What We Learned From Google’s Singapore Webmaster Conference

Back in October, Google held its first ever Webmaster Conference in Singapore, a conference designed for digital marketers, website owners, SEO professionals and web developers alike. The Google Search team hosted the conference, along with the help of other influential online practitioners. The goal of the conference was to provide interested parties with knowledge they could use to enhance their websites and their performance in Google search. Though the conference was full of fun information, some was more notable than others. Below are five things we took away from the event.

Emoji Search Is Now a Thing

Google sees over one million searches a day that include an emoji. Yet, it only just now added the ability to crawl, index and rank emojis. However, that doesn’t mean that it hasn’t been trying — it took the search giant a year to add these capabilities. 

Tips for Crawling and Rendering

In addition to sharing info on upcoming trends, the conference also shared tips webmasters, developers, site owners, etc., could use. The tips pertaining to crawling and rendering were as follows:

• Google minimizes its fetches, which enable site owners and developers to test how Google crawls an URL on any given site
• Don’t place a lot of emphasis on caching rules, as Google doesn’t abide by them.
• Your robots.txt play a major role in whether or not Google crawls your website.
• If Google cannot reach your robots.txt because of a 5xx error, it will not crawl your site.
• Most of the time, Google uses the Chrome browser to render what you see, regardless of which browser you use.
• Google performs 50-60 resource fetches per page, which equates to a cache rate of about 60% to 70%. That is about 20 fetches per page.

Duplication Tips

The conference also touched on duplication or, more specifically, how the search engine handles it. It also provided tips on how webmasters can help Google figure out the official version so as to avoid penalization:

• Use meaningful HTTP result codes, which are the server’s response to a person’s request made to the server.
• Use redirects.
• Use hreflang, a technique webmasters can use to specify the language and optional geographic restrictions or any given page.
• Check your canonical links, which are the links that help to prevent problems caused by duplicate content that may appear multiple times on your site and throughout the web.
• Protect dependencies for secure pages.
• Report cases of content theft to Google.
• Avoid mixed signals with your canonicals. For example, do not canonicalize page X –> page Y and then do the same with page Y –> page X. This will only confuse Google and defeat the purpose of canonicalizing at all.

Google’s Beginning to Understand Synonyms

According to presentation by search lead, Paul Haahr, Google assess searches it got wrong, why they got it wrong and how they can prevent similar mistakes going forward. Many of the wrong results involved search queries that used synonyms. To demonstrate how Google is getting better at discerning between synonyms, he uses a few tweets as an example.

The first tweet involves Google’s ability to discern between the three commonly used meanings of “gm”: General motors, general manager and genetically modified. The tweet demonstrates why contact is so important (i.e., “gm truck,” “gm barley” and “baseball gm salary”). With context, Google was able to produce relevant results.

In a second example, Haahr explains how Google is now able to return relevant results for “York hotels” that do not include “New York hotels.” Apparently, Google built something into its IR system to exclude results with the word “new” if the search is for “York hotels” only.

Why You Should Care

Though the conference didn’t yield any new information, it did provide clarity and insight that SEOs can work into their daily lives and use in future strategies. Any info you didn’t know could be used to improve your strategies and therefore your search positions in the future.