Google and the Case of the Disappearing Location Modifiers

One of the most interesting tidbits that I’ve come across in the past week is the following on the Think with Google site:

In September 2015, we shared that “near me” or “nearby” searches on Google had grown 2X in the previous year. Now, just two years later, we see that behavior has continued to change . . . we’re now seeing a shift toward dropping location qualifiers (like zip codes, neighborhoods, and “near me” phrasing) in local searches, because people know that the results will automatically be relevant to their location—thanks to their phone …

In fact, this year, search volume for local places without the qualifier “near me” has actually outgrown comparable searches that do include ”near me.” Over the last two years, comparable searches without “near me” have grown by 150%.

(emphasis added.)

So while these “near me” mobile queries have grown significantly over the past couple of years, Google is saying they’re now outpaced by implicit local queries (e.g., best sushi) that carry no modifier to alert Google to “local intent.”

Google has reported different numbers over the past several years regarding the percentage of mobile queries that have local intent. First it was one-third (33%) then 40% and back to 30%, which is the current official position. (Informally I’ve heard Google employees express larger estimates in the past.)

Putting aside the percentages, an interesting question arises: how is Google now determining local intent as modifiers disappear? At LSA16 Google director of engineering Chandu Thota said (I’m paraphrasing) that mobile search is inherently local and Google regards all mobile queries as potentially local in nature.

Google must be assuming queries associated with certain categories of information (e.g., restaurants, medical, automotive) are inherently or predominantly local. Another tactic may involve watching the relationship between mobile queries and offline visits.

I’m speculating here entirely, but it would go something like this: Google sees “sushi” and then later sees some percentage of those mobile searchers showing up in sushi restaurants. Over time Google would fulfill those “inherently local” queries with a local pack, etc. in the absence of a modifier.

I’m sure there are other inputs and data sources that Google is using to determine which results to show. One might be the relationship between ambiguous queries and clicks on local results, which would also reveal local intent. If more users click on the local results vs. non-local results we have a pattern over time that can reveal the intent behind ambiguous searches or categories of keywords.

My position has been that if we look at fulfillment — where people go to buy — then many more searches should be considered “local” or potentially local. For example, my search for a stove or a jacket is probably ultimately a “local query” because in 9 out of 10 cases I’m going to buy the item in a store. Most service-business related queries are local for the same reason.

Yet some of these searches, ultimately fulfilled offline, may have no specific local or non-local intent at the outset. Teeth whitening is such an example; I may be curious about the procedure at first and then I ultimately go to a cosmetic dentist to perform it. Accordingly, so-called “upper funnel” queries may be more research oriented and “neutral,” whereas “lower-funnel” queries may be more “local” in the sense I mean here of offline fulfillment.

It’s interesting to consider how Google is making these determinations about which searches are “local” as location modifiers disappear. The company must be assuming — and acting upon that assumption — that much more than 30% of mobile searches are local.

You can follow any responses to this entry through the comments feed.

No Responses to “Google and the Case of the Disappearing Location Modifiers”

  1. Video: Last Week in Local August 28th - Local University says at

    […] ScreenwerkGoogle and the Case of the Disappearing Location Modifiers […]

Leave a Reply