Mobile Optimization Terminology and FAQ

A header refers to the source elements in the HTML code of a website that send and receive information. This includes the browser being used to view the page, the page that is shown and the server where data is sent.
A mobile user agent is the field in the header that identifies the client software and browser originating the request (e.g. Android, Blackberry, etc.) The ability to properly detect mobile user agents gives marketers control over the content that is displayed and how it appears and functions on the particular device.
A redirect transcript is our terminology referring to just the portion of a web server’s header response code used to detect and redirect various user-agents (like Android or iPhone browsers) to mobile web pages. For each major mobile device, the redirect transcripts show what type of redirect a server used (e.g., 301, 302, 303, etc.), the URL or address where the user is redirected to, and the Vary tag value (see below) which tells Google whether the site is being transparent about its use of user-agent redirects (traditionally a very risky SEO practice).
A bookmarklet is a powerful Javascript tool that instructs your browser to do something such as save a web page or take an action on a particular site. When you save the Mobile Redirect Viewer bookmarklet to your browser’s toolbar, it acts as a bookmark that immediately launches the tool pre-populated with results for the web page you are viewing. Basically it removes the hassle of copying a long URL, going to our website, finding the Mobile Redirect Viewer, pasting in your long URL, etc. Get a bookmarklet for our Mobile Redirect Viewer tool.
Websites are organized in terms of a hierarchy. A deep-link is any request for a page other than a website’s homepage. Many retailers redirect mobile devices that request deep desktop URLS (say an ordinary website product page) back to the retailer’s mobile home page. That is a bad user experience and Google views that kind of redirect as an “irrelevant redirect.” The trouble is, it’s hard for marketers to confirm those configurations because its a server setting.
Sometimes referred to as a status code, response codes indicate how a specific HTTP request was completed. Possible responses include a 200 code (indicating a page was successfully served), redirect codes (which include 301, 302, 303 and others), error codes (like a 404 error) and other informational responses. Most retailers that employ mobile redirects use a 301 or 302 status code. When they do, the redirect status code is usually accompanied by a location field that indicates the address where that user gets redirected. For example, when desktop users request Amazon’s home page, the server responds with a 200 status code indicating content was served. But when iPhones and Androids request Amazon’s home page, their server responds with a 302 redirect that routes those devices to a different location ( How do we know that? Our Mobile Redirect Viewer shows that in the transcript for each device, which you can see in this example.
A user agent string is the string of code that identifies the browser being used by the customer to request the URL. When any browser requests a web page, standard internet protocol requires the browser to communicate its user agent string to the web server in exchange for receipt of the web page content. For example, when an iPhone 4 requests a web page, its Safari browser will automatically tell the server its “name” (so to speak) which in this case is: Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_0 like Mac OS X; en-us) AppleWebKit/532.9 (KHTML, like Gecko) Version/4.0.5 Mobile/8A293 Safari/531.22.7 Most retailers with mobile sites (like Amazon, Walmart, Staples, many others) configure their servers to detect mobile browser user-agents so that when the server sees user-agent strings associated with popular devices (see the “iPhone” in the above string?) the server then knows to redirect those users to the “m.” mobile page or deep page. Sometimes IT unknowingly implements such redirects improperly, or in a sub-optimal way, which hurts marketing results. Without the Mobile Redirect Viewer, marketers have no way of validating their own site configuration against best practice.
A redirect Vary tag is a standard field included in a web server’s header response that marketers have not needed to worry about until now. Now, Google is recommending all sites that employ user-agent browser redirects populate this tag to be transparent in telling Googlebot that the content which lives at a URL varies by user agent. That means retailers with mobile sites that use user-agent redirects must program their server to include the words “user agent” in this field any time a user-agent redirect is triggered to route mobile devices from a desktop URL to a mobile URL page. (Notice the Amazon example follows best practice and includes “user agent” in the Vary field)
Both are valid ways to redirect mobile users who request a desktop page to an equivalent relevant mobile web page. The method most big retailers employ is server-redirects because these get executed faster and introduce less load latency for the end-customer. However, browser-redirects are becoming popular among retail brands with fewer IT resources because installation simply requires adding some Javascript code to the page to handle device detection and redirection. (The problem is this code gets executed by the customer’s browser instead of by the server which means extra load time on the customer’s mobile device.) In either case, the new redirected page is reflected in the browsers’ address bar so as not to confuse the user about the page they are viewing. Again, both are acceptable methods but if retailers use server-redirects, search engine optimization best practices require they properly configure the server redirect, as reflected in the redirect transcript. The Mobile Redirect Viewer lets the marketer easily see whether they have properly configured their redirects and pages to satisfy Google’s guidelines.
We welcome your feedback via