Just search in Katakana

UPDATE 4JUL2010: The ICANN board has now approved Chinese character URLs. I doubt this will change the marketing strategies for Japanese companies, however.  I think we’ll just get a bunch of redirects.

Americans invented the Internet, and it’s going to stay nice and readable to Americans as long as they have anything to say about it (daggumit). Most people in the Americas, Western Europe, and Australia are blithely unaware that a battle has been raging for several years among the internationalists over the format of the URL (the web address you type to go somewhere on the web). Why? Because these countries all use the roman alphabet. Our slavic friends would love to use cyrillic URLs; Chinese want Chinese characters, Koreans want hangul, etc. The ICANN (the international standards cops) are still only testing and allowing non-roman alphabets in limited trials.

Last month in Tokyo, I noticed something about all the adverts on the trains: the Japanese companies have found a way around the common restriction: they simply tell you (the customer) what to search for, using Japanese kanji and gana, nevermind what the actual proper URL may be. I’ve highlighted two examples here (forgive the low photo quality– lighting on subways isn’t so bright).

On the good side, by asking the commuter to search for a specific keyword, they’ve immediately screened out all the content from the English-language Internet (which can be overbearing in volume). Every mobile phone has this rudimentary web search capability, so the call-to-action is immediate.

Solar backpacks-- just search for

On the bad side, however, this strategy assumes some risk: if I ask for my customers to find me by searching for “スマート外国人” (smart gaijin), what happens when someone else comes along with a better SEO value for that term? All my advertising now goes to benefit someone else.

Mitigating this risk are three factors: 1) campaigns on the trains cycle pretty quickly, so keyword hints can be updated often; 2) the keywords are pretty specific, and the campaigns only bring more traffic, which then reinforces the lead position; 3) Rakuten has an almost monopoly-level stranglehold on web publishing in Japan. If two companies conflict over a desired keyword, Rakuten can probably work it out between them (ah, the Japanese sense of controlharmony). It’s kinda like when AOL or Yahoo could get away with this back in the day.

I was talking with a friend of mine, who recently got the .mobi version of his company’s domain, but hadn’t done anything with it.  He asked for advice, and I offered up the following crude thumbnail that Internet (not web) information can be broken into three main types:

1. Transactional snippets of information: flight times, restaurant addresses, bank account info, sports scores, short text email, answers to salient questions.  Mobile phones do very well at this– iphone or not– just about everyone with a phone participates in this type of internet transaction.  If you see it that way, the iphone app store really becomes a collection ot targetted info queries all eye-candied up.
2. Search-centric information: Any business model that centers around the shear number of stuff for sale/rent/download/sharing is really a search-centric model and not a browsing model.  Amazon, zappos, walmart, piratebay, and wikipedia are all search-centric.  So is Yelp.  Notice how all of these operations are either exploiting a mobile strategy fairly well (half-way), or are likely close to one.
3. Browsing ecommerce information: anything that is a context-rich window-shopping experience doesn’t do well on a mobile phone.  The screen isn’t big enough, and we don’t have the patience to make the Solomon’s decision of either viewing a stripped down version of the site or spend the time endlessly scrolling around to see the website in it’s original layout.  Etsy, Borders magic shelf, Dell, or other ecommerce sites that are inherently dependent on the web-surfing serendipity of the site won’t do well on a mobile phone– at least not without some major rethinking on what works and what doesn’t.

My friend is in an information-sharing business.  He handles sets up relationships between business partners, and brokers deals where he sees possible matches.  His real strength is the depth of information, but he certainly has plenty of small transaction-level updates that would be more valuable if offered in near real-time.  My advice to him was to look through all the activities he does, find those that are in the first bucket, and go with the .mobi version.