A summary of interesting tidbits provided by a press release from mInfo of China:

-WAP is currently actively used by only 30-40 million of the 450 million mobile users in the country, whereas SMS has over 90% adoption

-Searches were spread fairly evenly amongst the basic subject areas of Local Search (41%), Informational Search (31%) and Rich Content Search (28%).

-Local search involves finding directory information for locations such as bars, hotels and ATMs

-The five leading search categories in 2006 were: 1. Dining/Recreation 2. Jokes/Riddles 3. Ring tones/Pictures 4. News/Stock 5. Weather forecast

-The query model for mobile search is also quite different from web search. Mobile searchers tend to use longer query strings (5-6 words/query) vs. web searchers who average about two words per query. Mobile searchers tend to input phrases with qualifiers to improve specificity in an effort to increase relevance and get more precise answers.

-The average web search session usually involves 5 or more iterations of keywords with countless clicks on the resulting links. Contrarily, the average mInfo search session requires just 1-2 queries for users to find their desired answers.

-Search diversity is much higher on mobile vs. the Web. According to traffic data from a leading Chinese Internet search engine, the top 1000 keywords account for over 70% of all search traffic. However, for mobile search, the top 1000 queries account for only about 20% of the searches

-The 19-24 year old group seemed most interested in ring tones and pictures while the 25-29 year old group was more heavily weighted towards local search. The 30-34 year old group tended to spend more time on travel and news related searches.

There is a fair bit of optimism that mobile (see Get Ready for a Surge in Mobile Search) is going to dominate local search in the very short term. The numbers are mouth watering to venture capital: 200 million cell phones in the US alone and research that 30% will use local search on their current or next cell phone. I am sure that it will happen…. someday

Research on actual current users tells a different story. Here is the data from Comscore’s recent research:

Wireless Users’ Internet Accessibility (via Phone) October 25, 2006 ­ November 1, 2006
sample size = 1,708

Top Reasons for Accessing the Internetfrom a Wireless Phone

(Among those who subscribe to Wireless Internet Option)

Let’s do the math: 200 Million cell users of whom 17% have & use internet access and of those only 24% (lets be generous and assume some more unique users from the other categories and say 33%) do some sort of local search. Total audience:11 Million. Nothing to sniff but not yet a huge market

Continue reading Comscore’s numbers on mobile search →

Greg Sterling has provided an excellent summary of local search volume and profitability by analyzing ComScore’s December data:

comScore basically defines “local search” as Internet yellow pages, the queries on the local products of search engines (e.g., Yahoo! Local) and searches on general engines with geographic modifiers (e.g., “Denver Attorneys”). comScore (7/06) estimated local search to be 13% of overall Internet search volumes….. it would mean that on an industry wide basis “local search” is today hypothetically worth just over $104 million per month or in excess of $1.2 billion per year….The local ad market (all media) is worth roughly $100 billion….So one would reasonably expect many billions in SMB and other locally targeted ad buying will flow online,

But Greg points out: But in search in order to capture more of these local dollars there need to be more local searches or those with “local intent” need to be better monetized….But, fundamentally, if you’re Google, Yahoo! or MSFT and want to get more local money into search, you have to build more inventory: local searches.

Ultimately despite all of the startups in local and all of the talk of delivery of ads on cell phones, the local search market won’t really take off until there are more searches by real people. Certainly the current crop of portable devices dosn’t make that easy and Yahoo and Google actively hide their local data on-line.

Perhaps the iPhone,or something like it, will allow this significant hurdle to be overcome by both making local searching easier and not requiring every website to reformat their output. Another strategy might be that if you can’t get people to come to the mountain you move the mountain to them…with a Google Kiosk in every mall. (Now that’s local search.)

For now, Local is a niche within a niche and it will remain so for a while. In some ways though, this is not such a bad thing (other than hundreds of startups going out of business), in that much of the data and the interfaces are not quite ready for prime time…when they are they will be adopted and used and the revenue opportunities will be evident.

Google Earth Meets Windows Live “Virtual Earth” Search Engine Roundtable –
Extremely impressive release by Microsoft with Virtual Earth. It is basically a Google Earth product with “dramatic 3-D mapping” in …

Windows Live Local Gets “Virtual Earth” 3-D Cities Search Engine Watch (Greg Sterling)-
Tomorrow Microsoft will bring dramatic 3-D mapping to the browser window. Windows Live is set to launch photorealistic 3-D imagery for the “urban core” of 15 U.S. cities: San Francisco, San Jose, Seattle, Boston, Philadelphia, Los Angeles, Las Vegas, Detroit, Phoenix….

Microsoft tries 3D maps, fails ZDNet –
A very cool (but for me, almost unusable) version of Virtual Earth was released today by Microsoft called “Virtual Earth 3D”. Instead…

I, being a Mac guy on a PowerPC, have not been able to see this work…

In yesterday’s issue (11/2/06) of the NY Times they published an interesting article called: Basics: Pictures, With Map and Pushpin Included on the nascient field of photo geocoding.

I had been oblivious to this technology before but after reading this article I realized that it will have a huge impact on Local Search going forward. Clearly the technology is just now getting into the hands of people who are interested in geographically tagging their images with GPS data. Like all new technologies it might take 5 years or maybe even 10 to be widely adopted and deployed so that the average housewife is geotagging her family vacation images and sharing the resulting map with Grandma.

The possibilities are endless and Bill Slawski pointed out an interesting patent for some of those. He also pointed out a Microsoft technology demo. It gives some idea  of how maps and photos might interact.  ( I can only imagine that while it might change the details of  the husband  & wife  mapping argument it won’t change the nature of it.) Continue reading When will GeoCoding impact Local Search? →

Greg Sterling had an interesting piece on local database accuracy

What did they find? Shockingly, they discovered there was something wrong (minor to major issues) with 47% of the listings! That’s a remarkably large number. I was truly surprised.

The discussion in the post is of equal interest and well worth a read. My experience with Google’s data is similar. I beleive that it was originally purchased and maintained by InfoUSA and was riddled with problems. The question is whether the partners/local businesses/Search professionals of Google can get the data straight and keep it straight.