The recent Google Local Update (aka Pigeon) was very disruptive. Although as I wrote at the time and reiterate, only when we actually look at measurable results can we both understand its actual impact and perhaps understand more of what was happening and what changes Google has made.
There were two main parts to the update as expressed by Google:
– Local search results that are tied more closely to traditional web search ranking signals.
– The new algorithm improves their distance and location ranking parameters.
My tea leave translation of that?
– We are moving the algos that dictated page and location prominence closer together. We are integrating brand preferences and entity rank into our main algo and using more of that as a common ranking process across both the web and local results.
-We are better able to ascertain location in both mobile and desktop environments and have, for the most part drawn smaller radius around the presumed location of the searcher from which to draw the local search results.
From where I sit, the first part of this change has been happening for a while and as David noted at SEL was largely seen first with the Hummingbird roll out.
It also appears to me that the improvements to their distance and location ranking parameters and the attendant redrawing of the local boundaries, at least on the desktop, is actually causing the more dramatic shifts in measurable desktop results.
Barbara Oliver & Co Jewelry (and others that I have looked at as well) was strongly impacted by that second change. The search radius, which had previously expanded to include the shopping suburbs of Buffalo (Williamsville, Amherst etc) had once again contracted to just include the city of Buffalo proper. You can see this when you do a search like Jewelry Buffalo. Google had effectively reduced the diameter of the search radius from 12 to 6 miles
The impact of the reduction in local Pack visibility for Barbara is obvious in Google’s My Business Insights with reduced impressions:
But Barbara Oliver has always done well organically and the loss of the Pack results still have left many of her organic results above the pack on that same search. So what happened there?
When we compare the last 9 days (since the update) to the previous 9 days and drill into her primary geography we see a somewhat different, more positive story about her website traffic. Obviously data is limited and it is still early but the net affect was upward.
Even more positive though was the increases she saw in neighborhoods surrounding Buffalo:
This last bit of improvement shown in the analytics at the more granular levels of suburbs and neighborhoods is interesting. As the users move more and more to mobile search not only has there been a decline in city level searches (i.e. jewelry buffalo) but the searches will be occurring with Google having a much better handle on the actual users real location rather than a very approximate desktop location.
As I drove up to Buffalo last week, I did an experiment where I requested a non geo modified search query (self storage) that returned mobile Map Packs and I did so about every 5 minutes as I approached Buffalo from the rural south towns.
Initially the iPhone Google app returned results within 7 mile distance of me, this dropped to 5 miles and then as the density of storage companies increased closer to urban Buffalo the stated distance from me dropped to 3 miles. (It is interesting that the iPhone app actually states the distance as part of the results when using voice search.)
When I added the modifier “nearby” the distance dropped to 2 miles. And about every 5-10 minutes, as I headed down the road, Google returned different search results within those increasingly restrictive radii. Upon arrival to the suburbs, I voice searched for “Jewelers” and the distance noted was 5 miles. The search for “DUI lawyers” (Google must think I am a reprobate) was 7 miles and did in fact return a pack local result unlike the desktop search.
This real world testing of searching via mobile as you move through an urban landscape is something that MozCast or Whitespark can not begin to measure using their tools. To a large extent we are forced to drive blind on what is really happening across the broad range of devices & locations that searchers are using.
The bottom line in all of this? Searcher behavior is changing and Google is changing right along with it. The problem that we have is that we don’t really have the tools to fully understand some of these changes and really measure their impacts on the local business.
So far though, at least in the case of Barbara, while she did give up some important pack results she has seem some modest gains in web traffic. Let’s hope that the (as yet unmeasurable) impact on her business is neutral.Pigeon - An Anecdotal Impact Report by Mike Blumenthal