Yesterday Microsoft announced that in 24 countries, including the US and UK, Nokia (Navteq) will be delivering traffic data to Bing Maps. The company also said that it would start using Nokia’s geocoding services in several (unnamed) countries to improve directions.

In May last year I wrote Bing Maps To Be Powered (Replaced) By Nokia? A source argued to me that over time Nokia would gradually replace most of the infrastructure behind Bing Maps. I was very surprised at the time but it appears to be happening.

This mapping integration was allegedly, according to my source, one of the sticking points that caused Nokia to pass on Android as a platform. Google wouldn’t similarly agree to a commingling of Google Maps with Nokia on the back end. However if Windows Phone fails to deliver for Nokia — so far results have been modest — the company may be compelled to embrace Android as a survival move.

Nokia is also behind the new Yahoo Maps. My suspicion is that Nokia hopes to eventually make these third party mapping services part of its local ad network.

Related Entries

• Microsoft And Nokia Unify Maps On PC, Mobile
• Report: Apple Replacing Google Maps With Own Product In iOS 6
• Will Apple’s Move Bring A Real & Perhaps Better Google Maps To iOS
• Microsoft Makes Venue Maps More Visible On Bing
• Bing Maps To Be Powered (Replaced) By Nokia?
• Nokia Now “Powering” Bing Maps
• Powered By Nokia, The New Yahoo Maps Goes Live
• Google Maps Vs. Bing Maps: Summer Vacation Planning Showdown

Wolfram|Alpha now knows the locations of some 2.4 million retail establishments for 1,300 major chains in 220 different categories—gas stations, restaurants, department stores, and much more. It also knows the typical business hours for roughly one third of those locations, which makes queries like that one possible.

When I tried it online (PC) the user experience proved to be very poor. It located me in the wrong city — I’m in an LA hotel right now and it’s probably a case of IP targeting getting it badly wrong. However in search after search (even with geomodifiers in the query) it located me in Dallas, with no easy or obvious way to manually change location. My efforts to do so were consistently frustrating.

When it works the information provided is very basic. Using Google, Yelp, Foursquare or IYP sites is going to be faster and much more effective. There’s also no evaluative (e.g., reviews) or much enhanced information available. In other words, there’s nothing to set local search on Wolfram Alpha apart from other local sites.

I tried the HTML mobile site and the results were no better. I did not use the Wolfram Alpha mobile apps and so cannot report on that experience. However I cannot imagine it’s much better than the PC or HTML mobile site experiences.

It was a mistake for Wolfram Alpha to publicly announce this capability when it’s in such poor shape. The company needs to  work through the usability problems but more fundamentally needs to answer the question: Why would I want to use it for local search?

Chitika “looked at hundreds of millions of impressions between April 8 and 14 and broke down traffic depending on operating system, search traffic, and the type of search query.” The company also broke out local searches (often inferred from the query category).

These data don’t measure activity within or involving apps in any way.

Chitika reported that 54 percent of browser-based web traffic from iOS was being driven via search (Google, Bing, Yahoo, etc.). Chitika didn’t report the breakdown of that traffic by search engine and told me that would require an additional analysis. However other data suggest that Google represents about 95 percent of mobile search traffic coming from browsers.

The data argue that iOS browser activity is more “search-centric” than usage behavior on PCs (Mac or Windows). This is also somewhat counter-intuitive given than search is generally easier to use on a PC than on a mobile device.

In terms of local search, the directional relationship among iOS, Android and Windows is roughly the same as in general search chart. However Mac and Linux-based searches reverse positions in the local search data chart. It’s curious and Chitika didn’t really have an explanation for the phenomenon. I asked whether there could have been an error and I was told essentially “no.”

Chitika is saying that 36 percent of search query volume coming from iOS devices has a local intent. This compares to Google’s 40 percent figure for mobile. In contrast, 28 percent of Android search traffic carries a local intent according to Chitika. Figuring out why is an interesting exercise.

Why would iOS users be doing more local searches than Android users? That’s a version of the larger question about why iOS users might be doing more searches generally than Android users? (Once again, these data don’t reflect or measure activity in apps.)

One partial explanation might be that iOS includes iPads, where there is a lot of search activity. But that doesn’t fully explain these findings.

Below are the overall most-searched categories and the fastest growing across the YP Local Ad Network (Q3 vs. Q4 2011):

The top mobile search categories are similar but not identical to the list in the graphic above:

1. Restaurants
2. Automotive
3. Movie Theaters
4. Beauty Services
5. Hotels

Mobile searches now represent “more than 30 percent of all searches” across the YP Local Ad Network. This number is higher than the mobile percentage of general search, which is more like 15 – 18 percent overall, though higher in some categories.

Last quarter AT&T Interactive said that its mobile users click ads 2X more than PC users. Mobile users are also more inclined to call a business than their online counterparts, who are more likely to visit a website.

The graphic below reflects actions taken by consumers on a business profile/more information page, after the initial click:

In Q4 2011, AT&T said that the top advertisers on the YP Local Ad Network by ad spend were the following:

1. Building Contractors
2. Legal
3. Medical
4. Automotive
5. Financial Services
6. Home Maintenance Services3
7. Pet and Animal
8. Moving and Storage
9. Real Estate
10. Insurance

Apparently not.

Roughly a month ago I spoke with Marc Brombert, the CEO of Implied Intelligence. His company provides a range of data-related services (e.g., enhancement, cleansing, de-duplication) to marketers and publishers. At the conclusion of our call I suggested that Implied Intelligence test the accuracy and completeness of the business listings data on several of the leading local search sites.

Surprise: Yellow pages beat Google for local search

Several weeks later Implied Intelligence sent me the results of its test. They’re a bit unexpected and illuminating. Google, which has probably devoted more effort and resources to local search than any of its competitors, did not come out on top in the test. Overall it placed third. Two yellow pages sites beat it.

Implied Intelligence crawled and hand checked 1,000 independent local business websites in the US (no chains or franchises were included in the test) and compared the information it captured to the data contained on the following sites:

• Bing Maps
• Citysearch
• Dexknows
• Foursquare
• Google Maps
• Mapquest
• Superpages
• Yellowpages.com (YP.com)
• Yelp

The criteria and results

Implied Intelligence evaluated and scored the local search competitors on the basis of the following criteria:

• Coverage (was the listing present)
• Number of duplicates
• Accuracy of information
• Richness of information (presence of additional information beyond business name, address and phone)

The first table below offers a comparison among these sites in terms of basic listings coverage and accuracy. The yellow highlighting indicates the winner in each category.

The table reflects that Google Maps had the most complete coverage: 80 percent of the 1,000 local listings were present. No site had 100 percent of the 1,000 listings. Foursquare had the worst coverage at only 16.7 percent.

In terms of error percentages, yellow pages site Superpages outperformed the others. YP.com had the fewest duplicate listings in the test.

In terms of enhanced information, YP.com was the winner. Reviews and check-in data were not considered because Implied Intelligence felt this didn’t allow for an “apples to apples” comparison across sites. However, had reviews content been included Yelp, Google and Foursquare would likely have fared better.

Superpages the overall winner

Overall Superpages was the winner, followed by YP.com with Google Maps coming in third. Foursquare was the overall loser. However Yelp also didn’t fare that well either.

To many people these results will be a surprise. (They were to me to some degree.) And some people may charge bias. While I didn’t supervise the test and was not involved in its design I can report that Implied Intelligence has no agenda here. I would and do take the results at face value.

Yext began as a pay-per-call network and shifted last year to focus on local listings syndication. Yext works with small businesses but also brands with multiple locations.

Emulating the now defunct Google Tags, Yext developed a real-time data distribution network (“PowerListings“) across the major search engines and local directory sites (including Yelp but not Google). UBL had been trying to develop a similar “real-time” data update capability.

The Yext PowerListings distribution has been incorporated into UBL’s new “UBL Express” product. According to press materials, “The Yext platform allows UBL customers to manage business listings on over 35 search engines and directories via direct-connect instant feeds, as well as view user activity through a monitoring dashboard.”

The combination offers a convenience for many local marketers who might have been trying to choose between Yext and UBL. UBL’s sydnication network fills in many of the blanks in the Yext distribution, which didn’t offer Google Places: “The UBL components will include establishing Google Places profiles as well as syndication of the listings to the databases that feed hundreds of other search engines, social networks and mobile/GPS services.”

The UBL Express package is priced at $599 per year and is the most expensive of the UBL data distribution bundles. Yext is not reselling the UBL product on a reciprocal basis.

Bruce Clay and UBL were embroiled a couple of months ago in a strange controversy over a “Local Paid Inclusion” product Bruce Clay announced. UBL denied that it was involved in any way.

FindTheBest is a kind of “anti-search engine” and the brainchild of DoubleClick founder Kevin O’Connor, who launched the site to provide better information than he believed was available in general search results:

FindTheBest.com was created out of my desire to organize part of the Internet, filter out the excessive junk and present information in a simple, comparable way. While the web is great at making seemingly infinite amounts of information accessible, the Internet can be a little overwhelming in that it doesn’t allow for straight up apples-to-apples comparisons, nor is it transparent enough to allow you to spot those hidden marketing schemes influencing the data you receive.

The site aggregates and presents structured data to enable users to compare products, institutions and services. It aggregates that data from mostly public sources. However the site is not a product or travel search engine per se.

With the addition of local classifieds FindTheBest aims to provide more context and data around the listing to help people make more informed decisions. For example, car listings show how the price of the vehicle being considered compares to the average price of others in the same model year, as well as all cars in its class/category (e.g., sedans). There’s also additional information about the make and model, including gas mileage and safety recall information.

All the listings data right now are coming from third party sources and feeds. Eventually FindTheBest will add self-service but will likely charge to prevent spam. CEO Kevin O’Connor believes that the depth and quality of the data the site has differentiates its classifieds listings from most others in the market, including but especially Craigslist.

FindTheBest would also be a potentially good candidate to provide structured information to Apple/Siri, not unlike Wolfram Alpha. I previously observed that FindTheBest is, in one way, the realization of the now defunct Google Squared.

FindTheBest has raised a total of $6 million in venture funding to date.

Related Entries

• FindTheBest Succeeds Where Google Squared Failed
• “Anti-Search Engine” FindTheBest Adds Buying Guides, Top 10 Slideshows
• DoubleClick Founder Launches Structured “Decision Engine” FindTheBest.com

In this article I want to focus on a single finding: 49 percent of smartphone and tablet owners are using apps to find local information. On one level this is unremarkable and makes sense; apps are popular and there are lots of apps that use location in one way or another.

We know from Google that 40 percent of browser-based mobile search has a local or offline intent. That already represents significant query volume. But what if half of the local search query volume on mobile devices (and tablets) goes through apps?

That’s not exactly what the survey data show, but it’s a very logical and plausible inference.

We don’t know the total volume of app-based search or local search because nobody is tracking that right now in the same way that PC query volume is tracked. Yet if half of mobile users in the US (including tablets for this discussion) use apps for local lookups and research it stands to reason that perhaps half their queries go through apps vs the browser (read: Google).

Source: comScore, Localeze, 15 Miles

The absolute volume of local-mobile queries could be greater in apps than the browser. This is because smartphone and tablet users (in particular) are typically more active and engaged than PC users as a general matter. The chart above shows that tablet owners conduct (on a percentage basis) more daily and weekly local searches than smartphone and PC owners.

This is “informed speculation” on my part, but I would imagine that if half of mobile consumers are using apps for local search then at least half of local search query volume is happening inside those apps. In terms of which apps smartphone and tablet owners are using for local searches and lookups, comScore found the following:

Source: comScore, Localeze, 15 Miles

My understanding is that this question involved a predetermined list of choices so it does not reflect all apps being used by consumers for local search. For example, 20 million people are using Foursquare but it’s not represented here (unless reflected in “other”).

The top app being used for local search is Google Maps according to the study. However, Yahoo, Mapquest, Bing and others are not that far behind and apparently in a more competitive position vs. their relative positions in search on the PC. Indeed, there are some very interesting implications for Google if Google Maps were to be replaced by an Apple mapping product at some point.

Essentially the UI has been simplified, colors have been muted and there’s an improved “visual hierarchy” at each level of zoom. It’s challenging for me to specifically tell what’s new on Bing Maps, however. Here’s more from a Microsoft spokesperson:

This week, as an extension of the Microsoft-Nokia partnership, a new joint map design will begin rolling out across Bing Maps, Nokia Maps and Windows Phones. This unified map style will feature key elements from Microsoft’s metro design, including strong typography, improved readability and a clean user interface to help people find and use mapping information more quickly. As part of the update, Bing Maps will also be improving its global mapping coverage in countries such as Egypt, Israel, Venezuela and many others, refreshing the maps with new roads, subdivisions and additional refinements.

The new mapping UI and features will appear on both companies’ PC and mobile sites. While there are still UI and user experience differences between Bing Maps and Nokia Ovi Maps, the look and feel is very close.

Below is comparison of maps on Bing and Nokia in response to the query “San Jose Convention Center,” where I am now at SMX West:

Bing Maps

Nokia Maps

For comparison purposes, here are Yahoo Maps (also powered by Nokia) and Google Maps results for the same query.

Yahoo Maps

Google Maps

The new Bing-Nokia maps are less visually cluttered than Google or Yahoo Maps. But some may equally characterize that as a “washed out” quality. Indeed, the assessment of the new mapping UI for Bing and Nokia will depend on your subjective aesthetic preferences.

Whose maps do you prefer?

Related Entries

• Bing Maps To Be Powered (Replaced) By Nokia?
• Powered By Nokia, The New Yahoo Maps Goes Live
• Google Maps Vs. Bing Maps: Summer Vacation Planning Showdown
• Bing To Become Default Search (And Maps) On RIM BlackBerry Devices
• Bing Maps’ “Game Changer”: Hi-Res Aerial Imagery Coming To Entire US and Western Europe
• Bing Maps Overhauls Interface, Exposes Map Apps To All
• Bing Maps Adds Transit Directions

The following were the most frequently searched local content categories in Q4:

Ad performance reported by xAd exceeds comparable online CTRs for both search and display. The network said that average CTRs in Q4 for local search ads were 7 percent, while locally targeted display ads yielded 0.6 CTRs.

Beyond the initial click xAd captures “secondary actions” (i.e., calls, map lookups). The company said that among those who clicked on ads, 37 percent of search clicks and 5 percent of display CTRs delivered these secondary actions.

The secondary actions were further broken down by xAd. Calling a business was the most frequent secondary action taken by local searchers, while maps/directions lookups was the most common action among those clicking on display ads.

Another interesting observation is how display responders (above right) sought more information or reviews but local searchers did not as frequently. This likely indicates the more “directed intent” of search-ad responders vs. display clickers who were being exposed to new information or businesses.

One of the most interesting findings in the xAd report shows how search and display ads reach people at different times of day. While the complementary nature of search and display advertising online is well established, xAd shows by daypart how the two can work together in mobile.

Local search peaks during the middle of the day and declines in the evening. However display exposure peaks “after hours” when people are in leisure mode.

Finally, xAd said that among all mobile subscribers, “mobile browsers are still the primary access method for local-search information.” However for iPhone and Android owners, “in-app access is preferred by a large margin over in-browser access.”

Others have reported similar data, confirming the preference and trend toward apps among smartphone users. “We expect that trend to continue as more users of feature phones switch to smart devices,” explains xAd.

While Google “owns” 95 or more percent of mobile-browser based search the same is not true in the app world, which is more verticalized.