Ranking factors for local search (and Google maps navigation and business listings) differ from signals that might be used in Web Search.
I’ve seen suggestions that requests for driving directions to a place be a possible indication of a place that should rank highly in local search. A patent application published this week points to another real world signal that could possibly be used to rank places in Google Maps.
On a Google Help page titled, “Improve your local ranking on Google“, Google tells us that local rankings in Google are based upon “relevance, distance, and prominence.” Relevance is often how relevant the name of a business might be to what you searched for. Prominence has to do with how frequently a place might be mentioned across the web, in a manner that provides factual information consistently and correctly. And the “distance” factor has traditionally been based upon a distance from where you are searching from. Imagine if Google changed that to a different way of looking at your location. One that works better with mobile devices that have built-in location tracking.
If you use Google Maps, you can see that Google provides a location history that you can edit. Google keeps track of the places you visit and that it may think you visit and it tracks different times and distances to possible destinations that you travel to. When you perform a search, it may consider a likelihood that you may visit certain places, based upon your location history, and a distance from places in your location history could be what Google uses to rank your local search results. If you haven’t looked at your location history, it’s worth looking at. Since, Google may start using your location history when suggesting local results to you. This is one of the benefits of carrying around with you a phone that can keep track of your location and share it with Google all the time.
If you’ve visited a pizza place in the past, Google might consider that when deciding upon a place to show you in search results, because it might consider that to be the most likely place that you might go again.
This change in how Google may treat distances as a ranking factor in local search results is described in a patent application that Google published at the start of November. The patent application is:
Ranking Nearby Destinations Based on Visit Likelihood and Predicting Future Visits to Places From Location History
Inventors: Guang Yang, Tushar Udeshi, Andrew Kirmse, Emil Praun, Pablo Bellver, and Keir Banks Mierle
Publication Number: 20160321555
Filing Date: September 11, 2016
Publication Date: November 3, 2016
In some examples, systems and techniques can determine a respective visit likelihood for each respective destination of a plurality of destinations based at least in part on a respective distance between the respective destination and a geographic location from a location history associated with a user and a comparison between a time associated with the geographic location and a visit likelihood distribution across time. The systems and techniques can then sort at least some of the plurality of destinations. In other examples, systems and techniques can determine whether a user is likely to visit a place during a future instance of a timeslot based at least in part on a location history associated with the user. The systems and techniques can then output information relating to the place prior to the beginning of the future instance of the timeslot.
The patent provides a lot of details on how local search engines work, and this section on how location prominence scores are calculated captures that aspect of local search in a way that is worth repeating:
The local search might also return a “prominence score.” A prominence score can be used to rank more prominent or well-known businesses ahead of less known businesses within the radius because it may be more likely that a user visited the better known business than the one that is geographically most proximate to the geographic location measurement. The location prominence score may be based on a set of factors that are unrelated to the geographical area over which the user is searching. In some implementations, the set of factors may include one or more of the following factors: (1) a score associated with an authoritative document (such as the Web page for the business); (2) the total number of documents referring to a business; (3) the highest score of documents referring to the business; (4) the number of documents with reviews of the business; and (5) the number of information documents that mention the business. In other implementations, the set of factors may include additional or different factors. Further information regarding prominence scores can be found in U.S. Pat. No. 7,822,751 to O’Clair et al., entitled “Scoring Local Search Results Based on Location Prominence,” issued on Oct. 26, 2010 and assigned on its face to Google, Inc.
The patent also provides a way of calculating the likelihood that a person might visit a particular business, based upon such things as whether or not it will still be open or open yet, if someone were to drive there. Another indication of the likelihood of a person visiting a place would be if they had visited the place before. The patent provides some additional things that might be considered in determining a likelihood that someone might choose a particular place.
So this patent takes the concept of distance as a ranking factor and changes how it may be used to rank business results based on a location associated with a mobile device that tracks a persons’ changing location instead of a stationary location associated with a desktop computer.