Questions I’d love to have answered about SEO

When done properly and in a serious manner, SEO is truly a fascinating field! However, I have some wonderings about how Google and other search engines treat and index certain things.

What SEO means to me

First, let’s acknowledge the fact that the world is riddled with shady companies working with SEO, trying to make some extra bucks. Common implementations are link farms, lots of hidden text for indexing etc etc. As with any business/technology, there will be people around in it only for personal gain, and without realizing that good business ethics will pay off in the long run.

But, luckily, there are also people and companies who care about doing it the right way, and not just some quick fixes/hacks that might work temporarily. Last year, I attended the Web Standards and SEO: Searching for Common Ground, Part 2 panel at SXSW, and what was thoroughly pleasing is the reply the panelists gave to the question “If you could do something which might improve your SEO ranking, but that would hurt the accessibility of the web site, would you do it?”.

The reply from all of them was no; making it accessible is what counts and eventually results in good SEO as well. If something inaccessible helps you in search engines, it’s just a temporary phase, and it will be removed in the future. And that’s exactly my standpoint as well: create good web sites with semantic code, good structure and outstanding copy (keyword density, explanatory texts and so on) and you will get a good search engine ranking. The only thing to add to that is naturally backlinks to your web site, but with good content, it is bound to happen sooner or later.

I’ve written more about my personal experiences in How to get a good search engine ranking.

The questions

With all that said, there some certain scenarios that I wonder about which is the best and recommended way to achieve them. I would love to have access to someone like Matt Cutts to get direct answers and guidelines. Not just for me, but for the world; to know what we all can do to have accessible code that is also indexed as appropriate as possible.

How to mark up a preamble/introduction?

Let’s take a normal web page, and break down the general content part:

  • Heading
  • Preamble
  • General text, eventual subheadings

The heading will be an H1 element, displaying the heading of the web page, the preamble and general text will be within paragraphs (P) and any possible subheading will range from H2 and on. My wonder here, though, is how to put any extra emphasis to the preamble, which is often a very important text. Sure, it will, to some degree, mean more since it is directly after the main heading and as far up in the code as possible.

But is that sufficient? I’ve been thinking about if the preamble text would be wrapped in an em or strong element: would that be correct, to make it stand out as it should? Or would that seem like over-optimization and maybe even be punished?

Sure, copy is king, but what if I want to complement this with a good semantic representation in the code?

Hidden text

In many web sites, a page (for instance, the start page) can contain information within the code that isn’t initially visible. The way to get to see it is either via client-side scripting, for those with JavaScript enabled, or regular links that will send you to a web page where that content is presented. However, this text can be just as important as the other text in the page.

Is this text indexed, even though it is most likely hidden via display: none? And if it is, is it regarded to mean just as much as the other text in the page, or less? And, finally, how does Google and others differentiate proper hidden text from hidden text there for solely spam/SEO purposes?

Cookie-based start pages

From a usability and customer perspective, it is pretty common to have a start page presented to first time visitors (and, consequently, search engines), and then give returning visitors a slightly different version, based on a cookie. Would that be seen as cheating, although it is for the best of the visitor?

Freaky onclick implementations

I’ve seen examples regarding banners on web sites, implementing a, to me, shady method. The idea is that the href part of the link points to the start page of the web site, but that it has an onlick event applied that takes you to the actual URL that the link should take you to. For example:

<a href="http://www.google.com" onclick="location.href = 'http://www.google.com/?id=ksldfhsklHKHH87Jjhi'">The banner</a>

The reasoning behind this approach is to get more links directly to the start page, trying to give an impression that it is more popular/interesting than it actually is. Personally, I think this is a crappy and incorrect way to do it (except for the fact that it wouldn’t work well for users with JavaScript disabled, of course).

Give me help!

If you have any hard facts or personal experiences about these, if you work for a search engine company (or if you’re listening, Matt :-)), please let me know!

 

Related reading

Firefox SEO add-ons

Posted in Developing,Google,HTML5/HTML/XHTML,Technology |

27 Comments

  • Nick says:

    Robert -
    I had a buddy who worked at the SEO firm icrossing, and I remember discussing the Hidden Text question that you posed here. What he said was that Google is getting smarter at recognizing hidden text. They are able to tell if a block is hidden via display:none or the color of the text is extremely similar to the background through CSS. He did say that you are able to accomplish the same thing via client side javascript though, and search engines would not be able to notice this. If I was to do this type of practice, I would hide the functions inside external JS pages instead of doing inline DOM manipulation (at least make the search engines work a lil to find out you are scamming them ;).

    Anyways, we talked about this a few months back, so Im not sure if everything still stands ecspecially since SEO has recently become so popular. Furthermore, I have not tried any of these methods myself as I rely on Google for a good portion of my traffic and don’t want to risk being banned. Id be interested to hear if anyone else has actually implemented any of these techniques.

    -Nick
    agridwork.com | free advertising and link exchange for bloggers

  • Nick says:

    Also, just found this:
    http://www.micropersuasion.com/2007/03/report_says_aja.html
    When reading through my daily news. Further confirmation that Google can’t parse Javascript…

    -Nick
    agridwork.com | free advertising and link exchange for bloggers

  • [...] uestion “If you could do something which might improve your SEO … Original post by Robert Nyman This entry was [...]

  • I can tell you that the way you describe making the "preamble" should work fine for on-page ranking factors.

    You do realize that Matt Cutts always says one basic thing: "Build websites for the users". Google hasn't changed from that philosophy in the whole time they've been online.

    If you write text-heavy pages that emphasize the main concepts concerning your "targeted keywords", Google does a pretty good job of ranking the pages. As the competition gets tougher, the tricks start to matter a lot more. But almost anyone can get "long tail" rankings just by writing indepth content that contains authoritative outbound links and semantically dense copy.

  • telga says:

    You're not alone in having uncertainties about the most effective way to mark-up the preamble or introduction, which I consider essential content in the header (or "branding" or "site description") section. If the preamble is short, I have put it in a span in the h1 element; longer descriptions have been enclosed by blockquotes; and I've also used the p and strong elements together as you suggest. Lately, considering the heading structure of the entire page, and using the document outline feature of the Firefox add-on web developer toolbar, I've been embolded to markup the preamble as an h2 heading, seeing it as a legitimate subheading of h1 in the site–or page–description section. From an accessibility point of view at least, or even if accessibilty and SEO aren't exactly co-terminus, I think having a short, highly-descriptive, hard-working preamble as an h2 heading must be helpful in determining what the page is about–and if it is worth continuing on through the heading structure.

    As for hidden text, I've replaced display:none with 2 span classes, marked-up semantically as "more-information" and "description" and put them off the page for visual users using position: absolute, left:-9999px. I use "more-information" to include information in headings that would be redundant for visual users and as h2 headings for navigation and footer or "site-description" sections. Conversely, using a:hover, I can also make the content in the "description " span available to visual users. Although I am hiding text from visual users, I don't see anything illegal about it, since it is not contextually different but just text that is marked-up differently.

    In all this, I haven't been pursuing any particular SEO strategy, but rather, thinking about usability and accessibility–No hard facts here, Robert! And, while I agree with Darren that content is king and search engines follow, in practice, effective content is also about worrying things, like how to mark-up your introduction.

  • I don't like any of telga's suggestion for the intro text. Span inside the H1 or the whole intro text inside an H2? Ugh. The preamble is not a heading and shouldn't be marked up as such.

    I wouldn't mark up the whole intro with or tags but rather mark up the keywords within the preamble appropriately. This keeps things semantic while also drawing attention to your opening paragraph.

    For example:

    "When done properly and in a serious manner, SEO is truly a fascinating field! However, I have some wonderings about how Google and other search engines treat and index certain things."

    "Things" isn't a great keyword but you get the point.

    Personally, I'd vote for an <introduction> or <summary> tag with the same behaviour as a tag but with further semantic meaning. Until then, I'd avoid the hacks.

  • Ross Johnson says:

    On page SEO is getting much less weight that it has in the past, and I wouldn't worry so much about giving different parts of the text extra emphasis over ensuring that some how or another google can index the content. Google does a very good job at figuring out what is and isn't relevant on the page regardless of good/bad/any mark-up.

    There is a lot more emphasis on domain/site trust than on page factors now days. That is not to say that they don't make a difference, just that you should do the best you can with in good practice and then start thinking about off page strategies.

  • Binny V A says:

    For marking up the preample, I use a with a 'intro' class.

  • The thing I find with SEO is that people make it seem harder to do than what it is. When I first started to learn about it (still am in a sense), I was receiving all these tips that just seem silly until I got the same good advice that was mentioned above "design and build for the user". Since doing that and following the basic rules of accessibility, I have found my website growing up googles ranking.

  • Robert Nyman says:

    Nick,

    The thing is that you can probably always get the text indexed, some way or another. But how do you make sure that it get s the same value/weight as all the other text?

    Darren,

    <blockquote cite="http://www.robertnyman.com/2007/03/14/questions-id-love-to-have-answered-about-seo/#comment-42488"&gt;

    Build websites for the users

    Oh, absolutely, and I agree a 100%. But when doing that, it is good to know if there are any important semantic factors that will, as the same time as making the web site more usable and accessible, also get the most correct and suiting search engine indexing.

    telga,

    I'm not sure about having the preamble in a <code>H"</code> element; if it's a proper sub-heading, sure, but if you have one or several full sentences it isn't really a heading anymore, but a regular paragraph (although more important than the following ones).

    But, like you say, accessibility and usability overall.

    Aaron,

    That sounds like a good approach and one that I've been thinking of. However, at the same time, it looks like exactly the same way a spammer/cheater would tackle it, and how would search engines know how to differ those two?

    And your examples of new semantic elements are perfect, and I wish we had them!

    Ross,

    Good input. I'm still that good semantic code in conjunction with great copy will help a lot, but absolutely, domains and trust are indeed important factors.

    Binny,

    Semantically, that sounds great. I just wonder to what degree search engines parse and analyze semantic class names, though.

    Jermayn,

    Yeah, definitely. People will shower you with tips, and it's only a few of them that are actually good ones. But hey, that seems to be applicable for any area in life, right? :-)

  • [...] d love to have answered about SEO – Robert s talk Here you will find posts about news, trends and developing f [...]

  • telga says:

    Robert and Aaron,

    I just read, at USC Annenberg Online Jouralism Review, http://www.ojr.org/ojr/stories/070312ruel/, the results of a Nielsen eyetracking study where Web pages were re-designed for faster reading AND better retention or comprehension. Interestingly, in the first example what appears to be a paragraph-long preamble is stripped out and replaced with a subheading. Overall, the recommendation was to increase white-space and use more subheadings and bulleted-lists. What this amounts to, of course, is there being less text in each section and on the page.

    The question arises, if less text in an area is a usability feature–for everyone? or for visual-content users?– does this conflict with SEO strategies and "text-heavy" or keyword-dense copy-writing?

  • eugene says:

    imho, make your site for your visitor (eg. relevant content). That, in combi with semantically and structurally good website, will be 'eaten' by Google anyway.

    And if you have really relevant content, eventually other relevant sites will start linking to you, there goes the PageRank up.

    Simply follow the (google) guidelines and your site will be indexed, by any search engine.

  • Robert Nyman says:

    telga,

    Interesting! Thanks for the link, it is well worth to ponder about this.

    eugene,

    Oh, absolutely. But everyone wants to be #1, right? Both in accessibility/usability as well as search engine ranking. :-)

  • Jules says:

    If you build your page correctly, it will get a higher ranking in a search engine but that takes time. There are people/businesses who are releasing a new site/product/service and want their site pushed to the top of the stack as quickly as possible. How would this be accomplished?

  • Robert Nyman says:

    Jules,

    In my experience, ranking goes fairly fast, if you have better code and copy than your competitors (being mildly popular also helps, of course :-)).

    But yes, what you mention is a very common problem: people expect to be no. 1 in search engines immediately, and more often than not, uses shady methods in some desperate vain. In the long run, choosing the correct path together with patience is the only valid option.

  • Kiper says:

    Well, SEO is here to stay and will probably get more important by the hour. So the "shady" guys will get a lot of money from people that are in a hurry to get to no. 1 in the search lists.

    Hopefully Google and the others will be able to take measures against this or the search engines will loose credibility.

    I'm afraid that the SEO business will attract the same type of guys as the virus crowd…

  • Robert Nyman says:

    Kiper,

    Yep, definitely. I also sincerely hope that the search engines will b e able to stifle such bad practices, so it doesn't pay off in any way.

  • [...] Questions I’d Love to Have Answered About SEO. Informative and Interesting, Robert Nyman wrote this article about his viewpoints of what the mechanics of SEO means to him. He talks about the difference of making a quick buck with quick fixes and hacks for personal gain, as opposed to good business practice and how the later will pay off in the long run. There are a lot of related links for informative reading at the end of this article. [...]

  • David M says:

    I like the new 'onlick' event you describe in the paragraph 'Freaky onclick implementations'.

  • Norm says:

    I recently lost ground in search results on my site by packing extra keywords into my titles tags I wanted better results on. I've since adjusted the tags to be more descriptive of the page and will be looking at the results… and also feel better about doing the right thing. I agree that making the site the best it can be for users is the way to go for the long haul. Thanks for the post!

  • Robert Nyman says:

    David M,

    Thanks. :-)

    Norm,

    Interesting to hear. And I agree, the sensible thing is for the right thing to pay off.

  • [...] Nyman has questions he’d love to have answered about SEO. I’m not the person to answer these questions, certainly, but I can certainly provide [...]

  • Portland Seo says:

    Great article. It is really informative and helpful. It did answer some of the questions I have about SEO.

  • Leave a Reply

    Your email address will not be published. Required fields are marked *

    You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>