SEO: Back to Basics

So now that we’ve thoroughly established that SEO is not dead and we all still have jobs, livelihoods, and incomes, let’s get back to actually getting some SEO done, shall we?

I caught a tweet last week of a fellow in Northern Ireland who, in all seriousness, described himself as a ‘seasoned SEO guy’. In this tweet he advocated the use of PageRank Sculpting. Naturally I felt inclined to put the matter straight and educate the poor chap on his dependence on outdated info. Unfortunately this is wasn’t the only case of ill-informed SEOs using outdated practices that I’ve come across recently.

So let’s discuss a few contested SEO issues and try to sift through the disinformation to distil some proper best practices.

1. PageRank Sculpting with nofollowed links

PR sculpting is the practice of using the nofollow tag on internal links to direct the flow of PageRank through your site. The idea is that by using nofollow on a link, the PageRank attributed to that link is instead redistributed to other links on that page.

The problem is, it doesn’t work. This practice has been thoroughly debunked over a year ago – the PR doesn’t get redistributed, instead it just evaporates. Even Matt Cutts has chimed in on the matter and has stated in no uncertain terms that he really doesn’t recommend using nofollowed links on any internal link, unless you really don’t want a page to show up in the SERPs.

2. Title Tags character limit

I was involved in interesting debates on this matter on a forum and a blog post comment thread. There seems to be some uncertainty among even experienced SEOs about what the best length is for title tags.

The consensus seems to be that due to the Google SERPs cut-off at between 65 and 70 characters, it’s best to keep your title tags to 65 characters or less. Some even go as far as to state that Google doesn’t index more than 65 characters of a title tag.

That is, of course, bullshit. Some valuable tests done by the likes of Hugo Guzman, Shaun Anderson and Paul Carpenter show that Google indexes pretty much the whole title tag, regardless of how long it is. That’s not to say it’s a good idea to have long title tags, but it does goes to show you shouldn’t be too focused on that 65 character limit.

Make sure it’s a good title tag with your most important keywords at the beginning, and don’t worry too much about staying within the limit.

3. Meta Description and Keywords

You’d be surprised at how much disinformation there is about meta tags. The meta description tag for example is often misunderstood. Some SEOs say that the description tag is a ranking factor.

It’s not. The meta description tag gives no ranking benefit. What it does do however is help with click-through rates on SERPs. Most of the time when your site appears on SERPs, the meta description tag is what is shown as the site snippet text (though not always, sometimes the search engine will choose to show an excerpt from your page if it thinks this is more relevant).

A good meta description can ensure you get more clicks from SERPs so it’s a good idea to optimise it. But it’s not a ranking factor.

The meta keywords tag is a different story. There’s very little misinformation around, everyone seems to know that meta keywords provide no ranking benefit whatsoever in Google. Many SEOs choose to omit the meta keyword tag entirely.

I’m not one of them – Google might ignore the meta keyword tag, but there are other search engines out there, and some might look at it. I tend to include the meta keyword tag on my pages, but I don’t spend too much time on it.

4. Paid Links are Bad

White-hat SEOs tend to frown on paid links and proclaim, often quite loudly, that you should never use paid links as they provide a temporary benefit at best and will eventually get you caught and penalised.

This is another load of bollocks. Yes if you’re stupid and pay for obvious spam links, you deserve to get banned from the SERPs. But the fact of the matter is that paid links work. They work really, really well.

Paid links are a big grey area anyway. Is a link on a paid directory site bad? Is a link embedded in a press release a paid link? You pay for the distribution of the PR, so effectively you’re paying for that link. Yet no search engine that I know of will see links embedded in press releases as spammy paid links.

Investing in paid links is usually a good idea, especially for young sites and/or competitive niches. It shouldn’t be your primary focus – be sure to invest in a solid natural link profile as well – but eschewing paid links entirely will probably put you at a competitive disadvantage.

5. Social Media Sites and SEO

Many SEOs still haven’t quite wrapped their heads around social media websites and what they mean for SEO. They needn’t worry too much: for purely SEO purposes social media isn’t all that useful. Most social media sites nofollow their links or hide them entirely, and poorly coded social media buttons can actually harm indexing on your site (not to mention load speed).

Having said that, social media does have its purposes, especially when it comes to content discovery. A link posted on Twitter that gathers some retweets will be crawled by Google within seconds. If you want to get content out there quickly, there’s no substitute for social media.

There’s also some indirect benefit associated with social media websites. If for example your content gets picked up by sites like Digg, Reddit, and StumbleUpon, you’ll not only receive an immediate traffic boost but you can also expect some extra natural links as bloggers write about and link to your content.

If you have any SEO misunderstandings you want to share, leave them in the comments.

About Barry Adams

Barry Adams is one of the chief editors of State of Digital and is an award-winning SEO consultant delivering specialised technical SEO services to clients worldwide.

21 thoughts on “SEO: Back to Basics

  1. Every now and then it’s nice to have your standard practices confirmed as best practices. Great summation of these points, Barry. And you’re right about social media. I came to this post via Twitter (my RSS Reader alternative).

  2. Hey Barry! Right on brother… the debate on title shouldn’t be how many chars but what is “weighted as title” because test indicate Google indexes all the title. The only real way to test is to use the “intitle” operator that indicates truncation point as cutoff for what is “weighted as title” is perhaps a tad pessimistic.

    Paid links are advertising and I wouldn’t listen to a radio salesman who tells me I can’t buy telvision ads… why would I let Google tell me where I can advertise it is beyond my comprehension why people allow a Search engine to dictate their advertising policies.

  3. Hey Barry. I agree with pretty much everything you have said there except one thing that I have always been a little muggy on. the whole follow/no-follow links idea has always made my head hurt and its for this reason (I know Andy Beard will probably kill me for this but here goes, before I go into it though I like the idea of do follow).

    Lets take WordPress. Standard install it is nofollow all the way along the link structure, be it internal and or outbound. you can pro-actively make the links do follow with plugins or coding. Who does bar the real hardcore advocates?

    Look at any “normally” designed website and its links will be nofollow unless it has been definitively decided that they should be do-follow. If this is the case in the very vast majority of websites out there, then …how can do-follow matter?

    The links that people are achieving are working anyway, so why get on the do-follow boat?

  4. @Justin, when it comes to using nofollow on blog comment links, that’s a different story. First they’re external links (and PR sculpting is about internal links), and second nofollowing blog comment links is a good practice imho to discourage the link spammers.

    I don’t know what version of WP you’re using, but all my installs have always had all links followed except links in comments.

    And CMSs like WordPress are intrinsically SEO-unfriendly. I have yet to encounter an inherently SEO-friendly CMS. WordPress gets around it with plugins and smartly coded themes, but these are still hacks to correct the flaws of an inherently flawed system.

  5. Hi Barry, well written article. On the point above, it depends on the theme as to whether links are nofollowed. Out of the box WordPress has a standard theme which does not add nofollow attributes to internal links, but I have come across themes which do in fact add nofollow to internal links which will cause the above issue when installed. This can as mentioned be rectified with rudimentary coding knowledge and a bit of patience.

    Fortunately, this is not the case for “the vast majority of websites”. See this data from SEOMoz from July 2010 which states that the percent of links that use Rel=Nofollow was calculated by them to be 2.08% of all links on the Internet.

  6. Thanks for the mention, Barry, and nice job on this article. It’s always good to tackle controversial/conflicting SEO topics head on.

    (and test, test, test)


  7. Hi Barry,

    You take the time to attach ‘through no follow” to the tactic of “link sculpting”, why not talking about link sculpting done through Java Script?

    At SEO Samba, we’ve released the open source CMS SEOTOASTER with a very effective point& click interface to build link silo through the use of Java Script. It is therefore now accessible to the masses and it works beautifully. Here is a link to a short video here:

  8. @Michel link sculpting through JavaScript is better known as dynamic linking. I didn’t want to muddy the waters too much and instead chose to stick with what the SEO masses (correctly or not) associate with the link sculpting terminology. Po-tay-toh, po-tah-toh, and all that.

    And dynamic linking via JS is in many ways a form of cloaking, which should be handled with care and deserves more thorough examination than I can do it justice in the broad-sweeping, generalised blog posts I’m known for. 🙂

  9. @Barry. Got it. I however disagree with the cloaking association.

    Intent is what matter, and I recommend using LS via JS when it makes the robots perspective closer to the actual click distribution on website. On most sites, menus collapse and links therefore not click-able. However robots understand all links to have equal chance of being clicked which is obviously false, and then triggers improper distribution of link juice.

    Link sculpting with JS can help correct robot’s skewed vision. I can’t wait to debate Matt Cutts on that one.

  10. @Michel, I understand where you’re coming from, but technically showing different things to users than to search engines is cloaking, especially when it is done to ‘manipulate’ search engines – even when you think its a benevolent manipulation. Darren Slatten (aka SEOMofo) posted a question along these lines in Google’s webmaster help centre and didn’t receive very encouraging answers (

    Also, not all links are equal for search engines. There is a growing body of evidence out there suggesting that Google attaches different weights to links depending on where they appear on a page – see the (in)famous ‘Reasonable Surfer’ patent (

    I’m not against cloaking per sé, but I do think it’s something that needs to be carefully considered and handled delicately and appropriately.

  11. Barry, I think you’re missing the point.

    Darren talks about manipulating outbound links to affiliate properties. The so-called user experience benefits built around load-time is rather thin. In addition, the only use is to prevent robots to see some external site links that users do see.

    On the other hand, we recommend using link siloing in order to re-conciliate what robots see with what users actually see. It is in fact quite the opposite intent o what Darren is preconising.

    Users do not see links contained in collapsable menus, hence they can’t possibly click on them. Why should robots see and count these links?

    As for your point about link value, note also that the point & click siloing interface provided in SEOTOASTER builds silos on your own website, not to external sites. Nothing in that patent suggest that somehow your menu links should be discounted/penalized by their algorithm.

    Finally, a point of agreement, as I said all along, this feature is provided as a way to correct the flawed vision of robots.
    You mention appropriateness, here is the acid test i propose, and that is whether the link graph after applying the silo is closer to the actual user click distribution. If that is the case, then you’re off the hook.

  12. @Michel, oh no I’m not missing the point – I understand perfectly what you’re saying, I’m just trying to make you aware of how a search engine could perceive what you’re doing.

    The bottom line is that you’re showing users one thing and search engine spiders another thing. That is inherently suspicious behaviour, regardless of your motivations.

    Search engines don’t think it’s an SEOs job to correct their link graph – the Googles of this world believe themselves to be quite capable of doing that for themselves. (Whether that is a correct POV or not is an entirely different matter.)

    I see you passionately believe in your own sales pitch (which is a good thing), but I do feel you need to understand – and communicate to your customers – that what you’re selling has risks associated with it. You may not agree with the legitimacy of those risks, but that doesn’t mean they’re not there.

    Re: your example of collapsible menus, I do advise you to read up on that ‘reasonable surfer’ patent – search engines are getting cleverer every day at identifying the context of links and adjust the value assigned to those links accordingly.

  13. For the moment I’m focusing on just basic SEO, and content generation. Probably the best seo work possible 😀 nice read.

  14. Barry – Nice reminder of the basics. Like Bill states above, it’s always nice to have your standard practices confirmed as “best practices”.

    Thanks for the share!

    BTW – I had only tried PR sculpting once and I never really saw any benefit from it. So I’m gonna have to agree 100% with your assesment on that part.

Comments are closed.