SEO Myths Reloaded: Clarifications, Consensus And Controversy
I have to say I wasn’t expecting the impassioned responses to my last article, 36 SEO Myths That Won’t Die But Need To. But that’s okay, as I wasn’t expecting all the positive buzz and re-tweets either (1,153 and still counting — how exciting!)
In this follow-up, I’d like to revisit some of the most fervently defended myths, clarify any potentially vague statements, and provide the occasional caveat.
Myths are born when folks mistake correlation with causation. And when they make inferences and draw conclusions without rock-solid data or methodology. Next thing you know a “feeling I’ve got about this” is espoused as fact.
It’s the nature of myths that they aren’t easily dis-proven or dismissed. So the myths persist.
And as Rand Fishkin so astutely recognized, there is strong incentive for someone to defend a myth if they had advanced that myth previously to a public audience, boss, client, etc. It’s self-preservation instinct, to “save face.” Consequently, some folks hold on to certain myths for dear life. Their very careers hang in the balance — or so they think.
But I’m not infallible either, so I wanted to get confirmation from at least a dozen other industry veterans that I’m not off-the-mark. For the most part I received vigorous agreement. Adam Audette summed it up nicely:
I can’t believe the hornet’s nest you stirred up with that post! I was THIS close to commenting on that post, but didn’t. I was going to say that I agree completely w/ your list and that everyone disagreeing needed to SIT DOWN :) Decided I’d take the easy route and stay out of it….
A hornet’s nest indeed! Well put.
Rand did a great job debunking the myths where I faced the strongest resistance. I won’t recap and rehash all that here, but I encourage you to read his post.
In the comments of my previous article, I received criticism that I was merely providing assertions rather than citing research and disclosing hard factual data. If I fully backed up every point with enough research to satisfy everyone, the article would have turned into a 10,000 word tome. Fitting for a white paper or ebook, not so much for an article/post to Search Engine Land. As it is, this follow-up article, at over 5,000 words (divided into two parts), has exceeded my last one.
My choice of studies to cite was also brought into question. It’s always possible to find flaws and things to pick at in someone else’s study, but in my opinion, having some research is a lot better than having none. Otherwise you’re flying blind, drawing only from your own observations, conclusions and hunches. In general I think we’re all relying on flawed data. Would anyone’s SEO experiment truly stand up to scientific scrutiny?
There are so many moving parts — so many influencing factors — that we can’t control or isolate out of the system when conducting experiments. Search is in a constant state of flux — the algorithms, indices, competitors, data propagation across data centers, etc. Given this, how can one possibly create a proper control group for scientifically rigorous experiments? But we do our best. That’s SEO. It’s the nature of the beast.
Now let’s address some of the SEO myths on which I was challenged…
Myth #2: Don’t use Google Analytics because Google will spy on you and use the information against you.
I’m not one of those tinfoil hat conspiracy theorists, but who knows, maybe someday I’ll be sorry that I wasn’t. Michael Geneles says:
“I am definitely NOT the most paranoid SEO out there… However, I’ve been burned enough times to use any Google services (Google Analytics included) on some of my “more aggressive” experimental sites. Thank god for Piwik. Some of us don’t want Google to learn our IPs and have the ability to make the association between different web properties that we play with. That’s why I never “touch” my client’s Google accounts without a proxy.”
Myth #3: Your PageRank score, as reported by Google’s toolbar server, is highly correlated to your Google rankings.
For further evidence, there is some exact data refuting this myth; you can repeat this process yourself and you’ll get similar results.
Myth #4: Having an XML Sitemap will boost your Google rankings
Google goes on record to say on the Google Webmaster Central Blog that “A Sitemap does not affect the actual ranking of your pages.” It doesn’t get any more explicit than that. One caveat that Ian McAnerin notes (and this partially relates to myth #14 about fresh content): Until your site is fully or mostly indexed, you won’t usually rank as well as you should. Therefore at first you will often see an improvement to your rankings if you have an XML sitemap and/or frequently updated content (and enough links to justify the visits to that content), but this is just due to faster indexing. Once a certain “critical mass” of your site is indexed, there will usually be no further effect. Ian’s never seen a difference when he does this for established sites, only brand new ones.
It’s not about the updates (which encourage more visits, and thus more opportunities to index new pages) or the XML — it’s the indexing. You can get roughly the same effect (plus link weight and anchor text benefits) simply by doing a link building campaign instead (or in conjunction with).
Myth #5: Since the advent of personalization, there is no such thing as being ranked #1 anymore because everyone sees different results.
I contend that you are unlikely to see a drastic reshuffling of the Google results from personalization, particularly if this is not a query that you repeatedly make. A single website may shift drastically in a particular SERP when personalization is on. Nonetheless, a head-to-head comparison of a page of SERPs personalized and non-personalized will return mostly the same results with minor shifts here and there. A listing may disappear or be replaced, but we’re not talking about a completely different set of results being swapped in and out.
Even for such an ambiguous query as “dolphins” (do I mean the football team or the cute sea mammal?), the first page of results doesn’t change drastically even if I do a bunch of football related queries first, such as “football”, “patriots” and “nfl”. I still get mostly animal-related results, with a bit of movement in the lower half of the page. #6 shifted to position 7, #7 shifted to position 8, and #8 shifted to position 6. My point: if you’re ranked in the #1 position with personalized web search set to 0 (“&pws=0” added to the end of the Google SERP URL), I think you’ll find, generally speaking, that you’ll be at #1 for the majority of searchers, whether personalization is enabled or not.
Mike Moran states:
“What I think is true is that, for many queries, a single search result is #1 the great majority of the time, but I think it is less true than it once was and that ranking reports are not nearly as valuable as they were a few years ago. It’s better to focus on traffic and conversions and not get too hung up on rankings.”
Myth #6: Meta tags will boost your rankings.
Note that I did not say that ALL meta tags are a waste of your time. Meta descriptions (not meta keywords) are still worthwhile, even though they won’t improve your rankings, as they can influence the snippet that’s displayed as part of your Google listing. More on how this works in my article “Anatomy of a Google Snippet“.
Myth #9: Having country-specific sites creates “duplicate content” issues in Google.
One caveat that Thad Kahlow notes: Google is much more likely to handle this well if the domain is set up on a ccTLD or if it’s been set up in Google Webmaster Tools under a common folder that allows for the webmaster to set the country specification.
Myth #14: It’s important for your rankings that you update your home page frequently (e.g. daily.)
I think freshness is mildly helpful but generally not required to maintain a high ranking (QDF “query deserves freshness” searches notwithstanding). Particularly if you have strong domain authority, you can maintain a top position in Google for a very long time without updating the page content. Take for example http://kidshealth.org/kid/ill_injure/sick/food_allergies.html which ranks on page 1 in Google for “food allergies” — yet this article hasn’t been updated in years.
That said, I wouldn’t dissuade you from keeping your content up-to-date. Indeed, regular updates are good for users and can make a site more link-worthy so that it attracts links at a faster rate than the competition. Just consider that there are a limited number of hours in the day, so your time may be better spent proactively going after quality backlinks than incessantly updating the page content.
Myth #15: Trading links (en masse) helps boost PageRank and rankings.
Building a massive reciprocal link network comprised of irrelevant sites is at best a waste of time and at worst an invitation for a penalty. Generally speaking, reciprocal links aren’t necessarily worthless, but they sit pretty low in the food chain. There are of course exceptions. Nevertheless, I’d always steer clear of link farms and link exchanges. I wouldn’t turn my nose up to a relevant reciprocal link from a trusted site, but my preference would be for a one-way link from that trusted site. Make sense?
Aaron Wall‘s position on reciprocal links is that as long as the trades are strategic, with quality sites, and don’t look like run of the mill trades, then they can help out a lot. He says this is especially true if they are part of a broader strategy (rather than being the entire strategy).
Myth #16: Linking out (such as to Google.com) helps rankings.
In my experience, the benefit of an external-pointing link is minor to nil. I am not alone in this view. Major caveat: I’m assuming there isn’t anything abnormal going on in your link neighborhood. If, on the other hand, you’re hoarding your link juice, or linking out solely to low quality sites, or your site’s nothing but broken links, then by all means correct that with some quality outlinks and I bet you’ll see an uplift. Michael Geneles has done some testing on this, and it is his opinion that outbound linking behavior does have some effect on rankings. He of course doesn’t recommend link hoarding, and he prefers to link out to sites with equal/higher mozRank.
Myth #20: H1 tags are a crucial element for SEO.
H1 tags used to be more important in years past, but as of late, they are only a minor signal. Prove it to yourself: try taking a site with headlines marked up with a font tag and turn that tag into an H1 in the template — without changing the copy within the headlines (or anything else for that matter) — and see what happens to your rankings. I think you’ll find the result to be not very impressive.
I’m not just basing this on anecdotal evidence, or on someone else’s study. Here at Covario, last year we did our own correlation study of page features that influence ranking. We did it to validate the scoring methodology for our Organic Search Insight product which our clients use to audit and track their sites’ SEO health. The 17-page report of findings is only for internal distribution — sorry, I can’t share it! The data collection and statistical analysis were rigorous though. We found H1 tags towards the bottom of the list of signals we examined, just above bold/emphasis tags.
If you’re getting benefit from H1 tags, consider that it may be because of the headline’s keyword prominence in the HTML rather than the fact your headline is inside an H1 container.
Myth #31: Home page PageRank on a domain means something (important).
I admit I’m being a little bit provocative here. This isn’t to say that toolbar PageRank is completely useless. At a minimum, PageRank scores have entertainment value. Seriously though, I just wouldn’t bet my business on that little green meter.
And since when is describing an entire site by a page-specific metric such as PageRank somehow accurate? As an indicator of domain-wide importance/trust/authority I’d take domain mozRank and mozTrust scores any day over a single page’s (truncated, possibly randomized) PageRank score.
A home page’s PageRank score can’t indiscriminately be extrapolated to internal pages. It’s not a given that a PageRank 4 will lead directly to a PageRank 3. The PageRank dropoff could be abrupt and marked if the site architecture is a mess. It’s a sad state of affairs when a high-PR home page stands alone among an army of low- or no-PR internal pages, but it happens.
Another key point about PageRank: the higher the integer score, the more seriously you can take the score (thanks to the logarithmic nature of PageRank). A toolbar PageRank of 8 is way more meaningful and less likely to be off by several integers from the true PageRank used for ranking purposes, versus a PageRank of 4.
Minimally, if the page registers some amount of toolbar PageRank, it is 100% guaranteed to be crawled, and most probably it will be indexed too — according to Hamlet Batista.
Myth #33: The clickthrough rate on the SERPs matters (for ranking.)
…should any SEO ever discover that it substantively impacts rankings, we’re going to be faced with an army of zombie botnets trying to take over our computers not to send email spam, but to click on links through our “reputable” Google accounts. Just look at the hacks of Facebook, Twitter & WordPress over the past few weeks and ask yourself – if any spammer could show any financial incentive or ability of clicks to influence Google, would we really have as (organic) click-fraud free a world as we do today?
Personally, to date I’ve seen nothing to indicate that CTR influences rankings. If anyone has evidence to the contrary, I’d love to hear about it. I’m not precluding that Google may some day use CTR as a signal.
Michael Geneles sees it differently. His view is that a good CTR combined with a low bounce rate “helps to show proof of relevance and user engagement.”
Myth #34: Keyword density is da bomb (i.e. measure keyword densities of competing listings and aim to match them.)
Honing in on a particular keyword density value while editing page copy is a colossal waste of time, a distraction from what really matters. Granted, keyword density measurements can help identify extreme situations — such as where there is no use of the keyword, or the term is being spammed into the page too often. But outside of those outlier scenarios, what is the point? There is an ideal score to aim for. There is no value in benchmarking against competitors’ keyword densities. If your SEO consultant tells you otherwise, turn and run!
Myth #35: Hyphenated domain names are best (preferable) for SEO.
Certainly a hyphenated domain can still perform well in the SERPs. The operative word here is preferable. Although a hyphenated domain can be an acceptable substitute when better choices aren’t available (such as a hyphenless .com, .org, or .net), I would never favor a hyphenated domain over its non-hyphenated counterpart if both were available. Years ago I bought information-architect.com, but I would have bought informationarchitect.com instead if I had the choice — for reasons of branding, usability, and SEO. The more hyphens, the spammier the domain looks, and the less desirable it is. To me, the domain san-diego-real-estate-for-fun-and-profit.com is not at all appealing.
Caveat: I wouldn’t be very keen on the unhyphenated domain therapistfinder.com for a directory of psychologists, for example; I’d rather have the hyphenated version — thank you very much.
Myth #36: Great content (automatically) equals great rankings
This is not a foregone conclusion. Great content doesn’t necessarily rank because it’s great content. The content may deserve to be ranked, but if no one knows about it, or if the site architecture is so atrocious that it repels the spiders, then it won’t rank. It’s as important to actively promote that great content as to have created it. I’m simply making an argument against that tired old phrase “Build it and they will come.” Don’t let these comments dissuade you from creating high quality content. Indeed, it’s a likely prerequisite for SEO success, especially when the keywords being targeted are highly competitive.
Read part two of this series, 36 More SEO Myths That Won’t Die But Need To.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.