
Thats really useful information Rand. Unfortunately, we were still using Keyword planner in Google adwords for this. I guess its time to stop estimating from that!...Thanks! With regards to the press release not being useful for SEO or google penalizing it, is it still a concern if we are using reputed companies such as PRWeb, Businesswire etc.
I completely agree with it, As many of our folks do not practice most of these activityies these days. But its good to have little refresher to avoid as such thought.
I have seen companies using a mix of both old and new tactics and they based the use of those tactics in google not revealing some their algorithm to guide us better in what we should be doing.
I think that we should focus on creating great content and great UX changing the thinking from a systematic perspective to a more user-friendly content.
I am so glad to watch this video.
I found this listing very depressing - those were the days my friend I thought they'd never end but then they did... Incredible how far search and ranking signals have come in the last 5-6 years in terms of closing the gap on all of the or most of the rather little cheats that existed back then. Internal keyword rich anchor text worked wonders for a time & required very little imagination or nuance.
Great post. I remember doing all this stuff (article submissions, directory submissions, keyword rich anchor text etc.), when I started working as a SEO trainee back in 2004.
But honestly I still see lots of SEOs suggesting these trick to clients and actually implementing them.
Hi Author,
After reading the fourth point, i Have a doubt. How to do SEO for same services targeting multiple location. For example, i am running a web design company and willing to target multiple locations world wide. whether i have to create different page for each location or i can use same page to target multiple location based keywords. As I am new to such concepts, please assist me. Eagerly waiting for your reply. You can also contact me on kovalanj23(@)gmail.com.
Nice post Rand,
Sadly I've realized I'm still doing the last one, I'm gonna have to change that! I think that it may happen on the older SEO companies, maybe the younger ones have already grown in the new techniques, so somehow they can escape this list.
Love how you're always giving us new ideas on your WBF!
Hi,
Interesting to read about the old tricks in the SEO book. I started with SEO only recently (couple of years) but i have to admit that i have been taught some of these methods by " SEO experts" in The Netherlands. for example #7 using adwords cpc to check the competition on keywords in the organic section. Also #4: getting backlinks by applying your website to as much directories as you can is something i heard more than once.
Ah well in the end in SEO you can only learn you so much by consuming every topic there is about SEO. I find that by doing SEO (and making mistakes) you will know what works and what not.
Greetings,
Gert
I think this is a great article but a little depressing for some who are still doing these things. For me, many of the tactics I don't use anymore. If I do I make sure they are done with the highest integrity and not over done. However, because of this, I'm wondering if a better whiteboard Friday would be:
SEO Practices That Are Still Effective- Whiteboard Friday
Link Building Techniques That Still Work after Hummingbird and RankBrain - Whiteboard Friday
Hello Rand Fishkin your post is very good. its beneficial information for my "Hindi Blog" Thank you very much for this information.
Hi,
What about the practice of social book marking? I've noticed it doesn't seem to do anything anymore but is it something that is penalized?
Rand, you look rather grand in your top hat. Unfortunately it's an unregulated industry and there is plenty of black, old SEO around. It really infuriates me to see how many digital experts have no clue nor support on-going professional development. You have to keep updating yourself, it's part of the job, not churn old advice... grrrr. This is my overview after reading a range of articles... you have to google my name and add how to seo as links are not allowed here.
If comment links don't work, why do they show up in Google Search Console? Seems like they are a kind of a human editorialized link. Or are they just there so you can disavow them???
Great, informative video, especially for those trying to revamp the SEO of an older, stale website.
I have a question regarding the first point touched upon.
I'll use your Tobacco Pipe example. If a person is trying to rank for "Handmade Tobacco Pipes", will they be penalized or not rank so well if they place other words in the string, such as "Handmade Burl Wood Tobacco Pipes" or "Handmade Stylish Tobacco Pipes"? Or is Google smart enough to pick the words out of the title/description that fit best, and understand the slightly conversational tone of the title with extra words put in?
Whole article is true! Confirmed!
Hi Rand!
Yes. Old techniques SEO is not that they are not working well (I personally give me still ...), but it is true that we must take into account more than ever that factors like choosing better and more segmented our keywords
The problem is the continuing evolution of Google algorithm. Let's finish crazy.
Great list. I would challenge your thought on exact match domain. I ranked (usually 1~3 on page 1) an EMD in the fitness industry (using geo + bootcamps) two years ago and with literally no effort it is still there. In fact, I was toying with rank and rent but gave it up as the client didn't want to rent. I left it alone and swapped my contact for clients and it goes between 1st ~ 3rd but never below 3. I should note that there are many local gyms trying to rank for their bootcamps but are not out ranking me. TBH, I was blown away as I ranked it after Matt Cutts started warning people about EMDs.
My thoughts are that EMDs are here to stay. Business (pre internet) routinely register legitimate businesses as city + service (New York Fries, L.A Haircuts, Chicago Pizza). Google is not going to penalize legit businesses for their ranking while at the same time they are not in the business of policing business licenses from website owners.
In summary: Write high quality content and forget about "quick fixes"
I would disagree somewhat with the exact match domains. Whilst I agree they don't look branded and they are honestly my biggest hatred about SEO at the moment, they work (provided the URL matches the brand name) and a lot of legit businesses change their names to Service + Area in the same way they changed their business names to AAA Service back in the Yellow Pages days.
I disagree on most of those pointers. I have any case studies that it still works. I have many websites that are ranked well for those SEO strategies. For example if for using variation of keywords. A website that whats to target 50 US states it's better to have a separate page. For example SEO NY, SEO NJ, SEO MA and etc. Each page for each state with unique content and keywords still works very well. That is the best way to target each state.
For the exact domain name I still see many domain name keywords ranking on top search results.
For internal linking that still works as long as you don't over do it.
Microsite still works. Same thing as linking from relevant websites. If the content of the microsite is unique and relevant that is fine.
Who really knows what Google wants other than Google. I agree with you on most of these points but some of these methods do still work. You must apply them properly, and this fact is based on the assumption that doing otherwise would not make anyone any money. It is costly and time consuming to do SEO 100% white hat. People cut corners because they have to, but solidifying a brand's reputation first will let them slip under Google's radar. Personally, I only use white hat, but I have known others who don't. They make more money than I do.
Great info Rand :)
Helpful video Rand!
On point no 5, how does it work for having multiple domains for different products vs having them as sub-domains? In our company we have been evaluating and not yet found a satisfactory answer.
Thank you.
Hi, Rand. Thanks for the information. I agree with you but not completely. I do have some questions to ask you rather to show you some data that proves these theory wrong by one of my competitors. How can I contact with you? Is there any email ID i should email you? Please let me know.
how about writing useful articles in several different domain name and different audience against writing in subdomain ? thank you
interesting points useless i guess! because that topic interests me, would that hurt my website. Thank you in advance Do meta keywords matter anymore?
seo sounds complicated now :(
Great post! Yes, you are right about Unfocused, non-strategic linkbait, Because without quality content your link has no value.And Press release without nay news, ignoring designing and creating thin content are the other SEO methods that we need to stop.
(Off The Topic)
The Green Ad label finally comes to india, but i have figured out why Google Made it Green! as of my view, they are actually trying to get more clicks on Google Ads, instead of organic results, as you can see clearly the URL color and The Ad label color is matching with each other, so its hard something in a hurry to identifie whether it is an ad or organic result.
So Google is clearly looking to increase their earning via Google Adword. this is the strategy of these Green label ads, i think
Hi Rand.
Great post. Loved your images, real effort must have gone in there.
WOW!!!! With so many NO NOs, what does one do today to get on Google top ranking?
Regards.
Veena
In the "old" days many people link from every place they could even if it was with, just like the last period in https://moz.com/blog/how-to-handle-downtime-during-site-maintenance if you look at the source the there is a link to S360 but after that there is a small link to a other page just from the period. I just hope that Google would stop counting this kind of links.
Hi Rand,
Great video...and I'm relieved we don't do most of these things. However, I have a question about microsites.
We have an organizational website focused on all of the services we offer, as well as a microsite with a different look and feel (though complimentary to our master brand) focused on just one of those services. This microsite is part of a branding awareness campaign for this one service, and the information on it is completely unique.
Is this an acceptable use of a microsite, or would it have been better to house this awareness campaign within the main website even if it has a different look and feel?
Thanks!!
It's awesome to explore this type of unique insights for the sem/seo practices and I believe it'll help many digital marketers to sharpen their knowledge. Deep analytic and plenty of information. Love it all.
I work with a lot of service area businesses, and exact match domains rank all over the place. Many companies are creating 50 different EMDs, and ranking in 50 different cities with them, based purely on the domain and sometimes a fake location. Does Google consider this spam and can it be reported and eliminated? Its very difficult to preach no exact match domains to businesses, when the only thing ranking organically in some cities are spammy sites with repetitive content for every city in the area.
For the love of the web pass this on. More interesting reading on Keyword Useage here:
https://moz.com/blog/optimize-marketing-with-keywords
Got to love the fact loads (and I mean the majority) of SEO's still don't understand this... many will go out of business and still wonder where they went wrong and why their clients left them... come on people listen to uncle Rand... you don't need to be a big brand to write authoritative interesting content that gets shared / liked / talked about, it's not about any of these old school techniques now as engaging content outranks anything.
I'd like to see into the future and see how many SEOs have gone bust in a couple of years
Hi Rand,
Thanks for this =)
Do you think there is a whole WB Friday session in determining how tough competition will be for ranking organically? Would love for you to elaborate on this some more!
Yell is a general directory are you saying it would be bad practice to enter your details in Yell? Or are you saying some directories are bad and some are OK?
Hi Rand, thanks for another great WBF.
Just wanted to comment on the AdWords part - CPC (and Competition, but a bit less) metrics actually can be worth to look at when doing keyword research for organic rankings, in my view.
The logic is this: if one keyword has a higher CPC than the other, it must mean there is
a) either greater competition for the keyword (we can compare it with Competition metric to confirm or deny that)
b) or it holds a higher commercial value (in terms of high click rate or conversions)
In other words, the advertisers must be fighting more over that keyword than the other for a reason and one of the most likely reasons is that they are more likely to convert the users that search for it. So we could look at the combination of CPC and # of monthly searches (and possibly Competition) to determine which keyword would be more worth to rank for (if it brings commercially more focused traffic and eventually more conversions than the other keywords).
So to sum up - AdWords metrics aren't that good for determining organic competition for keywords, but could be good for determining the general value of keywords (which includes organic and PPC), if approached with logic.
One other thing you could maybe consider (but it's more sci-fi) -> the AdWords Competition metric could be more important since the 4-ad pack was introduced, cause a very High competition (0.90 and more) could mean that most of the time all 4 positions are taken and the organic #1 position is further down and maybe even below the fold on some screens :) And since 50% of users (some recent study found that number, I didn't invent it) still don't recognize ads from organic search results, that means you could be better off targeting some keywords with less traffic but with lower PPC competition (and therefore being higher up the screen on the #1 position). Let's call it the #1 position's screen-height :D
I'd prefer to measure that directly (by looking at how many ads appear and where), but fair enough - it may be correlated with a metric that's useful for SEO, just don't rely on it to always be accurate :-)
Hi, Rand, great post as always.
I have one question relating to point 3, different pages for keyword variants. Where do you stand on FAQ pages? As although they aren't keyword variants, per se, you could argue that they could be grouped into one overarching topic.
I'm constantly having battles with SEOs who want an individual page for each FAQ. However, I'm of the opinion that it is much better to have all of those questions grouped onto one page. I think it is much better for the end user to be able to build up trust in a product/service by having all their questions answered at once instead of having to navigate through multiple pages to get those answers.
Thanks,
Craig
I generally agree that from a usability standpoint, it's better to have an FAQ page all together with a list of answers. However, for keywords that have real volume and are worthy of pursuing for SEO, I actually like to create more comprehensive pages targeting them specifically. A shorter FAQ list answer can then link to this longer, more in-depth version.
Awesome, thanks Rand. Unfortunately, I still see cases of pages being created for questions that have a yes/no answer. But I totally agree with you on the more comprehensive guide for questions that have real volume, and that's the process I'm trying to put in place at the minute.
Excellent WBF, Rand! We're in the process of creating an SEO glossary page for our website: http://topleague.in/seo-glossary-terms/ (feel free to remove the link). With regards to your answer on FAQ's, should we define the SEO terms comprehensively or write shorter answers for each SEO term and then link them to a longer post within our website (or an external source)? Appreciate you sharing your insights on this!
Note for other commentators: Please, don't thumdown this reply just because I've used a URL. I'd rather have it removed if found violating the rules than get thumb-downs for nothing. Thanks!
Hi Rand
Great tips
The truth is that circumstances change too fast and some aspects as the length of Anchor Text or not to use pages for every keyword variant does not work as well as before
Happy weekend
Oh Rand DA Man how awesome this WBF was I have to say that I am guilty of only #1 but have been revising my approach and after listening to this it is a solid change for me. I think that when building title tags just like we all know a catchy line utilizing them instead of stuffing them in there is key. I like how you brought up the thought of the user thinking that this is "Sketchy", and for the reason that many of us SEO's were in a place to rank, rank, rank, but forgetting that user trust is what drives the click.
I have seen some of the other tactics from various marketers I know and a simple conversation does not change there mind so I will let the power of Moz direct them with this. As always thanks for this and helping the community become on top of the latest tactics.
Great points, I recently interviewed candidates for a SEO position and was amazed how people that consider themselves as experts still belive those tacticts works.
I completely agree with all the points however one small thing, when I'm conducting a keywords research I do look at the potential bid, not as any metric to determine my needed organic efforts, for this I will conduct my own competition checks on organic results, but to understand the big picture- to know how 'busy' and 'wanted' this arena is in general.
It has been said for a long time that exact and partial keyword match domain names are not an important ranking factor. That may be true in the USA but in Canada, they still work extremely well. It seems that Google does not apply exactly the same rules outside of the USA. Is this true? Did anyone else notice this?
According to Google, they've got only a single algorithm globally. That said, in smaller markets, there's sometimes less competition and thus, tactics that wouldn't work in the US do work (e.g. link spam still works in many regions). I'd also be careful about ascribing the EMD/PMD sole responsibility for the effectiveness of the SEO campaign. Sometimes and EMD/PMD has done a great job of building a true brand, earning great editorial links, offering terrific content, etc. I've found it's often harder to get those great links, press mentions, and brand recognition with an EMD/PMD, but that doesn't make it impossible.
Good WBF Rand. I'm curious about your comment on Google possibly penalizing for exact match anchor text on internal links. I have been advising people that it still *is* good practice to use exact match anchors internally.
This video from Matt Cutts is a couple of years old, but he seems to say that it's ok to link internally with keywords:
https://www.youtube.com/watch?v=6ybpXU0ckKQ
The exception would be for sites that are severely overdoing it in a spammy way. Matt says, "It's not the kind of thing where internal links from a domain would cause any sort of issue...unless...there's always that one kind of person that takes things way to far and has a kajillion links all on one."
I'm curious to know if you have some data that shows that internal linking with exact match anchors could be harmful, or perhaps shows that it's not as effective as it once was. I think what you're saying in the video is that you could be penalized if you had super spammy internal linking, but I fear that the point that comes across is "don't link with keywords internally".
Hi Marie - yeah, we have seen a few examples, particularly the ones I talked about in this WB Friday: https://moz.com/blog/linking-internally-externally... where folks went overboard in their navigation or footers (or with in-content links) and saw penalties (or at least, dampened rankings) that lifted once they shifted away from those tactics. I think it's a fine line - there's "this keyword use is also helping users on the site" and then there's "I hope no users see this, because I'm just doing it to get a bump from Google." The latter, obviously, can cause problems.
Fantastic and useful article Rand - Cheers! Perhaps a typo in point 1 though (Studying instead of Stunning)?
People these days are becoming more and more quality conscious. These days I guess almost everyone in their right mind are able to differ what's quality and what's not. All these old tactics were business minded and not respecting the customer/visitor. But now the customer is king and the king is satisfied by the quality and respect he gets. Hence, it's always important to greet them with quality content that's inspiring and persuasive at the same time. I appreciate that Google has updated its algorithms to deliver quality results and strives hard to retain the user's confidence in the search engine giant. As someone has said, "Google is now a personal assistant". For me, Google really is! Thanks for this great article.
I think it's always valuable to know what the industry looked like to have a better grasp on why certain things are done today, especially for those new to the industry.
Very useful information. Although, you skipped to number five before telling us the substitute of those old school link building techniques that you mentioned.
As always Rand, a very good whiteboard Friday, it's good to get these points out there, as you said, we do still see these techniques used.
A question I did have about the variant for every type of keyword , but what are your thoughts on business locations pages, optimised for different services (or location pages in general), surely these are a type of keyword variation page, as they are all trying to accomplish or say the same thing but for a different location.
Don't get me wrong, I don't agree keyword page variations are a good idea at all, but there is some overlap with location pages.
Thanks, great summary.
On multiple domains, would that also apply in your view to a company with 2 sub-brands, one for the local market, one for foreign markets? Do you think it would pay more to just group them into one domain?
Thanks again
This is exactly why SEO is not dead, i know that is not what we are arguing about here but, it has changed and if anything it is exactly why we as an agency can justify ongoing relationships with clients. The value to them is that we know that things have changed. I realise it is hard to keep clients abreast of all changes but that is the value we should be providing. I think we can all agree that the old tactics did work and have probably been used by most SEO companies at some point because they did work in the short term. The clever people have evolved and taken that new knowledge together with ways of applying it and provided real value for clients.
In my opinion SEO is more about the experience. If something sounds slightly odd, spammy or strange...it probably is not a good long term investment. What you need is good, honest ideas and most importantly the time to execute, test and sometimes fail.
A Superb WBF Rand, I have seen people still relying on techniques like creating different pages for different but similar keywords which is obviously an obsolete technique now. The best point about this WBF is crafting a perfect content that seems legit and likely to earn links.
Still I would exclude directories from this list. Of course any type of resource if it's spammed to death, then it will not add any SEO value or worse. But big directories like Yelp, Yellow Pages or industry-specific ones still count. Also I see that commenting (not comment blasting) gives some effect if it's not the only type of backlinks in the profile.
Hi Rand, I agree you with all points except 4th one. While doing competitive research I see websites with lots of reciprocal, paid and low quality bookmarking links occupy top ranks for extremely competitive keywords. I think Google is still far from perfect when it comes to recognizing good and bad links. I also think, few manually edited quality directories can be done if used with variety of anchor texts. Thanks!
Very interesting Rand! These are exactly the tactics which I told my former agency to get rid of last year. It's nice to see you finally explain them in a Whiteboard Friday! Cheers!
I would love to find out more about free tools to use to find out about keyword difficulty if anybody has some suggestions!
Hey Rand,
You really quickly said 'PageRank is defunct'. Does this mean Google no longer take it into account, or are you saying that it's defunct from Moz's side of things?
Hi Rand, quick question. If you avoid all potential methods of acquiring spammy links and only try achieve links as a by-product of good content and brand awareness (so essentially don't actively seek to gain links), is there a need in 2016 to use the disavow tool, or are you risking losing potentially helpful link juice?
Hi Aaron,
Interesting point that I have been talking with my team about. We noticed that many of our clients develop backlinks to their content that we did not manually build. A lot of those sites we do not want to be associated with and are bad links, so we disavow them when appropriate. We avoid trying to acquire spammy links but they link to us anyway and so we have a use of the Disavow tool. Hope this helps!
I'd still use the disavow tool to pro-actively tell Google any links that come to you that you feel are spam. It's rare, but sometimes a number of spammy links could be interpreted by Google as part of a manipulative campaign on your part.
Great Article which recalled my Memory! Anyways Nice Hair style Rand!! Specially cross mark on back :)
It seems that Google is still playing favorites with exact match domains for the keywords "Rap Music". Even though the company I work for (rapmusicguide.com) has musch better content, (rapmusic.com) continues to rank higher than us in the search results. This has been an uphill battle for us!
This post was sooooo needed!! It is amazing how many websites, videos and tutorials are out there still preaching these old strategies and techniques that are not only obsolete but are detrimental to websites. Kudos Rand! Thanks you for keeping up the great work!
My view of "effective" (from a long term sustainability point of view) is probably the same as yours, I'm offering a counter-point for the sake of broadening the debate. It is important to note that PBNs still work because though I would never personally recommend it, many good, honest, SEO folk will be up against people using them in churn-and-burn sites, and you've got to know what black hat techniques still work and you could be up against in competitive SERPs. I work in-house on corporate B2B websites so I wouldn't touch them! However, to suggest otherwise of their potential effectiveness could be construed by some as being naive to the point of being unhelpful.
I always say the same: finally, we must do what the "logic" says.
Gradually, Google "forced" us to work as years he says. Make content that interests the user.
Hi @ Rand Fishkin its really great article like other whiteboard friday . But in this edition its really helpful for me just because few of those old tricks i used now a day. So from this article i really know about this get me some new idea thanks again .
I wonder what would be the best practice if you offer an informational site in different language. Wouldn't it be better to have different domain names. Lets say you have fitness-brandname.com (offering information about sports, health and fitness). Now you want to expand to the Spanish speaking market. Wouldn't it be better to it then aptitude-brandname.com (aptitude=fitness).
This information is exactly what I needed. My blog Makeup and Beauty Forever (feel free to remove the link) needs a lot of SEO brushing up and I think its time for the new upcoming bloggers to start following new ways. I am sure going to share this video with other bloggers so that everyone can benefit from it.
Rand - quick question...
If a company you're working with has several pages that are near enough duplications yet with slight variations in keyword (point 3), why not canonicalise these to a singular page? Is this a legitimate idea? I've implemented this in the past, not only for the scenario you outline above, but also when I have multiple pages ranking for the same topics, but at varying rankings.
I found that doing this pushed the target page much higher in the rankings, not only for what it was ranking for originally, but also for any keyword rank the other pages were benefiting from...
Be great to read your thoughts on this!
Yeah - I love that tactic Mark. We've used it plenty here at Moz, too, often with very successful results all around (better page for users, and higher rankings in Google, too).
Thanks for the response!
Yes, i agree with your point martin.
Thanks, Rand
Great topic and on time. One thing you didn't mention and I find quite disturbing is that so many people still think that more links mean higher rankings. People couldn't be more wrong. I see more and more website with fewer links outranking old and established brands.
Why?
There is a simple answer. Relevancy.
Penguin was the first google effort to somehow start 'rank' backlinks. Now we've got RankBrain. Many people associate RankBrain with on page content only and forget that RankBrain is learning from human decisions – specifically what they click on.
Focusing on 'quality' links coming from your niche rather than quantity is becoming ever so important.
The old SEO practices can remain relevant and eficases
Hello Rand Fishkin, Thanks for sharing good information and i have always appreciated your content which is very helpful to reader.
Do meta keywords matter anymore? Does Google or any other search engine take them into any consideration? What if my pages don't have any meta keywords in them at all? Will that hurt my rankings? Thanks in advance.
Hi Rand,
I'm a bit worried here, different pages for every keyword variations.
Though, these are post pages, not static pages.
Let's say I created a blog around the topic "Link building", that's very specific and all that is discuss on the blog is just around that topic.
There is all possibilities that most content focus keywords will contain the word "link" "building" "earning" in them.
Is this the same as #3 above?
Thanks.
Great one today, Rand. Thanks for always sharing interesting thoughts.
Thanks, great walk down memory lane. I appreciate information with an emphasis on DOs and DON’Ts. Item #2, heavy use of anchor text, was particularly interesting to me. I would love to have just a little more insight into what constitutes “heavy use” versus editorially sound, reader-serving anchor text. I’m trying like hell to serve the reader and use the anchor text to make it super clear (to the reader and Googlebot) what the link is about… but I can imagine that looking “heavy.” Not sure if I am doing it right or not!
Hi - great article/presentation, I agree with most of the content but this bit:
"...we know, as SEOs, that Google is using click signals to help documents rank over time and to help websites rank over time."
Rand, you were in the March Q&A talking to Andrey Lippatsev (Google) about this topic. You talked about how your CTR experiments worked, that is they sometimes did and their effect was very temporary when they did work.
Andrey said he'd like to watch one of those experiments sometime to try to figure out why Google was been 'thrown' by some of your CTR experiments.
Andrey went on to explain why Google could not use user signals like clicks from SERPs to rank sites, partly because it is so easily gamed... which echoes what Google has told us previously about user signals (e.g. Gary Illes at SMX).
So...
How can we say that we as SEO's know that Google is using click signals other than in test environments?
Well, we did have another successful CTR tests after Andrey and I chatted, but I actually wasn't referring to those instantaneous results. Rather, I'm talking about Google's very public statements that they use click and engagement data to determine whether they've got good results and whether they should be modifying their algorithms. Paul Haahr covered that nicely in his slide deck here: http://www.slideshare.net/SearchMarketingExpo/how-... (see slides 29-31 in particular). I tried to make that clear with my "over time" statement in there, but I see there's room for misinterpretation. My intent was to talk about how Google measures rankings over time with query/click/pogo-sticking data, which I think we're all agreed upon at this point.
Thanks for such a quick response Rand!
I think I see where you're coming from. My point is that Paul Haar's use of user behaviour data in their test environment to help improve the algorithm itself, is not quite the same thing as using clicks to rank websites, however it becomes a murky line between the two over time - totally agreed!
We also have another complication to the concept - Google can show different titles/meta desc's to what we include on pages which of course influences our click data!
I'm still unclear on how they might use pogo-sticking without a very careful correlation with user intent for the particular query (e.g. searchers just grabbing phone numbers). The last I heard 'bounce rate' was not used for that reason amongst others. At this stage I'm more a fan of the 'long click' concept (a la A.J Kohn) over the other user signals!
Thanks again for your clarification :)
Hello Rand,
Well, So surprised to see this covered because I just wrote about the same topic a couple of days ago relating it to Anzac day. You can find it here.I would apprecaite your input on that.
404 page not found - @Hardeep
Awesome as usual. Thanks Rand!
Rand never disapoint :D!
Great post thanks @Rand really help me to build new SEO strategies for white hat website promotion.
Hi, I write in my blog http://www.wealthstrategies.com.ph/ and from time to time I comment in unrelated blogs , because that topic interests me, would that hurt my website in I am putting my url in the url box . Thank you in advance
Rand always on point. Thank you for keeping the SEO community informed :)
Great info which I will use to the fullest !
thanks for sharing, good information.
Good refresher. The days of seo "tricks" are no longer...
depressing!