If you've been around the SEO world a while, you've undoubtedly heard the old adage:
Do what's right for users and engines will reward you with higher rankings
Along with its peer:
SEO tactics that focus on engines, rather than users, are manipulative (black/gray hat) and will eventually be discounted or penalized
In my opinion, both of these statements are utterly false and tragically misleading. In my view, SEO starts with the user (of course), but cannot ignore the incredible importance of search-engine targeted (and specific) tactics. When I first considered the issue, I thought that perhaps, years ago, these opinions were more accurate than they are today. However, after visualizing the issue, I discovered even that isn't true:
(please note: graphs like this are, as always, just my personal opinion)
The value of tactics from each set has risen/fallen over time, leading me to the conclusion that this was never good advice. And yet, thinking back, I'm almost sure that at some point, at a conference and during interactions with clients, I personally repeated this misnomer. I want to issue an apology for that now and set the record straight - SEO is a task that requires paying close attention to the needs of both users and engines. You can't be an effective SEO without it.
Just think of all the specific tasks we perform that we'd never do if it weren't for search engines:
- Title tags: We might still make them, but agonize over keyword usage and positioning, uniqueness and flow? I doubt it.
- Meta tags: Nope. No reason to even bother.
- XML Sitemaps: I'm pretty sure no human has ever visited this file in an attempt to sort out the pages on your site.
- Webmaster Tools Registration: Without engines, there wouldn't be any.
- Keyword Research: I think this practice would be more like advertising copy - think Mad Men.
- Keyword Targeting: Why worry about keyword placement for anything other than conversion rate optimization?
- URL Canonicalization: No need - visitors are getting the content either way.
- Accessible Link Structures: So long as you're not worried about the >2% of visitors who can't see Flash, go ahead and build rich applications to your heart's content.
- Robots.txt & Meta Robots: No engines, no reason to direct engines.
- Link Building: Unless it's specifically to draw in relevant traffic, why bother?
- Creating Vertical Search Feeds: That's going to be time wasted.
- Information Architecture: While there's good reasons to do some of this for users, a significant portion of the accessibility and link hierarchy arguments are made moot.
- Redirection: Without engines, we can use whatever method is convenient - javascript, meta refresh, 302 - it makes little difference to the user.
- Rel="Nofollow": Internally or externally, it becomes a pointless attribute.
I think the problem with the classic "build for users" advice is that it sounds so compelling and, on a surface level, makes a lot of sense. Maybe this is a good warning not to adhere to any advice just because it seems logical on its face - knowledge and expertise may not make for simple messaging, but, outside of politics, accuracy is far more valuable than fitting into a sound byte.
UPDATE: A lot of folks in the comments are under the impression that I'm recommending against building for users - nothing could be further from the truth. Websites are made for people, and users should absolutely be the focus of your efforts. My argument in this post centers specifically around the practice of search engine optimization and the idea that tactics which are engine-focused (like XML sitemaps, anchor text, link architecture, webmaster tools usage, etc.) can be ignored because they're not "for the user." The charts and points above are intended to illustrate that if you only focus on "user-targeted SEO" you're missing a huge chunk of the potential SEO opportunity pie.
1999 - Built it and they will come. (aka field of dreams syndrome)
2004 - Build it for users, not search engines, and they will come.
2009 - Build it for users, optimize it for search engines, create content for link bait, promote it through social media, reoptimize it for conversions and they will come and convert.
The advice has not changed over the years, there are just a few more pieces of it today.
i think you hit the nail on the head. good content is important but without seo it wont be found. Google wants to return results that are useful for users but you also have to help them out so they know you exist
Yes, yes, yes. I have been at this since 1999 and have found your assessments corrtect, James.
OK, i'll take the bait. I get what you're saying, but I also disagree.
As has already been said, making charts that seem to quantify real values that in reality are just your own personal opinions means little. I don't even know why it's worth questioning the made up figures, but....
Manual Link Building: Is now moderate? I mean in 1997, search engines were saying build relevant links. I remember, because I was writing about it back then.
Meta Data was exceptional in 1997? No. The meta keywords tag wasn't fully supported even back then, and it still have relatively little impact on ranking.
Engine Protocols are now exceptional? Rand, I could build a site with no sitemap, no robots.txt file, no use of webmaster tools and rank perfectly fine. You know this. Plenty of people do.
Site Submission: Yeah, I can roll with that previously being more important in 1997. Died out pretty quickly soon after that.
Canonicalization: Well, I think even in 2000 it was seen as important to know your 301 from your 302, but OK.
Look, the one consistent thing we've seen the search engines do over time is try and build ranking factors that mimic what humans want. Those who've understood that seem to have done well in the long term with little effort. But to take your list of what you say are just for engines:
Title tags: Yes, we think more about using the exact terms someone might use in a search to include here, but the title also has an impact on what the human reads when a page shows up.Meta tags: Because having a nice summary of your web site via meta description isn't for humans?
XML Sitemaps: Hey, don't do them.
Webmaster Tools Registration: Again, don't do them.
Keyword Research: If you don't know how humans are looking for you, how do you reach out to them again? You're doing it for search, but ultimately you do it for the HUMANS that are searching.
Keyword Targeting:Sounds like keyword research to me.
URL Canonicalization: Don't do it. Search engines often do the right thing anyway.
Accessible Link Structures: Or not worried about humans bookmarkging, those on iPhones, those who don't like Flash, humans on low bandwidth devices...
Robots.txt & Meta Robots:Don't need to use it.
Link Building: People go to sites outside of search engines? Why wouldn't you want humans there to find you?
Creating Vertical Search Feeds: Don't make one. Might get picked up anyway.
Information Architecture: Good IA is for humans, not search engines.
Redirection: Search engines often figure out redirection even if you don't explicitly help them.
Rel="Nofollow": Yeah, we do that if we don't want to get busted for selling paid links. Not a human thing. Is that the point of all this?
Whew... I'll try to respond as best I can:
Manual link building - had barely started to gain value for SEO in 1996-7, so I figured "moderate" was an appropriate palce for it, but I suppose I could have put it a little higher.
Meta Data - Maybe my memory is faulty; I had recalled that several of the popular engines then put heavy weight on the meta tags, specifically keyword and description.
Engine protocols - I'd argue that XML Sitemaps, using webmaster tools, etc. are very high value (maybe not robots.txt though), but fair point, I could have put that point a bit lower.
Site Submission - cool, we agree :-)
Canonicalization - cool.
-----
Title Tags - That's exactly what I said in the post
Meta Description - Precisely! Why would you bother writing a meta description if there was no search engine to display it in their snippets. Users would never see them without engines.
XML Sitemaps - don't do them?? I've found that they add substantive traffic benefits for nearly every site we've ever worked on.
Wm Tools - again I disagree; registering and seeing the data, particularly with regards to the crawl stuff is hugely valuable for SEO. And yes, you'd see some of that via log files and failed visitor paths, but much of it is important for SEO and the flow of link juice even on pages that receive little to no actual visitor traffic.
KW Research - Yeah, and with no engines, there are no keywords! There's just advertising copy and marketing focused terms (as I said in the post, more like Mad Men).
KW Targeting - Sorry if that's unclear. I meant the practice of employing the terms you want to rank for in prominent places on the site/page.
URL Canonicalization - Again I disagree. The engines do a half-decent job with some forms of it, but it's a big part of successful SEO. You even have whole panels dedicated to it at SMX!
Accessible Link Structures - yeah, that's exactly what I said in the post.
Robots.txt & Meta Robots - agreed, most sites don't need it unless they need to block something or keep it out.
Link Building - did you not read the post? I said specifically that all the link building activities that we'd do to gain relevant traffic would still be important, but those that are purely focused on engines (which, let's be realistic, is quite a lot) would dissappear.
Vertical Search Feeds - really? That's terrible advice! If you want to get in Local, you need to register and fill in your info. If you want to be in Product, you need to create a GG Base feed, if you want rich text snippets, you have to provide the engines with that data.
IA - Agreed, and I noted that some IA is for humans, and others is for engines.
Redirection - Yes, the operative word being "sometimes" which I don't think makes it "optimization" or "optimal"
Rel Nofollow - We originally did it so that engines wouldn't think we were linking to spam and spammers wouldn't benefit from leaving comments/guestbook entries/etc. I think we agree it's not a human thing...
Maybe I'm losing the meaning of your comment, but you seem to be suggesting a lot of bad SEO advice and I'm not sure exactly why...
Do I smell flame?
Both of you offer some very interesting opinions. As I was reading this comment thread, I keep thinking of a few lines from a popular movie you might have seen:
Obi-Wan: "Your father was seduced by the dark side of the Force. He ceased to be Anakin Skywalker and became Darth Vader. When that happened, the good man who was your father was destroyed. So what I told you was true... from a certain point of view."
Luke: "A certain point of view?!"
Obi-Wan: "Luke, you're going to find that many of the truths we cling to depend greatly on our own point of view."
IMO, our individual experiences and results can shape our memory and opinions on the value of certain SEO elements.
For instance I remember Exact OnSite Keyword Targeting being Exceptional early on (think scraper sites and pages that would rank that were just filled with keyword triggered search results) and then dipping in response to Exact OffSite Keyword Targeting and Over OnSite Optimization issues and now becoming slightly more important in relation to Title tags matching limited OnPage Keyword placement.
As Rand stated earlier, this is based on personal opinions.
Here's what I wrote on link building in 1997:
In my chart that went with that, 3 of 6 search engines were noted as using link popularity as a ranking signal.
Google started growing in popularity in 1998, and they came even more important. Whether they were moderate or high in 1997, I suppose we could debate. But when did manual links drop back to near moderate? And what type of linking is better, automatic?
On meta tags, yes, your memory is faulty. Here's again what I wrote in 1997:
Engine protocols, we have to disagree. Don't get me wrong, they are incredibly valuable tools. But you can build a site and rank with it despite never using any of them? How? Because the search engines try to do the right thing regardless.
On the meta description, yes, users would never see them if there weren't search engines. But you didn't say "let's pretend there's no search engines" (I'll get back to this, the most important point, at the end). You said advice to ignore what search engines would like is false. The meta description tag lets you please the user of a search engine, not the search engine.
Webmaster Tools: The data is hugely helpful in your opinion as necessary to rank well, because you're hyper into SEO. There are, again, many people who never try to download their link data or make use of the tools in other ways and yet manage to rank just fine. Wouldn't you agree?
Canonicalization, I agree -- there's lots you can do to help the search engines. But again, you don't have to do it, and people still do well despite what an SEO would view as serious errors. We've both been on enough site clinic panels where despite someone doing everything "wrong," they still rank.
Link Building, yes, I did read the article. You said why bother doing it unless it's to bring in relevant traffic. Now you say most of that in turn is just about getting link juice for search engines. I guess I view link building completely different. It's about getting a link on a page that has an important audience you want to reach. In turn, you might also get additional search benefits.
On Vertical Search, i agree -- it is terrible advice to ignore specialized submissions options for them. But if you're being all hardline, as you are, then yes -- you can also ignore them. You might, exactly as I said, get picked up anyway.
Look, the bottom line is this. You didn't title this article as "What If There Were No Search Engines." Instead, you've attacked the idea that doing the right thing for users will help you with search engines. I mean really attacked, right? These adages:
You said:
Wow. I mean no gray area in there? Later on, you get more reasonable:
Yeah, I'm with you there. That's reasonable. Thats just not the tone of your overall piece, complete with line charts that lend an air of scientific expertise to attack the "user" view. It doesn't advance that reasonable advice.
But I see elsewhere you're saying you've done this piece in reaction to those who seem to be hardline about doing things only for the user. Sure, I know that type. In fact, this entire debate has been going on for years, literally years. Change the names, the arguments stay the same
But Rand, I think you forgot about the more typical webmater out there. They aren't reading SEOmoz each day. They don't know about the 200 ranking factors Google users. They're trying to run web sites, and the very last thing you want to do with that group of people is introduce them to SEO by talking about XML sitemaps. Or worse, quadrangular link buidling. Or zebra pages.
Telling them to do things that they think will please humans is EXCELLENT advice for the confused novice. Someone wants a link from your page? Will I be in trouble with Google? Well, you could do a backlink analysis and try to determine the quality of that link neighborhood or you could ask yourself, "Would my users like this?"
Building a site all in images? Is that bad for search engines? Hey, would it be bad for your users? Who doesn't agree with that.
Where the user advocates go wrong is overstressing this as a good, fundamental piece of advice and being all hardline that you never act as if there are search engines. Of course there are, and of course you do things specifically for them as well.
Ok - I think I see your point (though I still think you're being really nitpicky about the specifics - 3/6 may give a boost for link popularity in '97 and you're still giving my "moderate" a hard time?).
In some parts of the post, particularly at the start, I overemphasized the anti-user tone, when what I'm really advocating is not ignoring the engine-specific side of things. I think you're parsing my wording a bit more harshly than is warranted, but criticism accepted and after reading your comment I think we share the same perspective - you CAN achieve good search results while ignoring many of the search engine protocols and search-specific recommendations, but it's not "good SEO" to do so, nor advisable if "optimization" is an important part of your business goals.
Just updated the two graphics to attempt to be more historically and currently accurate based on your comments and those of some others. Keep in mind it's still just my opinion, but I appreciate the feedback.
Everyone needs to stop with 1997 already. Geesh. Explaining 1997 to clients is TMI. Focus on today.
Hi Danny,
thanks for this. What I really like is to focus on good content. Because good content itself is a kind of optimization : it generetes links and traffic automatically.
But, as Rand said: this is not seo. In my eyes its a kind of definition for seo : to use all these methods to rank a content, that could not live on itself ( that means nobody will see it ).
I had some old sites where i never had done anything (without writing and showing art ). The pages ranked well. But when I began to optimize - quit simple - the pages ranked much better. So i would say : good content for users is the "i" and seo is the point on the "i".
Hey guys, thanks for this discussion. Its one of the best I have ever read :-)
Oh what I wouldn't give to see you both in the same room discussing this! Haha! I love it! :) Great info!
Rand -- xml sitemaps: two years ago at UW you said don't do them. I can look it up if you need.
Yeah - and I did it on the blog as well (a post I've since updated). That's certainly not the only time I've switched positions, either - we recently updated our stance on paid links; I've changed my mind about different H1s vs. Titles; we've changed our advice around PR sculpting and many, many others.
I wouldn't expect that to be the last time, either - although I wish our prognostications would be perfect, SEO is an opinion and experience based subject, and those are both subject to change. I'd hate to be held to a standard that doesn't allow for flexibility and even reversals of position on a semi-regular basis. This is anything but a static field.
I'm so interested in this article and especially this debate; Rand and Danny are two major stars for me in my chosen industry.
When they don't agree it makes me feel uncomfortable/insecure - like hearing your parents fighting!
But top marks Rand, your somewhat incendiary title drew in lots of views and stimulated a heap of useful commentary, even more interesting to see those that failed to read or perhaps understand the underlying point/s.
I guess that emphasises how we all tend to read different things into the same message - all communication is contextual. There's a lesson for content right there!
This back and forth with Danny is the most valuable part in mho, and helps make it one of the best seoMoz posts I've read in a long time.
Thanks guys
I'd even take that a step further. I'm sure we're all agreed that when you optimize for a particular keyword phrase and thereby get the page to rank for that phrase, when people click through to the page they expect to see those keywords almost immediately.
We want to use the words our audience uses, not just because that's how they search, but because that's what they want to see. It helps them to see that the page is relevant to them. If your page's target audience is neurosurgeons, referring to the cerebral cortex as "that big lump of goo that thinks for you" is a bad idea.
I think that would be true even if we took the search engines out of the equation (although keyword research would obviously be far more difficult if it weren't for search data from the engines). Is a medical student going to recommend the big lump of goo page to the members of their study group?
I have to disagree with a few things you just said.
Manual Link Building: Useful - yes, but necessary - no. Pinging the social web with great content is a faster and more efficient means of doing this. Creating a community that promotes you is much faster than doing per se a "Link Campaign".
Meta Data was exceptional in 1997? - regardless of factual exception. There's always been a perception that this tags boost rankings until recently.
Engine Protocols are now exceptional? Yes. You can build a site without using the tools or protocols and gain ranking, depending on the keyword. I have to ask though; do you really think Google created Webmaster tools just to joke us around? And for what reason is that prudent or strategic on their part?
Good article Rand.
I've always wonder a bit about those phrases myself. "Build websites for your users". Yes, of course you have to build website for your users, but there is so much more to it.
My personal opinion is that in the future SEO will need to be a part of an integrated (online) marketing strategy to succeed. A few years back you could get away with just doing a good job in the SEO department, however today you also need to have a good design, usability friendly site, watch your web analytics and improve when "holes" are found, comply with Google and other search engines, keep up with technology and algorithms, conversion optimising and so much more things.
Back in the days all you needed to make a "good" website was an editor. Today you can easily use 20 different tools, all focusing on different areas. In a way I'm glad this is the direction we're going in. I think this will make internet marketing more serious in the future!
I can't agree with this one. Focusing on the user is terrible advice? There is more to a successful website then just ranking in Google. What are you going to do with all that traffic once it comes? You have to be able to take relevant traffic and convert it into something (sales, leads).
Focusing on the user is the MOST important aspect, but obviously it isn't the only aspect to a great site.
This article may be accurate if SEO lived in a bubble all by itself, but it doesn't, it's part of an overall strategy to make money. Converting visitors into dollars is the name of the game - you need to focus on users in order to do this.
Also I think this "focus on users" advice is often given to non-professional SEO's. Basically it's saying don't get too hung up on the technical details - built a site your users will love - this will lead to traffic, links, and revenue.
Todd - again, I think you're taking the post out of context. I never said (and certainly would never mean) "don't focus on the user." I said and mean "don't take the advice that good SEO is just doing user-focused SEO." When you're doing SEO, you have to think about the engines - you cannot ignore them.
I'll try to edit the post as a good percentage of folks seem to be getting the wrong message.
Most people just read the title and assume you're slamming usability. Make them read the damn post!
Even as a usability specialist, I sometimes bristle at advice like "Build for your users". Partially, it's because it's vague and useless, much like "Build great content". Of course we want to make users happy, but what does that mean? Smiling and nodding at each other's vacant platitudes doesn't add anything to the user experience.
I think well-rounded practicioners realize that a lot goes into the user experience, which often begins with search. You can't just draw a dividing line and call everything on one side "SEO" and everything on the other side "usability". Engineering solid titles, search snippets, etc. is part of the visitor's journey.
There's also the issue of practical/strategic usability vs. some amorphous view of the user experience. As a professional, am I aiming to make everyone leave my sites with a warm, fuzzy feeling? No, I'm aiming to increase traffic and conversions, assuming that visitors who buy more often must be happier. I hope it's a win/win, and I won't create a bad experience just to make a few bucks, but at the end of the day I still have to have a practical, measurable goal, and often that comes down to the bottom line.
NAILED IT.
You my friend should work for SEOmoz :)
I think you're pushing a false dichotomy here, Rand. I've always been a "build it for users" person, but I don't know any serious SEO who's ever recommended that the search engines should be ignored. It's always been essential that you keep both audiences (people and spiders) in mind at all times, unless your goal is just to build MFA sites.
But in the long term, the people who are buying your products are in my opinion more important than the search engines that serve as one of the ways people find you, even if it's the biggest source of traffic.
If we consider extreme examples (which seems to be the premise here, or at least the way this article intends to grab our attention), building one site just for people and one just for search engines, neither site is likely to do well. The site built just for search engines may get lots of search traffic, with a 100% bounce rate. However, the user-centric site, with its great content, ease of use, and exemplary customer service, just needs some word of mouth and it will have a chance to grow in popularity, even if that popularity remains small because it never shows up on a SERP. A few sales vs. no sales. If we absolutely have to choose between those two results, I think I'll settle for a few sales.
It's also important to note that the things we do to make a site better for users are in most cases also signals to the search engines that the site is worthwhile. The things we do just for the search engines are just for the search engines.
The things we do to make a site better for users are going to benefit the site for as long as it exists. The things we do just for search engines often change over time.
Search engines are a path to traffic. Without question, they're the biggest potential path, and we'd be crazy to ignore that potential. As I said, I don't know anyone who recommends doing that. But it's how people respond to the site that's paramount.
Hey Bob - you might not know them, but they are out there :-) I've been getting a little exposure to the "only use an SEO tactic if it's good for users" crowd of late and hence, wants to write this post as a counterpoint.
Rand, I am sorry to say, but this is maybe the worst article I have ever read on SEOmoz. If this is not only a linkbait and you really mean the things you wrote (those about what we do for SEs and not for users), then you should consider to learn more about Accessibility and Usability.
Btw. You can say: "Focus on searching users." and then the statement is just OK.
I'm pretty sure he isn't saying that you shouldn't focus on the user. He is just trying to highlight that SEO is also just as important and that you can't soley depend on focusing on the user.
Am I right, Rand? Not trying to speak for ya, but that's what I got from this post....which was fantastic by the way.
I agree with you, Dana. There's nothing in this post that suggests Rand is implying that users should be ignored.
Look at SEOmoz UGC, the meetups, the SEOmoz blog, etc. User interaction is part of SEOmoz, and it would be hypocritical for Rand to suggest that users don't matter.
I think you might be confusing what I mean when I say "seo for users", which is really "SEO tactics that are aimed primarily at users."
I'm not recommending against a focus on accessibility or usability. On the contrary, I've marked both as extremely important. What I'm saying here is that if you're practicing SEO and ignoring the "engine-focused" techniques, thinking that only the "user-focused" ones matter, you're not likely to be very competitive or optimized.
Does that help?
I have to say that I agree with the “worst article” view. It seems to me this is just linkbait, while, importantly, it is also beside the point:
When we say “Focus on users, not search engines” we are talking mainly about things like CONTENT and navigation, not about whether a site should have an XML sitemap or whether META descriptions are useful.
So, the “old adage” is aimed mainly at those who think that writing CONTENT for search engines is good SEO.
(I want to add that I generally avoid commenting on SEO articles, and that I make an exception here because I hold the work of Rand and of SEOmoz in high esteem.)
Cheers!
Rand, I have to say I agree but would like to add a little thought. Sometimes it does come down to a choice between The Engine or the User Experience. Let me explain. If you have a site that you want to make a change to and it's something that you know will help the search enignes but maybe hurt your user experience, then I say you have to choose the user first.
What I usually explain to clients is that it's like a tight wire sometimes. You have to walk that line between pleasing the engines and still making your site user friendly. If it was all about pleasing the engines every website would be completely text based with no graphics at all, and if were all about the users every site would be (as mentioned above in another commen) like a Kiosk.
In the end if it's a really tough choice, the User Experience should almost always win.
Great Post!
Ignite - I'm actually of the opinion that those days are basically gone when you need to actually make the user experience worse in order to do better SEO. In the last 3-4 years, we've had several conflicts like this come up with client projects but always managed to find a solution that's been as good or better for users and still solid for engines.
My point wasn't that there's a sacrifice to be made, but that you have to respect "engine-focused" SEO tactics as much as you do "user-focused" SEO tactics in order to succeed with SEO overall. Hence, the advice of doing only what's right for the user (in the SEO field) is bogus.
OK, I think I can agree with that then. Thanks for the feedback.
I don't think the people claiming "focus on users" are suggesting that you should ignore SEO by any means. It's simply getting the point across that actual users should be the focus.
When users are your focus, you've got a site that's worth ranking well. This makes SEO easier because your site is actually worth linking to.
The "focus on users" mantra just puts things into perspective for people who are pushing the SEO side of things a little too hard. Similarly, it needs to be kept in perspective for anyone who hears it and decides they don't need to bother with keyword research, linking strategies, etc.
Basically, it's good advice as long as it's disclaimed.
Hi Rand,
Search engines focus on users and so should we. Search engines try to deliver what users want:
- relevant content,
- high quality content and
- content they can trust
I think that if we keep this in mind when creating websites we are serving both our users and the search engines.
But you are right that a lot of out actions are aimed directly on how search engines 'think'. But we should not forget why we do that: to help our users find us. When we have their attention, it is still necessary to present them content that is relevant for them, that has high quality and that is trustworthy.
I apologize if I'm off-base here, and I don't want to put words in Rand's mouth, but I think people are attempting to polarize this post a little too much. (The title helps to cause that, but hey, it got our attentions did it not?)
"Building for users" and "Building for search engines" are much too general to be taken on face value as the perfect truth, and I think that's what this article really shows. If you build solely for the search engines, no one will want to use your site. If you build solely for people, no one will find your site. There has to be a balance.
Every design/etc decision needs to weigh the benefits for the user vs. the benefits for ranking. If something doesn't change the user experience (and is white-hat) then yes by all means "build it for the search engines".If you're thinking about stuffing a page full of keywords that make it near unreadable...perhaps you should start "building for the user".
Our end goal is to make better websites, in every sense of the word. That requires a bit of both.
I like the post but have a question on something I have been wondering for a while regarding "Engine-Specific Protocols" I am a Google "Kool-Aid" drinking which means I adhere to all Webmaster Guidelines and as they change our clients website follow, but as far as Webmaster Tools for Google, Bing or Yahoo! other than submitting a XML Sitemap, I really don't see the point. Is there an important tool I am missing? Sure if I was a poor coder or incorrectly using 301s and my sites were riddled with errors I could see this being useful but only as a backup to the tests we run pre/post launch. Are there important Webmaster Tools I am missing? Any insight would be useful otherwise I find myself not really utilizing Webmaster Tools nor do they seem worth the time to continually add sites and monitor them.
Lastly I have read/heard chatter that if you have multiple sites in your Google Webmaster Tools they could be negatively effecting each other much like a general quality score given to a AdWords account where a poor campaign can negatively impact the whole account; is there any merit to this in the PPC or Webmaster Tools world?
Loved the article. I have used the whole "build for users", and I still see nothing wrong with it. What I do see is a lack of clarification, which you've provided (quite well, as a matter of fact) here.
Yes, make your website (or client's website) great for people. For instance, I push "relevant, engaging content" repeatedly. That doesn't mean leave out all the optimization, though. I definitely agree that if you only focus on what the users see, you'll miss a huge SEO chunk.
Two things I have to disagree with, however.
1. Meta tags - um... that description and title are the first handshake, Rand. How else do you draw the users in? Just because you make it to the first position on the SERPS? Don't think so. Okay, so agonizing of keyword usage is SEO, but also a little bit of marketing... i.e. user interaction.
2. Sitemap - Yes, people visit sitemaps. This is NOT just search engine based. Some visitors get confused by the varying navigations on websites and just skip to the sitemap to see what's there; most are all organized without fancy buttons and drop downs. That is, unless there just happens to be a handy-dandy site search available.
My two cents... loved the article!
Hey JR - Just FYI, I did note in the post a clarification that you'd still do titles without search engines (though probably not meta descriptions). And when referring to Sitemaps, I specifically meant XML Sitemaps - the file in XMl format that lists pages on a site, not the HTML Sitemap page that is used to direct users. :-)
Hey Rand ;-)
I caught the clarification about titles - my question is, why wouldn't you think about meta descriptions for users? Why would meta descriptions (specifically descriptions) be left out even if we weren't aiming toward search engines?
As for the sitemaps... okay, I missed the XML part. I withdraw my second point, but still hold strong on the first. ;-)
Where would a user ever see a meta description anywhere on the web if not in search results?
Meta descr are used to get users over to your site--to let them know what they should expect to find on the page. That's done for users. Right? Is this just semantics now? I like SEOMoz A LOT and I am so bummed that you wrote this link bait.
Hey Erin.
I think Rand means "Where else do meta descriptions appear, if not as search engine snippets?". As far as I know that's all they do - is there something I missed?
Assuming that it is already widely known that thengines don't use them for keywords.
In fact, you could say they are kind of link bait really!
;-)
Great reading Rand. The topic has crossed my mind many times too and sites must be built for the USERs along with all SEO tactics available for the ultimate benefit of the webmasters also.
Cheers
It's pretty interesting that Engine Specific protocols is at the top of your list for most important SEO tactics, but I find it hard to disagree.
I do have to disagree with "manual link building" being only "moderately" important. While I would agree that it's importance has diminished, I find it hard to believe that it falls so far below canonicalization in relative importance.
Certainly, link exchanges and hollow directory links are certainly far less important on the scale, I would include merit-based, highly authoritative links in the "manual link building" category and that can't just be moderately important, can it?
It really depends on what you consider "manual" link building. Content creation and promotion I put extremely highly, and that might overlap with some of the types of links you're thinking about. I really just meant the old style beg, buy, submit, leave comments, etc. for a link stuff.
I'm glad you clarified on that issue, that was just the question i was about to ask.
what i woult like to know is, what is the information source for the graphs? what are they based upon?
Would I be naive in thinking that the advice above (which is fantastic btw) applies mainly to established sites?
For example, if a small new starter wants to make a name for itself in the SERPs then surely manual link building is still a very relevant part of their early search strategy, however after the initial rush hopefully traffic caused by early promotion would then spread naturally.
I agree here. Rand make a good point. Links are organically created when great content is shared and put out there for others to read/use/reference. A great way to get the web to work for you is to share socially the content you've created via FB, TWTR, DIGG, Etc. This turns your peers into the litmus test - if it's good they'll share and you'll get links, more links that you could beg for on your own.
If it sucks - rethink what you're writing about and don't submit another crappy article or you will get ignored.
I wonder if this was written purely to manufacture a dichotomy... whatever side you're on, we'll all eventually agree that the accidental combination of your chocolate and my peanut butter is downright delicious.I say the end user IS the point of SEO - just like we wouldn't be doing this if there were no engines, the same could be said of users. The engines provide the means to connect with the user, so we have to appease the search engine gods (I like "YaBinGle", although I'd be willing to consider a variation that includes "Wolfram|Alpha") with sacrificial content and dispensation through link building. The end goal never changes - connection to people.
Hi Rand,
your article is really interesting and it's nice to see the development of the SEO techniques again in a time bar.
With most of your statements I do agree.
But in my opinion it's not enough trying to shown this in one graph. Because I think, that you have to make a difference between the different kinds of pages.
Especially you have to make a difference between small to mid pages and between really big pages with tons of pages in google index.
For big pages the "Information Architecture" has a really high to exceptional impact on SEO.
And on the other hand Vertical Search SEO is for small pages nice to have but mostly won't be the decisive factor.
Really like this post. Interesting to see the visualisation of your take on how SEO has changed over the years, with elements like social media rising in important and meta content falling away (esp with Google announcing that meta keywords are effectively dead).
I agree with the main theme - you need to ensure your site is optimised for the search engines to ensure you can be indexed for relevant content and links. You then need to ensure the on-page content and structure delivers great customer service and provides logical user journeys. Neither should be more important than the other, both are essential.
Thanks for the thoughts.
james
I'm just glad to see there are people on here who aren't afraid to disagree with Rand! Great debate!
I still think we should build for users. If it's good for a user it's good for an engine. And how can you say meta data is meaningless when most descriptions in SERPs are from the meta description? And where did you get this data? Footnotes, please.
Wow...I just now read this entire post and the comments and it's taken two cups of coffee.
So much has been said that I hesitate to add to the pile, but I'll compromise and try to be brief.
My take away from this article was that SEO is still very important and very relevant to achieve results from your website.
I would add : just like Design is relevant and Usability is relevant.
The end all be all for most sites is conversions. Whether that is measured monetarily or by eyeballs, it's all about conversion. It takes a usable site to convert, but the site has to be found, hence the need for SEO.
It seems like a very simple concept to me and I don't understand all the hullabaloo above.
Great Article, I think too often we go so far to one side or the other that we need a reminder that sometimes its good to do both.
The reality is that you have to do both, as your graphs (love those graphs!) indicate.
Great post.
Wow, this one should create controversy.
I completely agree with you, however I'm still saying build a site for users, yet we speak the same SEO language.
Every point that you said is done to satisfy the search engine doesn't enhance the value of the user, is wrong imo. For example, the search engine's goal is to determine the most relevant site for it's users. Since the SE decided that an XML site maps helps the SE know what's on the site, the SE can then more easily deliver the right content for that site to it's users.
Therefore, arguably, a task that enhances the value of the visitor.
Link Building --- well, that's simply exposing the website in the right areas of the Internet...this is marketing, not link building. OK, so this might not enhance the value to the visitor, but like Best Buy purchasing ad space, that's exposure in the off line world.
Perhpas the concept of enhancing the visitor value is taken out of context. I simply argue SEO is bullshit, quality SEO is simply quality marketing.
PR: wait... I: wait... L: wait... LD: wait... I: wait...wait... Rank: wait... Traffic: wait... Price: wait... C: wait...
Can you please clarify site submissions and manual link building? There definitely seems to be some bleed over in these two categories.
Very Interesting post, though lots of disagree's.
As of your point Rand yes its very clear "SEO requires paying attention to NEEDS of both users and engines". Fair Point.
As per my experience there comes a point where we sometimes should think of using SEO tactics, as well be fair to users experience. Sometimes Content is well prefered rather than image or so, as well image being used to be clear rather Content.
I would say yes its of 50-50 choice to make, though it depends on our clients as some might be rigid or damn specific of their course.
Good post. !
I've come late to these comments, and I admit I haven't had time to read all the feedback yet, for which I apologise. It's a shoddy gutter press tactic to use a shock headline but then when you read the article it isn't actually "terrible" adivce at all, it's just that great content for the user is only one part of SEO.
I see above you have noted this point yourself Rand, but I would have expected someone with your experience to be better than that in the first place.
Not that you ought to care when some freeloading SEOmoz user says they are very disappointed in you, but I am anyway.
lots of comments due to can of worms being opened. I think that this is ultimately a different discussion about where the jurisdiction of the SEO guy ends. Is it at the click, or at the conversion?
If it is at the click then it is time to pull on that black balaclava and do whatever is necessary to get the rankings, because user experience is not be a consideration.
If SEO jurisdiction ends at the conversion, then it is all about the user, because once you get the rankings (actually the easy part when you think about it), every other bit of the journey is a seduction to encourage the user into a particular course of action.
I don't think you can separate them, personally. The conversion typically also means a good user experience and a satisfied customer, which yields more sharing and more links, which yields... better SEO.
I'm getting really worried that because of how I phrased the title and the intro to this post, the message I was trying to communicate (that you can't only focus on user-targeted SEO tactics if you want to succeed) has been lost... A follow up post is likely in order.
I agree with you on that point, that the responsibility for the user should lie with whoever generated it, and that as SEO professionals we should consider implications of any work that is done.
Its also important to think about the expertise of others - all too often, SEOs working on their own do not have access to the kind of usability/conversion experts that staff in larger agencies have. I'm lucky that I can call on other people to optimise user journey, but ultimately, the client is who I optimise for.
After all, it is the user who decides what they are going to search for, not the search engine.
Yes Rand, no need to create an artificial divide - if we are doing SEO for our clients, who is it really that pays for our services?
It's the users, right?
So, unless we all have a sweet govt contract or something, it's inevitably All About Conversions.
"focus on users" is a drum that has long been beaten loud and clear by googlers, but SEO tasks mentioned in this article also clearly points out what's necessary. It's so true that if I only focused on users, such important things like robot.txt, site submissions, and webmaster tools wouldn't even be in the mix. Well written.
This posted has clearly sparked a large debate – good one Rand for commenting back to us SEOmozzers!
In my opinion users and SE’s are important as each other, SE’s bring users and users bring extra business and this is why it should be a 50/50 split.
I really liked this article and found it interesting, but I think it has some interesting ramifications for SEOs. Most of the high value SEO activities, according to your graph, are really more on the technical/developer side than they are on the marketing/SEO side. I agree that manual link building, paid links, and directory submissions are not what they used to be, but what do you then propose for the marketing-minded SEO (in other words, the SEO who is not also webmaster/cannot implement technical changes)?
My previous employer (a web design firm who hired me to basically be able to offer digital marketing services) built sites that were not conversion focused at all. I was able to get the rankings/traffic up for many clients, however the visitors rarely converted.
Conversion Optimized has been my mantra for a while now which encompasses search engines, conversions and most importantly, the clients ROI.
Great post Rand (and also good comments from Danny). Can't figure out which was better, the post itself, or the comments and discussion that followed. Also a good reading the somewhat "heated" discussion between you and Danny. Insightful angles from the both of you, keep it up!
I think the thing to remember is that spiders&bots should be considered an "audience" you are writing for, but they are not the only audience of interest.
Write to please both your audience and search engines is the best way to go. It would be a shame if your post were looked past because it wasn't searchable enough.
Cheers,
Melissa Smich, Social Media Specialist, SoftCom
I agree that anyone who tells you that all you have to do is create content and the engines and all their bounty will be headed your way is not a person to take SEO advice from. If your content rules but the engines can't see it, it doesn't matter - it's the textual version of 'pics or it didn't happen'.
I also find it amusing that while looking at graphs that show the value of content creation growing over time to become the top factor listed in the first graph, that Rand is trying to downplay the value of content in SEO.
Your content dictates the terms that you have the best chance of pursuing - not in just a one-keyword way, but dominating and owning a space - and is the primary thing that drives people to link to your site. Should it be visible? Targeted to the right keywords in internal linking, meta data, etc? Yeah. But good content that feeds all of these things as the basis for a wider SEO program is absolutely in a site's interest.
Not sure where/how/why you'd consider "Engine-Specific Protocols", certainly the ones you've listed, to be the top engine-focused seo tactic?
Personally I'd say IA is a hybrid tactic, and more important for engine-focused stuff than the 3 examples you listed under "Engine-Specific Protocols"...
Of course this just boils down to my own personal opinion.
Maybe you went a little overboard on making your point. If one ignores SEO, that failure is probable but if one ignores user experience failure is certain.
It was a good article to show my boss though. He knows SEO is important but doesn't like most of the things that must be done to achieve it.
Chris
It is common sense that a website should be focused on the visitors, or you wouldn't make any business? the 2nd step is how to make it more visible on the web, and it is where we come into play!
Rand: thanx for the good article review of the different types of actions which are directly addressed to the search engines. ;-)
Once the users show up at a site, it should be for the users. But you've obviously got to get them there and that's by designing for the search results. Good points.
Hi Rand, Could you please explain the context of the graphs you give? What are the figures based on?
Thanks
These are, as always, my personal opinions. You can see from the comments that some folks certainly disagree about the relative placements of specific items. I don't mean to suggest this is the shared opinion of every SEO, but it does fit relatively well with the aggregated data from the ranking factors.
Rand,
I thought the graphs were a pretty good representation..
p.s. good article.. too many people didn't read all the way thru.. we ARE supposed to focus on bounce rate AND conversions but if Google can't find the right pages via webmaster tools, etc.. then it's hard to measure bounce rates and conversions w/o any traffic.
Good technical SEO + good website visitor (THE USER) analytics = getting to #1 AND staying there.
Excellent post though as with the previous comment, I would be interested in knowing the background to the graphs.
Brillant article, especially the graphics. Thanx.
Engines (t.m. the algo) are build by humans. They try to find the "most human faktors" or something like this. Optimizing for engines means to optimize for those engeneers ideas ("what is the best result for all people"). Sometimes this is in conflict with special target-groups. And for this we can use the "non-user-optimization" methods.
Nice post. The graphs are a great visual to see the relevance of each item over time. Thanks!
Hi Rand
As I'm relatively new to the ball game: "Thanks for giving me a quick overview on the evolution of mankind... erk... SEO."
Brilliant article. This is really a great overview of how SEO has changed in the past decade (plus some) and I completely agree with you. 3 thumbs up!
you have 3 accounts?
ok im sorry. it wasn't funny id delete it if i could just stop voting down plz im teetering on a hundred and cant go back to no-follow or id die...
but lets be serious saying this will only make it worse. right?
the first rule about thumbs is you dont talk about thumbs
and the second rule about thumbs is YOU DONT TALK ABOUT THUMBS
Maybe he ment his two real thumbs and one SEOmoz thumbs up? =)
LOL. Well if that's the case, then he should have counted his toes and said 5 thumbs up.
I personally dislike most adages/sayings etc.
We have been interviewing recently and if an SEO constantly fell back on catchphrases and analogies it did not go well for them...
To me it shows a lack of imagination and probably also a lack of understanding of the subject at hand.
Harsh maybe...
Anyway - I've never been a huge fan of design for user and the engines will love you.
I can think of loads of examples, particularly with regard to site structure/navigation where user friendly is definitely not SEO friendly.
Nice post - always love seeing those ranking factor trends graphs :)
wouldn't life be easy if we could still use 1997 tactics to rank #1 for "insurance"?! :)
Good imagery used to illustrate the changes in tactics, do you have any figures/stats to back it up?
I have to admit your title disappointed me when I first read it but it didn’t take me long to understand what you’re getting at.
Some people do have a tendency to go overboard with things whether it’s focusing too much on search engines and too little on the user or focusing too much on the user and ignoring the fact that SEO is called SEO for a reason.
You raise a good point but my worry is that some people, who focus too much on search engines and too little on the user, are going to read the title of your post and continue down the wrong road.
I absolutely agree that you need to be very mindful of the Search Engines desires versus the user's desire. Obviously once you get them to your page you need to ensure you have the content the user wants in order to accomplish your goal (i.e. conversions, sales, ebook requests, etc..).
Recently, I've seen crappy sites showing up organically as #1-3 because of the number of inbound links. So in regard to the order of your rankings, I do disagree (Vertical SEO should be #1). In today's SEO environment, I firmly believe the number inbound links gives you a higher ranking. For example, I and a competitor both have a YouTube video, mine has over 600 views, his has 80, guess who is in the first page results? Not mine (which one would assume with a higher number of views has better content, which of course it does!). Nope, Google thinks the one with a network of low user friendly websites (with sole purpose of generating inbound links to the companies primary website) are much more credible than having a few, high-reputation websites with content that is actually better for the user (hence the over 5x number of views). Now we are scrambling to upload our video's on Viddler, Vimeo, and other video oriented sites to get more inbound (albeit low value) links to our primary websites. Let's face i, YouTube owns the video market, why would links from low usage sites like Viddler and Vimeo, impact organic SEO.
The fact is you still need to get the user to your page, and to do that you need to design with the Search Engine's in mind (of course, you also need to meet the users needs, or you will not convert).
Interesting stats, it's interesting to look at it all in a different way like that. I would, though, still be aware of your title tag since it is used when people bookmark your page.
I just met with a designer the other day in regards to a new client and was told that he thought I was going to be "another hokey SEO guy" because content was the only thing important in his eyes. Until I explained the fact that this is a synergestic effort and unless traffic is driven to the site by the masses, then it does matter if there is content there or not and my clients marketing dollars might as well be flushed down the toilet...maybe this was a little dramatic. Content is King, but without SEO it is only going to get us so far and seen by so many.
At last - sound common sense.
I've also repeated the 'make the site for the user/visitor' - but always followed it with - 'but keep one eye on the search engines and SEO'
Filed under SCS - Sound Common Sense.
Take THAT Mr. Cutts (aka Matt)!
The title is pure link bait. The graphs are pretty but, as others have mentioned, "where's the data"? I agree with the "false dichotomy" assessment above. It strikes me a little bit as "talking to talk" and too "meta" for daily use, where is the actionable intelligence that we all crave?
Sean - fair point. I think there is some actionable intel here (pay close attention to engine-focused tactics, etc.), but it's not as specific or high quality as it could be. I've been working very hard this summer to improve the blog, but it's quite challenging - I'll try to do even better.
This was a great read but I have to disagree with it. Sure, all of those on-page elements mentioned are for search engines but honestly, I've seen many cases of sites ranking without those elements. If they have incredible content and people are liking it and liking it, they dominate.
I believe that Google is actually quite smart (imagine that...) and they seek to build their algorithims not around who can add the best title tag, but by observing how people naturally use the web. (i.e. people like an article, so they link to it, and Google recognizes that if enough people are linking to it, it must be really good). Sure the system can be gamed (for a few months) but over the long haul, focusing on making your users happy - with just a mild emphasis on SEO - has proven to be the best strategy.
This is the exact message that we have been blogging about and discussing with our clients. One must totally understand the needs of the client. Increased revenue is the bottom line. That can be achieved on the web by customer service, authority, affinity, promotion, conversations etc etc. But it all begins with traffic. If no one is coming to a website or blog, the items above are moot. It is a balanced approach and the ratio of emphasis percentage on SEO to User Content needs to slide as the process (and traffic) matures. That is why analytics, measurement and adjustment are so critical.
XML Sitemaps, Webmaster Tools Registration, Keywords closer to the front of a title tag, short, static URLs, shallow link architecture and good anchor text all help you rank better in engines (substantially), but you'd never do any of those things (or at least, wouldn't focus much energy on them) to help users.
I disagree with your point - I think just doing SEO for users and hoping that Google will eventually get smart enough to count those tactics is a mistake. They've created an ecosystem with rules and protocols and opportunities and wishful thinking that engine-focused tactics will one day be less important is a dangerous precedent in my opinion.
If I could truly build a page 'just for users' it would probably look just like a kiosk with images, prices, words (grahics)with lots of cool TV-like animation and interstitials for all content delivery. We get the joy of using straight text to convey our messaging.
Imagine a moving-magazine ad with a big pretty buttons. Okay, maybe that's out there a bit but I pose this observation on-topic: How about the fact that we have to use ASCII text characters (and readable character sets) in our pages?
Also note that we have to build additional pages, too! Why not just have a single one-link-to-rule-them-all campaign where I could deliver a server-side application that contains ALL information? Your 'Information Architecture' item touches on that, I think, but you get the point.
We all know a 100% flash site is sorta indexable but I don't think it's going to out-perform a text-driven website.
The fact that we use the combination of ASCII for word messaging is all for the search engines, right? Maybe I need some sleep.
Cheers on (another) great article. TGIF
(thank God it's free)
I wouldn't agree 100% there Garry. I agree there are certainly times where the only function a page needs to serve is the presentation of multimedia yet we find ourselves writing ASCII copy to appease the SEO content gods but I wouldn't say that text is not of benefit to the user.
Perhaps it is coming from a nation with an embarressing national attitude to broadband (Australia) but depending on your market a reliance on readable character sets may have some advantages if broadband penetration is not ideal. In my deep dark days of dial-up (thankfully many years past) I would avoid 100% flash sites like the plague.
Obviously the need to cater to the lowest common technological denominator will diminish in time and then it is my sincerest hope that a world where we are limited by text presentations in archaic formats purely because Search Engines deem it just and proper is no longer in existence.
Mind you none of this should read that I'm against 'text'. Properly written copy remains one of the most effective ways to apeal to a wide variety of visitor types with differing consumption styles (skim readers vs those who spend time reading the content in detail to cite one example). But I digress.
The usual high standard we have come to expect from Rand, full of great pearls of wisdom.
Excellent article Rand! Thanks for writing it! I am so tired of hearing that "build for the users" story...
I love SEO articles. Conjecture, hypothesis, probability, and unsubstantiated opinion, o my!
Hey Brain - while I greatly appreciate comments on the site, I would note that we try to maintain a very high quality of discourse on SEOmoz.
I love that you're being critical of the post - we definitely encourage that. But, it would be great to get specifics rather than sarcasm. ANd, to your point, conjecture, hypotheses and probability (when applied to correlation or statistics data) are all parts of the scientific method, too (and normally accepted parts of advice and recommendations in nearly every field).
My criticism is misunderstood: hypothesis et al is where the greatest minds of the scientific community come to play, as it's from the fiction that the fact will grow. Put the best minds together and have them invent the hypotheticals, and it gives the less creative types a place to focus their energies. What I like the most about SEO is that it requires people to make stuff up and figure things out.
Thanks for the charts Mr Rand... they are very descriptive and so is your post.
now would please tell me why the url of you post is not identical to its title?
url:terrible-advice-do-seo-for-users-not-engines
title:Terrible SEO Advice: Focus on Users, Not Engines
thks in advance
I would assume that the slug in the url was created when he first wrote the article and he then changed the title but the slug was not changed.
I wish neopunisher was right, but in this case, it was just a draft issue - the original title was different than the final, but I have done posts before that were more strategic about this, so don't be super surprised if it happens again :-)
i think its because hes good at SEO. @mattcutts example was in his post url: change-firefox-default-printer title he had:changing firefox default printer that way he could get results for both change and changing but iono thats just what it made me think about im prolly wrong and swami is right it was just a title change what was it rand?