Pigeon Factors - 5,000 Page Case Study and a Big Takeaway


Linda Buquet

Moderator
Moderator
Joined
Jun 28, 2012
Messages
14,433
Likes
4,283
Adam Dorfman from Sim Partners just shared results of a Pigeon study which opens up a couple interesting theories.

<a href="http://searchengineland.com/factors-affecting-pigeon-5000-page-case-study-205862">Factors Affecting Pigeon -- A 5,000 Page Case Study</a>

At SIM Partners, we decided to do some homework first. We discovered that Pigeon has a number of implications for brands, such as the importance of attaching your local listings to strong domains.

We wanted to know how Pigeon affects crucial factors such as website traffic and ranking performance among businesses that depend on local listings. After analyzing the numbers, we concluded the following:
Below are the main points made and my comments in blue...

1. Pigeon Does Not Equal Improved SERPs (Click above to read full details on this point)

Agree with that one.

2. Strong Domains Matter (Click above to read full details on this point)

We are seeing more EMDs and brands with City in name rank higher that's for sure.

But check out his link analysis using Moz authority. Interesting...
(Then see Michael's comment and my retort.)


3. Location Matters More Than Ever (Click above to read full details on this point)

Everyone keeps saying this - but I'm seeing more outliers rank than ever before.
Oftentimes even 2 - 4 in the pack are from other cities.


4. Don’t Over-Optimize Your Website

He shows a graph that indicates businesses that were aggressive in adding multiple zip codes and/or areas served were dinged.

My reply in comments:

"Especially noteworthy is # of cities or zips on a page of the site. Google changed webmaster guidelines a couple years ago to say that listing a big bank of cities, zips or phone#s is considered KW stuffing. But spammy SMBs like plumbers still do it - in excess. Maybe now practice is being dinged.

If so, that's the only good thing I've seen come out of Pigeon so far."


Anyway like I replied to Michael:

"I don't think of studies like this as pseudoscientific nonsense. I don't take them as gospel either. But it's helpful to me to gather any data people notice and even just opinions like in the LSRF, then weigh it all together to see what makes sense or track patterns between different studies.

Especially with this particular algo - there is no scientific study that could possibly make any sense of it!

It changes 3 times a day, is totally unpredictable and makes no sense at all. Plus what Adam's study observed 2 weeks post Pigeon, is different now AND depends on what time of day you check - due to rotating datacenters with different versions of the algo."


So one big takeaway I have after reading this - one thing I'll be paying attention to as I analyze rankings is whether the site has a big bank of city or zip service areas. I'll be looking to see whether spammy ones are still ranking OR if spammy ones that used to rank are no longer. (Can check by viewing G.ca or Classic maps.)

And as Adam suggests, if you have a client that dropped after Pigeon and has a big list of service areas, you may want to try removing to see if it helps. And as I stated above, that's been a spam signal for awhile now - maybe now it's finally a negative ranking factor.


What do you think???
<meta property="og:type" content="article"><meta property="og:title" content=""><meta property="og:description" content="Pigeon study opens up a couple interesting theories.">
<meta property="og:image" content="http://marketing-blog.catalystemarketing.com/wp-content/uploads/2014/08/pigeonRealDC.jpg">
 

JoyHawkins

Administrator
Administrator
Joined
Jul 18, 2012
Messages
2,180
Likes
1,156
Linda,

I'm with you here - I'm seeing just as many SERPs that have tightened the radius as ones that are including listings from random nearby cities when there are tons of businesses in that city. So my conclusion is the way that Google determines location has changed.

I'm also going through a list of clients right now and looking at before/after Pigeon. The majority of our clients either increased in position or stayed the same. So I'm wondering if businesses that had good SEO practices weren't impacted as much.
 

Jim Froling

Forum Member
Joined
Jul 19, 2012
Messages
135
Likes
58
Too bad summer is over. A 5,000 page study on the Pigeon update would have made a great summer read.

In my view, Pigeon will be a good thing for local SEO and SMBs. Fewer strings to pull under the hood of a website, fewer cheap tactics i.e. listing dozens of zip codes for ranking purposes.

Notice I state "will be" because at the outset I agree wholeheartedly with Linda. Multiple changes in the algo on a daily basis, completely different SERPs depending on what G server handles the query, bogus and irrelevant SERPs for many queries, 7-Packs to 3-Packs to NO Packs (rinse and repeat), etc. etc. Chaos!

I only hope that Nietzche was right. "Out of chaos comes order".
 

Linda Buquet

Moderator
Moderator
Joined
Jun 28, 2012
Messages
14,433
Likes
4,283
Too bad summer is over. A 5,000 page study on the Pigeon update would have made a great summer read.

In my view, Pigeon will be a good thing for local SEO and SMBs. Fewer strings to pull under the hood of a website, fewer cheap tactics i.e. listing dozens of zip codes for ranking purposes.

Notice I state "will be" because at the outset I agree wholeheartedly with Linda. Multiple changes in the algo on a daily basis, completely different SERPs depending on what G server handles the query, bogus and irrelevant SERPs for many queries, 7-Packs to 3-Packs to NO Packs (rinse and repeat), etc. etc. Chaos!

I only hope that Nietzche was right. "Out of chaos comes order".
"Out of chaos comes order" Here's hoping Jim!
 

Lloyd Silver

Forum Member
Joined
Jul 21, 2014
Messages
136
Likes
12
What's your definition of a spammy listing of cities or zip code?

We typically build out location specific landing pages on their website, not just where a location is but the major areas that the business serves. And sometimes building out an entire subdomain with unique content. I haven't seen any sign of a rankings shift on those sites.

Certainly the main location is most prominent but if we want to rank organically outside of that city then location specific parts of the site are critical.

So are you referring to just the practice of things like listing a bunch of cities in a footer?
 

Linda Buquet

Moderator
Moderator
Joined
Jun 28, 2012
Messages
14,433
Likes
4,283
What's your definition of a spammy listing of cities or zip code?

We typically build out location specific landing pages on their website, not just where a location is but the major areas that the business serves. And sometimes building out an entire subdomain with unique content. I haven't seen any sign of a rankings shift on those sites.

Certainly the main location is most prominent but if we want to rank organically outside of that city then location specific parts of the site are critical.

So are you referring to just the practice of things like listing a bunch of cities in a footer?
More details and some examples from my post in 2012 Lloyd:

<a href="http://localsearchforum.catalystemarketing.com/local-search/908-warning-new-google-guidelines-local-seos-must-2.html">Warning: New Google Guidelines Local SEOs Must Know - Could Cause Penalties - Page 2</a>

Just one example and not the worst I've seen, by far...

www DOT rodenhiser DOT com (Footer)

www DOT rodenhiser DOT com/plumbing/plumber-watertown-ma.html (Bottom of page)
 

Lloyd Silver

Forum Member
Joined
Jul 21, 2014
Messages
136
Likes
12
Thanks. As of now I don't see anything wrong with having location specific pages that include unique, quality content. Don't over optimize in the URL. Don't over optimize in the content. Don't list a bunch of cities in the zip code. Use common sense and don't be spammy in general.

But take realtors, there's a legitimate reason for them to have community pages.

Now, does that apply to plumbers as well? Not from a city perspective, I guess. But I still don't think it's harmful and as of now properly optimized and architecturally sound pages are ranking.
 

Linda Buquet

Moderator
Moderator
Joined
Jun 28, 2012
Messages
14,433
Likes
4,283
Just note: the number of zips and cities listed on a landing page in that study and the corresponding drop.

See the 2 graphs under #4 at the original article.

Here is the city one, need to go there to see the zip one.



Source: Search Engine Land


Now I think it's totally appropriate to say we also serve city 2, 3, 4. And you would often naturally do that even in a TV or radio ad, so it's not KW stuffing for ranking. It's just when it gets extreme which I often see in service area businesses that it's a problem.

But still the study is interesting and if you have a client that suffered a drop that has a big bank of cities I think it would be worth taking out and just saying on the site, we serve the whole local region or we serve customers in a 30 mile radius. Or we deliver as far as city 8 and 9 and everywhere inbetween. Just test and see if it helps.
 

Lloyd Silver

Forum Member
Joined
Jul 21, 2014
Messages
136
Likes
12
I'm not reading that the same way as you, perhaps.

I see "Number of Zip Codes Listed on a Location PAGE" and Neighborhoods/Surrounding Suburbs/etc Listed per Location PAGE"

So yeah, that means don't be spammy on a page. But it's not saying don't have multiple location pages. Just to make sure that on each page you focus on one or perhaps a couple of neighborhoods.

To me this almost is a recommendation to use multiple (properly structured) location based pages. That the method of just listing a bunch of cities on a bunch is ineffective and that each city needs to be treated separately.
 

Linda Buquet

Moderator
Moderator
Joined
Jun 28, 2012
Messages
14,433
Likes
4,283
Maybe I said something that lead you astray or miscommunicated.

I never meant to say anything about multiple city pages or landing pages. Those are fine and best practice.

My only comment was listing A BUNCH of cities or zips on the same page, like the example I showed in a link above. You know a page with a bank of 20 cities or kw city, kw city for 20 different cities all on the same page.
 

Lloyd Silver

Forum Member
Joined
Jul 21, 2014
Messages
136
Likes
12
Ok gotcha.

Yeah, I read "the number of zips and cities listed on the site" and thought you were talking about site wide limitations. Like we couldn't have more than a few cities mentioned throughout the entire site and that didn't make sense.

I think we're on the same page : )


 

Linda Buquet

Moderator
Moderator
Joined
Jun 28, 2012
Messages
14,433
Likes
4,283
Ya I meant on a page not site. Will correct.

I knew exactly what I meant in my head. Just multi-tasking and typing too fast so didn't proof to see if what I THOUGHT is what flew out of my fingers.
 

Local Search Forum


Weekly Digest
Subscribe/Unsubscribe


Google Product Exert

@LocalSearchLink

Join Our Facebook Group

Top