{"id":319,"date":"2007-05-25T23:34:47","date_gmt":"2007-05-26T06:34:47","guid":{"rendered":"http:\/\/wapreview.com\/?p=319"},"modified":"2020-10-09T19:59:19","modified_gmt":"2020-10-10T02:59:19","slug":"local-search-shootout","status":"publish","type":"post","link":"https:\/\/wapreview.com\/319\/","title":{"rendered":"Local Search Shootout"},"content":{"rendered":"
This piece started as a review of Local.com<\/em>‘s new mobile site (mobile.local.com<\/a>). It’s got a simple, efficient user interface, understood addresses in every format I tried and it was very good at determining what I was searching for and returning relevant results. Most importantly, Local.com seems to return more results<\/strong> for my queries than other local search engines. I wanted to quantify what I was seeing and came up with some simple metrics for comparing local search engines.<\/p>\n The things I look for in a local search are, first and foremost, the number of relevant local results returned. Second comes usability which I define as a simple, intuitive interface that requires the fewest possible keypresses. Usability is also enhanced by features that save time and effort, like remembering the locations and search terms I’ve used in the past, or make the results easier to understand and use, like maps.<\/p>\n Here are the tests I came up with:<\/p>\n The winner! Local.com 27 points<\/strong><\/p>\n mobile.local.com<\/a><\/p>\n Sushi: 9 in 2 miles from 94132 = 9 Second Place: Windows Live 26 points<\/strong><\/p>\n mobile.live.com\/search\/local.aspx<\/a><\/p>\n Sushi 11 in 2 miles from 94132 = 11 Third Palace: Google 17 points<\/strong><\/p>\n www.google.com\/xhtml?site=local<\/a><\/p>\n Sushi 1 in 2 miles from 94132 = 1 Fourth Place: Go2.com 14 points<\/strong><\/p>\n wap.go2online.com<\/a> (review<\/a>)<\/p>\n Japanese Restaurant (‘sushi’ doesn’t work) 6 in 2 miles from 94132 = 6 Notes: Go2 uses a combination of category lists and search boxes. It has a quirky user interface with multiple paths to the same result. The first time I ran the searches it took me 98 clicks. I was able to perform the same search in only 55 clicks after trial and error to find the most efficient path. I had to search for “Japanese Restaurant” as searching for “sushi” returned nothing (I didn’t count the wasted keypresses required to discover that). Fifth: Ask.com 9 points<\/strong><\/p>\n m.ask.com<\/a> (Review<\/a>)<\/p>\n Sushi 0 in 2 miles from 94132 = 0 —<\/p>\n Sixth: Earthcomer 7 points<\/strong><\/p>\n mobile.earthcomber.com\/phone\/intro.do<\/a> (review<\/a>)<\/p>\n Sushi 2 in 2 miles from 94132 = 2 Notes: EarthComber requires picking a search category from a series of lists. There is no category for Western Wear or anything similar so the click count is based on searching for “Mens Clothing”. Conclusions<\/strong>: Local.com returns the most results and beats Windows Live overall by a single point, none of the others are even close.<\/p>\n One thing that surprised me is that the two search engines that let you pick locations and things to search for from lists required many more keypresses than the ones where you key addresses and search terms. It’s conventional wisdom in mobile interface design that users refuse to type on phones and that a wizard-like series of pick-lists is the optimal design. That may be true for some types of mobile sites where there are relatively few items to choose from. But for search, typing using T9 takes many fewer keypresses, is faster, and I think less frustrating and more powerful. For example, EarthComer had extremely long scrolling lists of search categories. It took a frustrating number of keypresses to find that there was no category for “Western Wear” or anything similar (I didn’t count those keypresses in EarthComber’s total).<\/p>\n All of these local search engines, at least at the URLs listed, are US only except Google which seems to cover just about every place in the world and Windows Live which searches Canadian locations in addition to the US. If there’s a good local search engine in your country please share the URL in a comment.<\/p>\n","protected":false},"excerpt":{"rendered":" This piece started as a review of Local.com‘s new mobile site (mobile.local.com). It’s got a simple, efficient user interface, understood addresses in every format I tried and it was very good at determining what I was searching for and returning relevant results. Most importantly, Local.com seems to return more results for my queries than other local search engines. I wanted to quantify what I was seeing and came up with some simple metrics for comparing local search engines. The things … Continue reading \n
\n– ‘sushi’ near ‘94132’ which is a largely residential San Francisco zipcode. One point for each sushi bar found within 2 miles
\n– ‘Western Wear’ near ‘Dublin CA’ a San Francisco suburb. I gave 1 point for each shop selling cowboy clothes within 10 miles of Dublin, Ca
\n– duplicate and irrelevant results waste time and bandwidth so subtracted 1 point if there were any.<\/li>\n<\/ul>\n\n
\n
\na) Click to call<\/strong> – This should be obvious, we are using a phone, the technology to make a call with a single key press exists and is easy to implement – 2 points.
\nb) Driving or walking directions<\/strong> from my starting point to my chosen target. Pretty much essential when you are in unfamiliar territory – 2 points.
\nc) Maps<\/strong> – appropriately sized, readable maps showing each business found in the context of nearby streets are a help in finding the businesses.
\nd) Saved Locations<\/strong> – Applications that remembered recent locations and let the user chose and reuse them got a bonus point.
\ne) Saved queries<\/strong> – Nice to have if you search for Espresso or Thai Food frequently – 1 point.<\/li>\n<\/ul>\n
\nWestern Wear: 8 in 10 miles from Dublin CA = 8
\nmaps = 1
\ndirections = 2
\nclick to call = 2
\nkeypresses 47 (tie with ASK and Live for best) = 5
\nno saved locations = 0
\n—-<\/p>\n
\nWestern Wear 5 in 10 miles from Dublin CA = 5
\n1 duplicate = -1
\nmaps = 1
\ndirections = 2
\nclick to call = 2
\nkeypresses 47 (tie for 1st.) = 5
\nsaves locations = 1
\n—-<\/p>\n
\nWestern Wear 6 in 10 miles from dublin ca = 6
\n3 dups = -1
\nmaps (113×113) = 1
\ndirections = 2
\nclick to call = 2
\nkeypresses 47 (tied for 1st) = 5
\nsaves locations = 1
\n—-<\/p>\n
\nWestern Apparel 3 in 10 miles from Dublin CA = 3
\nno maps 0
\ndirections 1
\nclick to call = 1
\nsaves locations = 1
\nsaves queries (categories) = 1
\n55\/98 (ave = 76) clicks (5th place) 1 point<\/p>\n
\n—<\/p>\n
\nWestern Wear 1 in 10 miles from dublin ca = 1
\nmaps = 1
\ndirections = 1
\nclick to call = 1
\nkeypresses = 47 (tied for first) = 5<\/p>\n
\nWestern Wear 0 (not on list) in 10 miles from Dublin CA = 0
\nmaps = 1
\ndirections = 1
\nno click to call = 1
\nsaves locations = 1
\nsaves queries (categories) = 1
\n119 clicks (mens clothing) (last place) = 0<\/p>\n
\n—<\/p>\n