SEOMoz gives some terrible advice about SEO and user focused development. Rand presents some graphs about different optimization components and how their effect on ranking has evolved over time and magically draws the conclusion that SEO should focus on engines, not users because additional SEO tactics are required to rank.
I am not going to comment here on how he ranks the impact of each tactic. I disagree with some of his analysis, particularly the importance of keyword research and keyword mapping but I agree with others. The larger point here is that Rand is arguing against “focusing on users”, saying this approach is bad SEO advice because it does not include other SEO tactics.
Rand is ignoring the considerable nuance in the meaning of “designing for users”. The art and science of designing websites for users has evolved significantly in the thirteen year thanks to the contributions of Jakob Nielsen, Gord Hotchkiss and tens of thousands of designers, conversion optimizers and site owners who have observed user behavior, tested and validated different theories to enrich the body of knowledge about user interactions with the web in general and search users in particular.
The overall consensus from this research is that the tactics embraced by push oriented designers and agencies who claim to be designing for user are not effective. Flash based sites, fancy formatting and dominant images are almost as much of an obstacle for users as they are for engines. What works for users and for search engines is delivering clear information scent that matches the search intent, bulleted content and the judicious use of bold text. Scent should be presented in easy to read text with high contrast and the content should be written with the assumption that the user wants to scan the page in 3-5 seconds to make sure they are in the correct place. If the scent is not delivered in that time frame, the user will hit the back button and “bounce” from the site. Users will rarely engage with the site navigation if the landing page doesn’t match the search intent.
During the last 12 years, SEO has evolved from the process of manipulating search engines by jamming pages full of hidden keywords to a high ROI discipline with a data driven methodology rooted in query frequency that focuses on developing a site which matches the search intent of users with content that meets that intent. I call this evolved SEO Website Optimization. The effective optimizer focuses on creating pages that attracts visitors and reduces bounce rate, engages users and creating conversions.
A majority of the changes over the last five years with regards to how the engines regard on-page and on-site factors are rooted in the engines trying to emulate human users as they crawl and interpret the page.
- Content in the golden triangle is more important than content in other locations on the page. The impact of keyword stuffed H1 tags has diminished over time but designing your site to deliver good information scent in your headline is more important than ever for both SEO and for conversion.
- Engines readily admit that the location of a link on the page matters. Links in the variable content area on the page, particularly links in the golden triangle, count more than links in the boilerplate, the left/right rail or the footer. The savvy optimizer includes the most important links as citation links above the fold in the variable content area of the site, where users are likely to see them
- Toolbar data, analytics data, conversion optimizer data and other streams give the engines an incredible assortment of information about how the user interacts with a site and we are seeing this data affect rankings more and more.
While the engines are tight lipped about how they use this data, Matt Cutts revealed at SES San Jose this year that Sitelinks are driven, at least in part, by the popularity of individual pages on the site. During a site clinic review of Meijer.com he observed that their Store Locator is buried in their primary navigation (and suggested they make it more prominent) but that is was a popular page “because it appears in your sitelink”. The Meijer Store locator page is also the second listing below the sitelink despite being buried in the global navigation and completely lacking content on the page. User behavior data explains how a boilerplate ONLY page with a few graphics is the second highest ranked page on the site.
I will not take issue with Rands observation that SEO involves a lot of tactics that are specifically for the engines nor do I embrace the Google party line that says “make content for users, not for engines”. SEO is a multi-faceted discipline and I completely concur that these factors cannot be ignored. On the other hand, any SEO who does not focus on designing sites for users is doing a huge disservice to themselves and their clients.
Conversion takes place at the intersection of search intent, content and user experience. More and more, on-site SEO occurs at the same intersection.
Updated: Rand posted an update on his site as well as a comment here reflecting that he was misunderstood. He was trying to say that it is a mistake to ignore SEO for the engines and ONLY focus on users. It is impossible to argue with that. Given that, I have toned down my remarks some and I hope we can agree that user centric design is essential for SEO…it just isn’t enough.