The Maine SEO Blog

This is a recap from a webinar SEOmoz did a while back, with Rand Fishkin. Follow Rand on Twitter here.

Goals of successful information Architecture

• Semantically logical structure (Zoo animals –> African Savannah –> Lions) – If your site architecture is logical, your users will spend more time on site, find what they want easier, and convert more often.
• Minimize click depth (not JUST for search engines!) – so that users and search engines can reach any point on the site in a minimum number of site. Usability and SEO best practices are nearly identical.
• Maximize usability of navigation

Tips for Semantically Useful Navigation

• Initially design without keyword research – so that you aren’t bias in the way you organize by the keywords you discover. Rand suggests first organizing your content the way it makes sense to you and then incorporate the keywords that make sense for search engines, where they make sense.
• Add  in keyword research based modifications to your draft IA
• Validate architecture/path with non-SEOs – make sure that your navigation still makes sense to non-SEOs and non-web users

Tips for Minimal Click-Depth

• Imitate the ideal navigation pyramid – in the first example, you’ll see you can get to 1 million pages with three clicks; in the second, you can only reach 150,000 pages with three clicks.

• Broad linking at top levels – at the top level, link to very broad categories; link to popular subcategories from the homepage. Rand uses Metacritic as an example.
• Editorial categorization > user-defined (hack: multi-level HTML sitemap – like this page at Rotten Tomatoes)

Tips for Usable Navigation

• Obvious navigation elements (like with MailChimp)
• Naming Conventions that Make Intent (not like Media Temple) – don’t use language no one outside your company won’t understand
• User & usability testing (using something like Silverback 2.0)

Avoiding Common “Big Site” Problems

• Duplicate content issues:
  • Rel Canonical tags (although sometimes it isn’t perfect) – you’ll lose a tiny bit of PR, but you’ll save yourself before bad things happen. Rand always suggests using the rel canonical for the absolute URL of pages for your article/blog/products section(s).
• Google Webmaster Tools – use to ignore duplicate content
• SEOmoz web app

Scraping and Re-Publishing

• Scrappers (good or bad) that take content can be shown instead of original content.
• Employ absolute URLs (as in <a href=”http://www.seomoz.org/blog”> anchor </a>) not relative (<a href=”…blog”> anchor </a>)
• Don’t go overboard with bot blocking

Incomplete Indexation

• Don’t look at the site: command and compare it day to day. (Read this post by Rand.) Use track referrals instead.
• Check page “types” that don’t receive traffic (see this post by Rand)
• XML sitemaps – helps search engines crawl large websites
• Content syndication (use the allintitle: command)
• RSS feeds
• Twitter for indexation

“Search Results” in the SERPs

• Create category “landing” pages
• Remove obvious traces of “search” on landing pages

Faceted navigation

• Rel canonical can help
• Use AJAX to reload pages
• Watch out for Google crawling Javascript
• Offer facets only to loggin-in/cookie users

Q&A

Sorry couldn’t stick around for the whole webinar, but here are two juicy tips:

Google Image search – less a new algorithm than a new interface. Text around image seems to be doing better than alt text.

Want a copy of Rand’s Firefox bookmarks? Here they are!

Four Methods to Accelerate Your SEO

1. Increase rankings on individual results
2. Earn rankings in alternative results types
3. Improve visibility in the long tail
4. Better convert existing search traffic

1. Strategies to Boost Individual Results

Competitive Analysis Process

Step 1: Determine weak vs. strong metrics

• Determine value vs. difficulty (Low difficulty + good search volume + high conversion rate = WIN) (How difficult is your keyword?)
• Paid search is a great way to test the value of a keyword

Step 2: Establish Analysis Process

• More root domains linking to the page
• Possibly more subtelty in anchor text
• Possibly higher value/relevance in links

Linkbait Ideas:

• Exact match domains (be careful here, and only use what makes sense)
• Widgets and badges (Example: picnik.com)
• Embeddable infographics (Example: this infographic)
• Content & technology licensing (Example: allrecipes.com)
• Link attraction targets with regular updates (Example: SEOmoz’s SE Ranking Factors – SEOmoz updates the same page with their biannual article, adding to the link juice collected from previous articles)

2. Earn Rankings in Alternative Results

Video Results & YouTube

• Huge opportunities for video by creating a keyword rich title, keyword rich URL, and good screenshot
• Title your blogpost containing your video something different than your YouTube video title – that way you’re not competing with your own video
• YouTube is the 2nd most popular search engine = huge opportunity
• Possible YouTube Ranking Factors
  • Title/description
  • Video length
  • Number of comments
  • Propencity for people to view video all the way through
  • History of account uploading video

Local & Maps Results Tips

• David Mihm just updated his Local Ranking Factors article
• Make sure your Google Maps result name matches your name, phone number, and address exactly
• Google will remove suspicious looking reviews, so gather them organically

Image Results Tips

• Not necessarily associated with traditional ranking factors
• For certain industries, image search can be among the highest converting traffic (but watch for first click attribution)
• Possible Google Image Search Ranking Factors:
  • Size – image has to be at least a certain size and below a certain size
  • Relevancy – Google’s doing a great job matching relevancy for images

News Results Tips

• Start by getting site submitted to Google News results, you can do that here
• Rand thinks Google incorporates Real Time Search into News Results
• Including an associated thumbnail image will increase click through rates

Real Time Results Tips

• Real time results don’t get a great click through rate
• HOWEVER, posts and links that get Tweeted a lot will do better for real time results and where real time results are used

Blog Results Tips

• If you have a blog and it has an RSS feed, you can appear in Google Blog Results
• Continue creating keyword rich titles to do well hear

Shopping Results Tips

• It appears a certain quantity – or a certain level of competition for products – must happen before shopping results are displayed
• Read this post to learn more about ranking well for Shopping results: How to Rank Well in Google Products Search & a Big List of Places to Get Reviews

3. Improve visibility in the long tail

25% of queries searched on Google each month are entirely unique and have never previously been searched for in the engine’s 10+year history. -Udi Manber, Head of Search, Google

• How the long tail works
• The long tail is much less competitive
• It’s difficult to compete here, because you can’t (really) do keyword research for it
• You’ve got to create high quantities of unique content (forums and Q&A sites are great ways to combat this)
• Is Blogging a Long Tail Strategy? Not necessarily. (Examples of great long tail strategies: NY Times, Wikipedia, IMDB)

Long Tail Strategies:

• Paid content writers
• Tweets and social updates as content
• User-generated Content (UGC), Examples: Webmaster World, Yahoo! Answers, HowStuffWorks, Answers.com, Reddit, Ezine articles, Etsy, SlickDeals

4. Better Convert Existing Search Traffic

Conversion Rate Optimization Funnel

• Visits to your website (100,000)
• Visits to landing page (25,000)
• Add to cart (500)
• Complete checkout (250)
• Customer lifetime value ($850)

Rand’s review of “Influence: Science & Persuasion”: An Illustrated Guide to the Science of Influence & Persuasion

Conversion rate strategies:

• Anchoring with pricing (Ordering highest price to lowest price, left to right)
• Pre-selected options (most popular, pre-selected, etc)
• Social proof (positive review with picture of person)
• Scarcity (Example: airlines proclaim “2 seats remaining!”)
• Reciprocity (Share with a friend and get more, example: Dropbox)

Rand Fishkin, SEOmoz, @randfish

Reverse Engineering the Search Rankings: Determining How the SERPs are Ordered and What You Need to To Do to Rise to the Top

Rand will cover:

• What is reverse engineering
• Search ranking factors
• Applying the process
• Examples in action

What is reverse engineering and why do it?

It is: a process to determine how a system is constructed. In this case, Google’s ranking algorithm for a particular query & result list.

Why do it:

Search Ranking Factors

There are over 200 ranking factors, some more important than others. This is the biennial survey SEOmoz conducts on ranking factors:

Trust & Authority of the Host Domain

Homepage Toolbar PageRank? – Rand cautions against this. Homepage PR is not about the domain, it’s on a URL basis.

(By the way, this URL will give you nice, clean search result screenshots.)

Domain mozRank – will give you the actual rank of the domain

Domain mozTrust – similar tool,

# of links to a domain – search on Yahoo for “linkdomain:yoursite.com”

Domain Authority – another SEOmoz algorithm (including Google’s ranking algorithm, page authority, domain authority, and many more). Puts all of these important stats together.

Links to a specific page

Toolbar PageRank – will give you a good idea of how many incoming links to a specific page

mozRank – if mozRank is high and PR is low, most of the time that’s because Linkscape updates far more often OR Google is not counting all the links to the page (might be a good indicator that that site is manipulating incoming links)

mozTrust – good anti-spam metric

# of links – relatively good metric, but you’ll want to break up internal/external links and nofollow/follow. Yahoo Site Explorer gives you raw number, but SEOmoz’s Open Site Explorer.

# of linking root domains – are there just a few with a ton of links from individual pages? Not good. Or a ton of links from a ton of different domains? Good.

Anchor Text

Exact Match Counts – you can see this under “Anchor Text Distribution” in Open Site Explorer

On-page/On-site Optimization

Important places for keywords:

• Exact Match Root Domain Name
• Keyword Matching in URL
• Title – in the front and together

Other important on-page element:

• H1
• Alt tag
• Body text

Don’t worry so much about keyword density. Instead, look at link metrics, etc.

Data isn’t perfect; Sometimes…we speculate.

Know what actions to pursue (and/or) What you need to Compete

Do you need:

• More anchor text rich links
• more domain diversity in your links
• More mozRank/PageRank
• More trusted link sources
• Better on-page optimization
• Freshness/Social & Sharing
• Something outside metrics (branding optimization)

Reverse engineering isn’t about trying to figure out why a site ranks where it does – it’s about figuring out what you need to do to do better (using your reverse engineering data).

Rand Fishkin, SEOmoz, @randfish