Reading this excellent article over at UXMatters, I was pleased so much to read this
Putting users’ goals first allows a design team to concentrate on the new opportunities a mobile application presents rather than seeing the challenges of mobile simply as barriers to implementing a Web application’s existing functionality.
So much so that I thought it worthy to put here… After many months of zero activity.
So nice to see someone efficiently articulate what I’ve tried to communicate on a few mobile projects now - please, put your desktop context away.
It takes a true optimist to see opportunities, not limitations - brilliant!
I’ll be the first to admit that this is more optimistic than I’ve ever been able to be, but now I have something to aspire to.
Emily Freeman recently invited me to pipe up on one of my recent areas of interest - Mobile SEO.
Catch the full article over at mobilista.com.au.
Mobile Search is experimental. The big search engines constantly change their search offering in response to market segmentation; global changes in the market; technological capability and a raft of other variants. Generally this is done by leveraging more than one collection of data—news, images, wikipedia, weather etc—and presenting the collected results to users in a helpful way. This is known as Universal Search or Federated Search and it’s not unique to mobile.
This case study looks at just how experimental—read immature—mobile search is and what this means for mobile SEO in a quickly changing market.
Let’s get straight to some examples.
A few interesting points:
Both device types share the same default collection – web. Come again, it defaults to web collection… on a mobile device… Interesting.
WAP configuration
Web | Local | Images | News | Mobile Web
iPhone configuration
iPhone defaults to web as above. Okay, but where is the option to select mobile web results? After all, this is a mobile device. Actually, forget mobile web results, where can I do a local search? With it’s Safari WebKit browser, Google presumably treats it like a full featured PC, offering results that look a lot more like those you’d find on standard Google search:
Web | News | Video
This is too brazen - those mobile web results must be in there somewhere. Say hello to Google’s Blending Mobile Search Results patent. Let’s not get into devilish detail here, enough to say that this patent describes a method by which mobile web results are blended into web results - removing the need to keep them separate.
Here’s an example
Note that only one listing—the first—from web results has a mobile icon next to it, indicating that it’s a mobile web listing. This is a blended results page. In the mobile web results, as you’d expect, there are only mobile web listings. You might also notice that many of the listings for mobile web look suspiciously like online sites. In fact, in this list of 10 results, only one is a bona fide mobile site. We’ll talk about this some more later.
So, why use Blended Search for some devices (iPhone) and standard Universal Search for others (WAP)? The answer lies in that dirty C word, Convergence. We know that iPhone handles standard web content with considerable grace—at least in context to its peers—and, it also brings a new school of mobile web users – those that may not even be aware that mobile web exists, but rather think in terms of accessing the web while being mobile. With this in mind, Blended Search makes sense for this type of device. Compare this to a crummy old WAP device with a Symbian/Opera browser, and you can easily see why Blended Search doesn’t make sense here. In fact, on these WAP devices, you’d expect to see web results buried under more usable tiers of content like news and weather. Yahoo does exactly that.
WAP Configuration
News | Mobile web | Images & Photos | Web | Wiki | Answers
iPhone Configuration
News | Web | Answers | Photos & Images | Wiki
Note the absence of mobile web for the iPhone - this suggests that Yahoo also now have Blended Search capability targeted at specific devices. Now, for comparison’s sake, let’s take a look at the difference between web and mobile web results for Yahoo.
You see here, Yahoo take a slightly more traditional approach - notice that mobile web results are all legitimate mobile entries and not blended with web results. They maintain discrete collections for both web and mobile. However, it’s pretty clear from the iPhone example that they also are moving toward Blended Results—in a similar way to Google—for devices that are deemed capable. Only time will tell if they retain the integrity of their mobile results, or go blended across the board as Google and others have done.
I’m surprised by this—not so much the change itself, but that I missed the change—because not even 12 months ago, mobile web search results were totally dominated by real mobile sites, you know the sort, remember .mobi TLDs? Putting anything else in here was risky at best, for fear of device unpredictability when trying to render sites not designed for small screen limitations and needless to say, bandwidth considerations.
SEO has long been a dark art and mobile SEO more so. There’s precious little information in the public domain about mobile SEO, and I’m yet to met anyone (SEO specialists included) who really understand this market, let alone how to optimise for it. I’m not an SEO specialist, but for the last few months, as a Mobile Web Producer I’ve grappled with a SEO implementation for a large enterprise mobile product. As with many SEO projects before this one, it was largely trial and error. There was some success (albeit limited) and the output from ‘error’ is learning.
So, here’s a few things learnt along the way
In the old model of Universal Search—where only bona fide mobile sites are included in a mobile web collection—these conventions are more important. It’s true that the big search engines often use special crawlers that act like mobile devices—Google’s for example behaves like a Nokia 6820—to collect mobile content. This might lead you to believe that information such as the existence of a Mobile DOCTYPE is collected from the HTTP header and is used to determine how ‘mobile’ the site is. This was found to not hold true. Of course, it’s still a good idea to have the correct DOCTYPE and to use conventions, but it may not deliver on business outcomes.
The Blending Mobile Search Results patent discussed earlier suggests that the search engine ranking for mobile is actually determined by keyword association in the content. This explains the perceived poor relevancy of the mobile web results compared earlier - results listings are not there because the search engine knows they belong to a site designed for mobile browsers, but because the content of the listing includes keywords that are associated with being mobile, or the site has mobile specific content, like ringtones and games.
SEO can do wonders for the web if implemented properly and it’s natural to see success in one area and expect the same in another. However, as we’ve seen in this analysis, mobile search appears to be trending toward convergence which is currently manifested as Blended Results for high capability devices. This means that any given mobile site is not only competing with other mobile sites for ranking, but with the internet at large. Little is known about how sites are identified and ranked, so we don’t know much but we do know there’s a long way to go.
By the time we have good search results for (true) mobile web—supported by a body of knowledge on how to optimise for mobile—the majority mobile web usage is likely to be on devices that can gracefully browse standard web. In fact, that point in time has perhaps already arrived, Google and Yahoo both seem to think so.
Information and opinions in this article are largely based on my own experience and understanding. Feel free to point out any gaps in this analysis, or just leave an opinion of your own in the comments section.
It’s not a thoroughbred case study, but:
…users are able to browse useful, noteworthy, and most importantly, user-rated applications. The fact is there’s no equivalent for well-designed mobile web sites, no matter how Safari-optimized they are.
Good point. Good article.
via Idlemode
The crew at Mozilla Labs are working on Geode for Firefox 3. It attempts to localise search for desktop browsers in a similar way to mobile web. Instead of using GPS or mobile-tower triangulation common to mobile devices, Geode brings location based services to the desktop using Skyhook’s positioning system based on wireless hotspot meta-data .
The implications are obvious for both consumers and advertisers, but it’s yet to be seen what might come of this capability. Like other major LBS enablement providers, Mozilla is attempting to neutralise widespread privacy concerns by promising their trustworthiness in lengthy privacy policies and giving users control over how much information about their location is shared.
A List Apart published a great article covering Test Driven Progressive Enhancement.
Looks like a good practice, but seems… difficult, for mobile.
Over here, we use a few different techniques to (try) overcome broad device support issues.
We know that: generally, mobile devices don’t support Flash; and Javascript support is patchy at best (although this is changing fast).
Box modeling isn’t as much a problem, but there’s certainly inconsistencies with other browser standards, like link/active/hover state for href which is idiosyncratic, even for browser whose behaviour should be predictable.
So what do we do?
Old-school targeted CSS policies based on user agent detection. Sound labourious? It is.
Volantis helps, but brings it’s own problems to the party.
Then there’s mobile specific functional capability. I’m talking about things like vCard support and telephony (WTAI, tel). For those, we rely on design solutions that cater for heavy degradation while still offering a good experience on capable devices. This relies first on the designer’s intimacy with each device’s capability and idiosyncracies. Secondly, we rely on some customised control systems (based on databases that hold ‘flags’ against user agent types from field testing) embedded in the mobile application infrastructure. It’s tedious work, and requires high maintenance.
Emily Freeman, specialist mobile advertising consultant and founder of Mobilist talks with Vishal Sharma. The interview is a good account of what’s currently happening in the Australian market.
Rummble crossed my path a few months ago. It’s well executed, and I like it because:
There aren’t too many sites out there that really have a handle on bridging the gap between desktop and mobile. Rummble is a pretty nice example of what’s possible.
desktop – http://www.rummble.com
mobile – http://m.rummble.com
This month, Google spruiks for Rummble, one of the first to use Gears Geolocation API.