Google+ Local Centroid - NOT City Center! Affects Ranking!

[Linda Buquet]

In our info-packed InsideLocal Webinar last week Mike Blumenthal dropped a big bombshell that I think surprised all of us. Even me! If Myles didn't have everyone on mute I think we would have heard a big gasp!

The centroid is not the centroid???
Not only that, but you'll see in my example at bottom the centroid MOVES & CHANGES!

Mike said the centroid is NOT city center. Which is what I've always assumed. He explained that it's calculated by Google to be the center of that particular industry hub within that city.

When he said that I about fell off my chair. I'm like "why didn't I ever think of that - I've seen it I just didn't realize what I was seeing." Mike has this way of finding things I'm looking right at and just missed the significance of. That's why we call him 'the professor'!

When you do a search and your client does not rank, check the small map right on Google search and see if they are off the map or outside the cluster of businesses Google thinks is the centroid. Examples below.

​

If you had a Dentist in downtown Atlanta or just south of town they WOULD NOT RANK. Because they are cut out of the map based on the centroid radius Google has determined to be the center for Dental offices near Atlanta.

(See my example of this problem with one of my clients below the next screenshot.)

​

In this example the proximity issue is magnified and the "centroid" for car dealers is much further north of the city. You could have a client near downtown and think "they are closest to city center why can't I get them in the pack?" Answer is - they are not in the centroid, even though they are near center of town.

REAL WORLD EXAMPLE:

I actually have this problem with a Pediatric Dentist. I have them ranked #2 in organic and #1 in maps. Once I finished all their optimization I moved them from C to A in the pack.

But they keep popping in and out of the 7 pack. Lately it's crazy! One week they are in and the next they are out. (And they REALLY insist on being in the pack and don't want to be in organic, because they've worked so hard on their reviews.)

The 1st time they disconnected from blended and dropped out of the pack, I found a dupe and thought that knocked them out of the pack. Fixed it. They came back.

But then they dropped out of the pack and started jumping in and out. Even though they have NONE of the standard problems that cause a listing to disconnect from blended.

They watch that pack like a hawk and email me every time they pop up to #2 organic (above the pack) because they don't want to be there. They want IN the pack!

Over the course of time, I've taken screenshots of their rankings and started noticing something - they are pretty far south of town. There is another high ranking Dentist on the north of town that would drop out of the pack every time they did. Then I started noticing that whenever they would pop out of blended, it would be when Google either shrank the map radius (cutting them both out) or moved it a little north - literally cutting my Dentist out of the picture.

Sooo with the new knowledge Mike provided about the centroid, I went back and looked at all the screenshots. Sure enough, that centroid radius moves and changes. The only time they pop out of the pack and up to pure organic is when they are cut out of the search radius Google has determined is the centroid for that industry - because they are just a little too far south!

In one screenshot where they were back in the pack at letter A - I noticed the map was zoomed WAY out to like a 5 city area. So they showed up because the map expanded wide enough to show them.

So possibly the map and centroid opens up or shifts, if some Dentists in the more outlying areas do some press releases, or get a bunch of 3rd party reviews for instance???

Mike referred to a patent that covers all of this. I'm not sure if this is the one he's referring to or if it's some other. I'll ask him to shed more light on this.

Patent US8046371 - Scoring local search results based on location prominence - Google Patents

Check out that patent. Seriously you need to read and try to understand it. There is a TON of rich info about the algo including weight of citations and reviews and how all that figures into the calculations.

The patent uses weird terminology and keeps referring to "documents" but essentially I think part of it at least is saying it uses # of reviews and citations (including 3rd party reviews), to determine where the cluster of businesses are that relate to the query.

I know a lot of you were on the webinar and probably gasped when Mike said that about the centroid and I think it turned on a huge light for all of us. (If I remember right it's about 17 minutes in. Sorry, the video is delayed. Hopefully will be posted tomorrow.)

Any questions, comments or examples to share?

I'll alert Mike to this thread and see if he'll weigh in if he has anything to add or link to another patent if that's the wrong one.

 

[Linda Buquet]

Forgot to mention, just discovered Ryan Hill blogged about this the day of the Webinar. He didn't mention the Webinar or Mike's comment but due to timing, I assume that's what prompted his post. Anyway, you can go see some different examples he posted.

 

[Russ]

I'm now very confused.

I have some clients but one in particular that isn't anywhere near the cluster displayed by Google on the map and is actually 15 or so miles south of the cluster; sitting there all by itself but it ranks B for the main keyword City + Roofing. Additionally the NAP isn't even in the City we want to rank for, it's in a distant suburb.

Not trying to be contrary or debunk Mike but for me this doesn't add up at all.

 

[Colan Nielsen]

Great post Linda

I was one of the people gasping when Mike talked about this.

Hmmm...if Google is shifting the map based on quantity of reviews and citations of the industry hub or cluster of businesses, it would make sense to try and aid the competing businesses within our cluster to get reviews and citations..

 

[Linda Buquet]

Great post Linda

I was one of the people gasping when Mike talked about this.

Hmmm...if Google is shifting the map based on quantity of reviews and citations of the industry hub or cluster of businesses, it would make sense to try and aid the competing businesses within our cluster to get reviews and citations..

I know I heard you gasp even though you were muted!

Actually I'm uncertain what causes the map to shift, shrink or expand. Just guessing based on the patent.

Now that I think about it I bet clicks affect it. So if a bunch of people are clicking to the far north businesses, the centroid moves further north??? Not sure.

I'll email Mike now and see if he can shed any light.

 

[Linda Buquet]

I'm now very confused.

I have some clients but one in particular that isn't anywhere near the cluster displayed by Google on the map and is actually 15 or so miles south of the cluster; sitting there all by itself but it ranks B for the main keyword City + Roofing. Additionally the NAP isn't even in the City we want to rank for, it's in a distant suburb.

Not trying to be contrary or debunk Mike but for me this doesn't add up at all.

But he's on the map that shows in Google search right? The point is the size of the map and where it's centered. Not specifically where the cluster of businesses are.

At least I think that's the point, but maybe I didn't make it clear.
The centroid cluster helps determine where and how big the map radius is.

Plus I think if you have a really high ranking client that has lots of the right stuff going for them it can force the centroid map radius to expand and include your client. (Just guessing here.)

I have a Dentist that ranks double #1. #1 in organic and A in blended. He's so far south of the city that most of his patients come from the next state south.

For a long time I could not get him on page one in the pack because the map radius was zoomed in too tight and he was off the map, down south. But now the radius expanded I think in part due to one other Dentist down near him that also gained some prominence so the map expanded down to include them too.

Again I'm only guessing. No one knows for sure. PLUS with Google there are always exceptions to the rule and anomolies, so nothing is ever black and white!

 

[Linda Buquet]

Just throwing this out there - not sure it's true and I'd need to do more research...

A long time ago, maybe a year (long in Google time)... I noticed that when there would be an Adwords pin way north for example, the map radius would expand to show that business. Then if the Adwords ad was gone, the map would shrink back down to normal radius for that particular area. So possibly paid ads play a part.

If so that gives me an idea for my client. They want to start doing Adwords anyway - Maybe if they have an Adwords pin on the map it will help widen the map radius so they show up in the pack. We'll experiment and see what happens.

Knowing Google, there are hundreds of things that weigh into the centroid calculation which makes it hard to pin down and which accounts for everything shifting like the wind at times.

However what Mike said plays a big part and helps make sense of things.

 

[billbean]

That's a really good idea.

 

[Stefano Romero]

I've seen this happen a few times and it depends on the city that you're client is in.
In less competitive markets, you will find it more likely that someone outside of the main cluster will be included in the 7pack because what Google considers the "city" will include the surrounding areas and suburbs. That being said, this also happens in major markets if the local signals are strong enough.

The cluster of businesses being the "Centroid", as Mike referred to, tells us where the map will be centered around, but that doesn't necessarily mean that you must be in the cluster zone in order to rank in the 7 pack. If you're local signals are above par when compared to the competition, then you may outrank them as well.

I can only speculate based on what I've seen in my experiences, but this seems to be pretty consistent. And I agree with Linda, nothing with Google is ever black & white. Hope this helps guys.

 

[Linda Buquet]

Just throwing this out there - not sure it's true and I'd need to do more research...

A long time ago, maybe a year (long in Google time)... I noticed that when there would be an Adwords pin way north for example, the map radius would expand to show that business. Then if the Adwords ad was gone, the map would shrink back down to normal radius for that particular area. So possibly paid ads play a part.

If so that gives me an idea for my client. They want to start doing Adwords anyway - Maybe if they have an Adwords pin on the map it will help widen the map radius so they show up in the pack. We'll experiment and see what happens.

Knowing Google, there are hundreds of things that weigh into the centroid calculation which makes it hard to pin down and which accounts for everything shifting like the wind at times.

Have not researched in detail but think the above is no longer true, if it ever was. (Again above was just sharing an old observation that I was not sure about.)

But for my Pediatric Dentist, when I checked yesterday their were Ad pins that stretched the map radius out way beyond where they are located and they are still on top in organic AND #1 on Maps, but still disconnected and not in the pack.

 

[Damian Rollison]

Thanks for posting on this topic, Linda. It certainly caught my ear as well, in particular because I have heard more than one local SEO offer the recommendation that businesses should seriously consider opening an office or moving to the center of their city in order to improve ranking. It sounds like that advice may not be quite correct, and that the ranking-by-centroid factor is much more dynamic than we've assumed. It's a little disturbing to learn this as it means Google is basically ranking businesses, in part, arbitrarily based on proximity to like businesses, solely because of the practicalities of map display.

 

[Linda Buquet]

Thanks Damian, I agree.

Google often works in mysterious ways that don't exactly match up with the real world OR what's exactly fair or right.

However when someone smart like Mike helps us figure stuff like this out, it feels a little less random and more predictable at least.

 

[drleo]

This centroid theory makes sense now. Funny thing is none of the so called SEO experts I've consulted or talked to knew what the issue was.
I'm an orthodontist and located in the Main City, but for some reason I am only ranked in Google Places pack 7 in South Main City. The bummer is nobody is going to search for South Main City orthodontist. Thus I'm left out cold.
Somebody needs to help me soon as I'm practically invisible...gasp...

 

[Linda Buquet]

Hi Drleo,

Sorry I did get your email, just have not had time to respond yet. Super swamped as always.

I already did some research for you and can see the centroid situation you are referring to, clearly,
however that's not the only issue in your case. I'll email you back as soon as I can.

 

[21Development]

This exact situation is killing me right now. I have a client that has her business running out of her home. It is about 3 miles south of the main cluster of industry related businesses. We cannot get into the 7-pack to save our lives and it is very frustrating. We are ranking first page organically, but the traffic is just too low.

If anyone wants to take a look, I'd be very appreciative. Thanks,

Robert