Edward Lewis (@pageoneresults): Below is my frugal attempt at assigning SEO point values to various HTML elements and attributes within a semantically structured document. This is a general guide for assigning points based on years of working with HTML and gut instinct SEO.
Points are cumulative depending on the scope of the document(s). There is no perfect score although in theory, the higher the point values, better document performance would be expected. There are many variables at play when it comes to assigning point values to HTML elements from an SEO perspective.
7 Comments
Points are cumulative depending on the scope of the document(s). There is no perfect score although in theory, the higher the point values, better document performance would be expected. There are many variables at play when it comes to assigning point values to HTML elements from an SEO perspective.
7 Comments
Comments
Love the list and love the conclusion even more. Well done.
Interesting attempt to quantify the various elemtns. I wonder if, by assigning a somewhat significant value of 1.00 to blockquote, Edward implies that text within that tag might be given higher value by the search enignes (ie. should try to fit keywords into blockquotes when appropriate)?
If you are using the blockquote element according to the guidelines, you won’t be forcing anything that isn’t already there from the quoted source. I would like to clear up one misconception about this element and that is its use to indent content which is the result of WYSIWYG misinterpretations. The blockquote element is to be used when quoting more than one paragraph of content. Then you have the attributes that are used with blockquote such as cite. I will typically provide a cite attribute for all blockquotes and q’s. I’ll style those elements via CSS and will usually use styled quotes for before and after. Don’t ask me how the search engines treat the cite attribute, I really couldn’t provide any concrete evidence that there is any SEO value. When I read the guidelines for UA development, that cite attribute is referenced multiple times. 1+1=2 Even if the point value was at .00, that would not deter me from using it. I decided to learn HTML via the W3 guidelines. They are the only official resource for all things HTML and CSS. Along with a multitude of other technologies. The W3 is big when it comes to establishing guidelines for developers of UAs, Browsers, Crawlers, etc. There are so many different names used and I just look at them all as bots. They are machines. They don’t read or interact with your page. They read what resides between the HTML Elements and Attributes you’ve chosen to use. Do they interpret all of that? I think so. We’ve come a long way since those HTML 4.01 Guidelines were established in 1999-12-24. Keep in mind that the core of those guidelines have been in existence since 1993 June. Many of the elements from HTML 1.0 have transitioned to what we have today. Here’s a trip down Memory Lane for some. Not me, I’m not that old! Hypertext Markup Language (HTML)A Representation of Textual Information and MetaInformation for Retrieval and Interchangehttp://www.W3.org/MarkUp/draft-ietf-iiir-html-01.txt
<div><div><div></div><div></div><div>careful @pageonresults...</div><div></div><div></div><div>Jill has already made it clear on her forum that </div><div></div><div>"I’m afraid you just can’t "score" different design elements like that for SEO purposes. It simply doesn’t work that way."</div><div></div><div></div><div>heh heh.. without assigning values to the elements there is no way to quantify them.</div><div>With an experiment you must assign some type of measurements in order to run calculations.. </div><div></div><div></div><div>Even in animal experiments, the monkey learns through rewards to assign values to particular stimuli in particular contexts. </div><div></div><div></div><div>green light comes on.. push button.. get bananna</div><div>red light comes on.. push button... BZZZZZTTTT!</div><div></div><div></div><div>anyways... lol</div><div></div><div></div><div>So are you "scoring a page" to quanitify your results?</div><div></div><div></div><div>as in.. run the numbers on a competititor’s website listed above one of your pages, then compare to yours? or are you just using this as a rough guide to which elements you should place more importance on? </div><div></div><div></div><div>food for thought.. </div><div></div><div></div><div>What if... (completely hypothesizing)</div><div></div><div></div><div>we assign point values to the factors.... as u have done.. </div><div>then we assign "Weighting" factors... as google does.. </div><div></div><div></div><div>like an H1 will beat a bold anytime...</div><div></div><div></div><div>so if you have 15 of the same keywords bolded but the competitor has the same KW in an H1 preceeded by an image named the KW and the ALT for said image is the KW.</div><div></div><div></div><div>(yes in the above example 15 occurances of the same KW bolded could be considered spammy)</div><div></div><div></div><div>so even tho the point score would be HIGHER for the total of the bold words.. the H1, image name and image ALT would be "weighted" higer due to page segmentation bonuses, image ref from google images DB,. etc... </div><div></div><div></div><div>just an idea.. </div><div></div><div></div><div></div><div></div></div></div>
Sorry, I got banned over there long ago. Haven’t visited since. paisley, [B] has no value, it is a presentation element and not a phrase element. If we are talking [STRONG] then yes, the semantics are different. Also, [H1] is at the top of the food chain naturally. And, you ready for this? You can have multiple [H1]s per page. Now watch what type of responses that one generates! :) So are you "scoring a page" to quanitify your results? Not really. I’m not into the "deep analytical" approach. I just follow the damn guidelines and it works, bottom line. The guidelines state that if you have this, use this. If you need to do this, you need to use this. Its pretty straight forward once you understand the structure of the W3 and related properties. If I told you that I don’t use any of the SEO Tools on the market, what would you say? I built my own a few years ago and we’ve been actively refining it as we continue with our research and development. I use it almost every day and it allows me to get a quick glimpse of page quality at which time I can then move to advanced features and look at semantic structure. You can do the same but at a limited level via the W3. Do this, go to our http://www.URIValet.com/ and enter your destination URI. Take it from there. Be aware, that page is loaded with information about the destination so please do scroll and interact with the functionality of the report. At the bottom, you’ll find a window that will allow you to perform various validation routines. There is a Semantic Extractor there too from the W3. If you get an error running the extraction tool, there is a good chance it could not extract the meaning of your page which is usually the first warning that something "may" be amiss. That is the beginning of the process for me and covers most page basics. A little more will be happening behind the scenes which will not be available to those outside our team. You know, trade secrets and stuff. But, what we have there for you to consume publicly is a good starting point for flushing out various challenges with page quality. Enjoy!
Here’s a little tip for the folks at Third Door Media. A quick glance at a URIValet Report for this Story. Links 75, 76, 77 and 78 need updating. And yes, the numbers coincide with where we found them in the order of source. 75 1 301 | http://www.thirddoormedia.com/200 | Redirect http://thirddoormedia.com/76 1 301 | http://www.thirddoormedia.com/privacy.shtml200 | Redirect http://thirddoormedia.com/privacy.shtml77 1 301 | http://www.thirddoormedia.com/terms.shtml200 | Redirect http://thirddoormedia.com/terms.shtml78 1 301 | http://sphinn.jp/200 | Redirect http://www.sphinn.jp/Could I have you update those links so they don’t go through the 301 redirect? They should be non www and are hard coded www. Hence the 301>200. Not the most optimal utilization of internal link equity. Maybe even nofollow the privacy and terms while at it. ;)
nice programming on the URI thingy.i don’t use tools either, except a java script for metricsand the ASP it feeds to, so you are not alone.happy easter!