from SEOByTheSea

Google uses a large number of signals to decide upon the order of pages shown in search results. Some of those signals measure the quality or importance of a web page, while others may indicate how relevant a page is for a particular search query entered into a search engine’s search box.

One fairly obvious relevancy signal is whether or not the words in a query actually appear upon a page that might be a search result for that query. If those words appear on the page more than once, the page might be considered even more relevant for that particular query than other web pages where the terms only appear once, or not at all.

Another factor that might indicate how relevant a page is for a particular set of terms is how close those terms might be on a page. While you could easily count the number of words between individual query terms to determine how close they are to each other, the formatting of web pages presents some challenges to the approach of simply counting words between terms, such as in a list like the following:

Imagine that the list in the image above is all that appears upon a particular web page. Since every item listed is about Saturn, as shown by the heading of the page, it could be said that semantically each list item is equally relevant to Saturn in terms of closeness, even though the items listed grow in visual distance from the heading of the list when calculated by the number of words between “Saturn” and list items.

This way of calculating semantic closeness means that the page this list appears upon is equally relevant for the terms “Saturn Mass”, “Saturn Volume”, and “Saturn Rotation.”

A Google patent granted this week explores how the search engine might view how close words are together when they appear in semantic structures like a list, to determine how relevant a page might be to queries that contain those words.

The patent was filed back in 2004, but it provides a way of thinking about how semantic structures on web pages might be interpreted by a search engine in a way that might not be obvious on its face.

Document ranking based on semantic distance between terms in a document
Invented by Georges R. Harik and Monika H. Henzinger
Assigned to Google
US Patent 7,716,216
Granted May 11, 2010
Filed: March 31, 2004

Abstract

Techniques are disclosed that locate implicitly defined semantic structures in a document, such as, for example, implicitly defined lists in an HTML document. The semantic structures can be used in the calculation of distance values between terms in the documents.

The distance values may be used, for example, in the generation of ranking scores that indicate a relevance level of the document to a search query.

HTML Formatting used to Determine Semantic Structures

One part of the process behind this approach involves a search engine analyzing the HTML structures on a page, looking for elements such as titles and headings on a page, unordered lists (<ul>) and ordered lists (<ol>), nested tables, divs, and line breaks (<br>) that might be used to layout a list of items on a page.

Page headings might use an actual heading element such as an <h1> or a larger sized font such as <font size=16>, and text below that heading might be considered to belong to the heading.

In other words, the search engine is attempting to locate and understand visual structures on a page that might be semantically meaningful, such as a list of items associated with a heading. We’re told that this process may also look for other kinds of meaningful semantic structures other than just lists as well.

The patent gives us the following rules about headings and list items, when it comes to the distance between words appearing within them:

1. If both terms appear in the same list item, the terms are considered close to one another;
2. If one term appears in a list item and the other term appears in header, this pair of terms may be considered to be approximately equally distant to another pair of terms that appear in header and in another of the list items;
3. Pairs of terms appearing in different list items may be considered to be farther apart than the pairs of terms falling under 1 and 2.

So, in the Saturn example above, the words “Saturn” (from the heading of the list) and “Distance” (from the last list item) are considered closer together than the words “Days” and “Rotation” even though “Days” is the last word of the first list item and “Rotation” is the first word of the second list item.

Conclusion

This Google patent was filed way back in 2004, but it does present some interesting ideas about how the search engine might look to semantic structures like lists to determine one aspect of how relevant a page might be for a particular query.

from searchengineland.com

If you go to Google.com and type “hamburgers,” “shoes,” “candy,” “grills,” “beer,” and hundreds of other terms of ambiguous local intent, Google will almost always show you local results on a map that’s tied to your IP address.

But type in “seo,” “seo company,” “web design company” and several terms related to these two types of businesses — heck, go ahead and specifically add in a city name like “seattle” or “houston” or wherever, and even include the state abbreviation like “wa” or “tx” — and Google willalmost never show you local results on a map.

Why does Google have it out for web designers and SEOs? That question has been going around for nearly two months, particularly in this thread in the Google Maps Help Forum. And, as Barry Schwartz reported on Search Engine Roundtable, Google employee “Joel H” (likely Joel Headley) posted an update today saying that it’s not a bug (as originally said to be), but a specific, conscious decision on Google’s part not to show these types of businesses on a map.

“Today, we’re intentionally showing less local results for web design / SEO queries. For example, [web design sacramento] doesn’t display local listings today. We believe this is an accurate representation of user intent. In some cases, we do show local listings, however (as NSNA/php-er noted) [web design in bellingham]. I’m sure some of you feel we should be displaying local results for queries like [Web Design Vancouver]. I understand that concern, but based on our understanding of our users, we feel this is the right decision for now.

“I’ll give the usual disclaimer that we’re constantly working on improving the user experience and results will vary over time. So, this could change in the future, but I wanted to be explicit about what we’re doing today.”

What Joel H. is saying is that certain industries won’t get local business listings even with a city name included in the query, but they might get a 7-pack if the query includes a word like “in.” So, for example, queries like “seo seattle wa” (even with the state abbreviation!) and “web design seattle” don’t show local listings with a map:

But if you search for “seo in seattle” or “web design in seattle,” you’ll get local business listings.

This doesn’t just affect SEO and web design companies; the results are similar for queries such as “ad agency seattle,” “advertising agency seattle,” “graphic design company seattle,” and “logo design company seattle.” These aren’t local searches to Google. (But “candy” is very local, even without a city name included. Riiiight.)

Joel H.’s statement only says, “We believe this is an accurate representation of user intent” and “based on our understanding of our users,” but offers no details beyond that. There doesn’t seem to be any logic behind why this group of companies isn’t considered to be local without the word “in” as part of the query. And vague statements about “user intent” don’t seem to fit the openness ideal, either.

Until these search results change, or until there’s a better explanation from Google, expect to see plenty of comments and posts about Google profiling SEOs in 2010.

A recent patent application from Yahoo explores ways that a search engine might consider the amount of time it takes different types of pages to render and other issues involving how quickly pages respond to a visits in ranking, classifying and crawling those pages.

Latency is a big fancy word that simply means the amount of time between when something was started and when you can see its effects. It’s a word that shows up very frequently in the Yahoo patent filing. It’s a word worth learning a little more about, especially when it comes to web sites, how people use them, and how a search engine might track that use.

A search engine may look at a wide range of information to make decisions about whether or not it will visit and index pages on the Web, how it might rank those pages in search results, and how it may classify those pages.

It’s likely that a search engine will consider at a wide range of informational signals. Those can include the content that appears on web pages, links and the text within links that point to and from pages, information about how people use specific web pages, and other information about pages and the sites that they appear upon.

A search engine might also look at how quickly pages load and render in a browser, how much people might tolerate when pages load slowly, and how good an experience web sites might deliver to their visitors.

When a search engine ranks pages in search results, it will explore signals that indicate how relevant those pages are to queries that might be used to find them, such as the use of words upon those pages that appear in those queries. A search engine may also look at signals that indicate the quality of the web pages that it might list within those search results.

A measure like PageRank is supposed to be an indication of quality rather than relevance, because it looks at the number and “importance” of links pointing to a page to try to determine how important a page might be. There are other quality signals that a search engine may use. Some examples might include things such as the amount of text upon a page, how readable that text is, if the page contains broken links, and possibly hundreds of other factors.

A search engine wants to return pages in search results that are both relevant and high quality.

Another set of signals or factors that a search engine may use involves how people interact with pages that they find on the web. These can include which pages people select in search results when they see them in search results for a specific query, how much time people might spend on a page they’ve selected before they return to the search engine, how far down a page they might scroll, whether they bookmark or save a page, and others.

User Experience Characteristics

The patent filing considers much more than just how quickly pages load into a browser, and it may influence more than just the rankings of pages.

It tells us about an information integration system that can be used with search engines, job portals, shopping search sites, travel search sites, RSS applications, and other types of pages, and how it might look at those in at least three different ways:

Access – How quickly it takes to access a page or other kind of document when sending a request to retrieve a page or document. Measuring access might mean looking at performance characteristics associated with a page such as server performance, and file performance. It might consider how quickly a page might load for visitors at different connection speeds, such as broadband and dialup. A search engine crawling program might simulate connections at different speeds to measure how quickly a page loads for visitors coming to a page through dialup or broadband connections.

Rendering – How quickly a page starts showing up within a browser (and it might emulate a number of different types of browsers), how a page loads in a browser, and how long it might take for the full page, or at least the part of the page above the fold to load in a browser. It contemplates that on some sites, some large pages might be set up so that even though they contain a lot of content, the content at the top of the page renders quickly so that a visitor doesn’t have to wait very long to start reading and viewing the content on the page.

It may also consider such things as “differences in complexity, size, number of files, user interface mechanisms, embedded sections (e.g., advertisements, audio content, video content, security features, etc), and/or the like,” to understand how a page renders, and how good of a user experience that might be.

User Experience – How do people actually use web sites, and how do they react to different access and rendering issues on different sites?

Different people might have different levels of patience in waiting for a site to load and render in a browser, and they might be willing to wait longer for some types of sites to load and render than others. For example, someone might be willing to wait longer for a page to show up that is associated with their bank account, than a for a “more generic” type of page.

Examples of other “user related performance characteristics” could include how visitors to pages react to things such as:

• Pages that fail to download or render within an acceptable period of time,
• Pages that automatically play video or audio content,
• Pages that include pop-up or pop-under advertisements,
• Pages that in some other way add further delays due to additional file downloading, additional processing, etc. These might include things such as Javascript, Flash, Embedded or externally links objects, and Plugins

How Measuring Latency and User Experience Might be Used

The inventors behind the patent application point to at least three uses that a search engine may have for measuring the performance of a web site based upon access, rendering, and user experience. They are ranking, classification, and crawling.

Ranking – The information collected about user experience characteristics could be used to possibly filter, promote, or demote web documents to improved desired user experiences.

Classification – The user experience information might be used to classify pages in some way. The layout of a page might indicate that a site might contain certain types of content related to certain types of sites. The patent application tells us:

For example, finance-related websites often display streaming data of the stock market, news websites also often stream content, and certain types of web pages might use frames or tables which may be useful in classifying the web document.

Crawling – When a search engine has a list of URLs to visit that it hasn’t seen before, or that it might revisit to check for new content, it might consider a number of different things in determining which to look at first. The user experience information might help making some decisions to look at certain content on pages that a search engine might not have considered before.

A search engine may simulate the amount of time it takes to connect to a page, the way and amount of time a page renders in a browser, and how people react to those times to influence how a page is ranked, classified, and how much of the page is crawled and indexed – including embedded material on a page such as javascript or flash content.

from MediaPost Search Insider

Flatter and more-responsive organizations. Working on SEO is like taking your Web site to the doctor: a good SEO consultant will tell you what you have to do, but the hard work is up to you. Companies that listen and respond will do better than companies that justify, finger-point and go on the defensive. Healthy companies look for ways to improve; dysfunctional companies offer reasons why improvement is impossible. Companies that refuse to do the heavy lifting required to whip their site into shape generally are equally negligent in other areas of their business.

Better communication channels. SEO is by nature a cross-functional exercise. It involves many different departments, all working together toward a common goal. This approach is well within the comfort zone of healthy organizations, but totally foreign to dysfunctional ones. An SEO initiative severely tests the communication and cooperative capabilities of an organization. It requires marketing, IT, product managers and often legal to all work together, and the faster they can do this, the more positive the results will be. SEO is not a one-shot tactic. In the most competitive categories, it’s a full-out and ongoing war. The companies that can respond and adapt quickly will win that war. The ones mired in bureaucracy and butt-covering will inevitably sink in the rankings.

Healthy community connections. The new era of digital communications requires companies to be engaged in an ongoing dialogue with their community of customers. Great companies do this instinctively. Bad companies put up huge corporate communication barricades, keeping the angry hordes at bay. Because much of this dialogue happens online, these dialogues tend to generate reams of content and links. Raving customers generate link love; angry customers generate link hate and reputation management problems. A company that can effectively engage in conversations with customers will find a natural lift in organic rankings is often the result.

Efficient execution habits. Companies that keep a clean house do better organically than companies that keep skeletons in the closet. Both approaches are symptomatic of the company’s overall approach to business. Highly effective companies constantly upgrade systems and infrastructure, both in their organizations and their online presence. They invest in best of breed tools and technology. And they are able to quickly prioritize and executive as the landscape shifts. Again, a clean technical online infrastructure makes SEO much, much easier.

Executives that “get it.” C-level executives who make SEO a priority realize that the marketing landscape is shifting quickly. They’ve been paying attention to customer behavioral trends and have committed to being proactive rather than reactive. This usually indicates well-placed intelligence gathering “antennae” and feedback loops. It also indicates an executive who isn’t hopelessly mired in “old-boy” thinking and outdated command and control management models.

Corporate pride. Content might not be the sole king anymore (SEO is more of an oligarchy now) but it’s still part of the ruling class. Great cultures tend to engender pride that naturally precipitates an explosion of content. People blog about where they work, people tweet and product managers enthuse verbosely about what they’re working on. All of this generates great, searchable content online.

Companies get the SEO rankings they deserve. I’m guessing that if you asked any SEO consultant in the world, they’ll tell you their favorite clients are the ones that are the easiest to work with: clients who listen, are proactive and for whom continual improvement is a religion. Based on what I’ve seen in the past decade, this attitude extends beyond the SEO team (indeed, it has to) and permeates the entire culture. There are those who game the system and gain undeserved rankings, but more and more, “organic” rankings are just that: rankings that come from the very nature of the company and how they conduct themselves in the marketplace.