Tuesday, April 11, 2017

Announcing the 2017 Local Search Ranking Factors Survey Results

Posted by Whitespark

Since its inception in 2008, David Mihm has been running the Local Search Ranking Factors survey. It is the go-to resource for helping businesses and digital marketers understand what drives local search results and what they should focus on to increase their rankings. This year, David is focusing on his new company, Tidings, a genius service that automatically generates perfectly branded newsletters by pulling in the content from your Facebook page and leading content sources in your industry. While he will certainly still be connected to the local search industry, he’s spending less time on local search research, and has passed the reins to me to run the survey.

David is one of the smartest, nicest, most honest, and most generous people you will ever meet. In so many ways, he has helped direct and shape my career into what it is today. He has mentored me and promoted me by giving me my first speaking opportunities at Local U events, collaborated with me on research projects, and recommended me as a speaker at important industry conferences. And now, he has passed on one of the most important resources in our industry into my care. I am extremely grateful.

Thank you, David, for all that you have done for me personally, and for the local search industry. I am sure I speak for all who know you personally and those that know you through your work in this space; we wish you great success with your new venture!

I’m excited to dig into the results, so without further ado, read below for my observations, or:

Click here for the full results!

Shifting priorities

Here are the results of the thematic factors in 2017, compared to 2015:

Thematic Factors

2015

2017

Change

GMB Signals

21.63%

19.01%

-12.11%

Link Signals

14.83%

17.31%

+16.73%

On-Page Signals

14.23%

13.81%

-2.95%

Citation Signals

17.14%

13.31%

-22.36%

Review Signals

10.80%

13.13%

+21.53%

Behavioral Signals

8.60%

10.17%

+18.22%

Personalization

8.21%

9.76%

+18.81%

Social Signals

4.58%

3.53%

-22.89%

If you look at the Change column, you might get the impression that there were some major shifts in priorities this year, but the Change number doesn’t tell the whole story. Social factors may have seen the biggest drop with a -22.89% change, but a shift in emphasis on social factors from 4.58% to 3.53% isn’t particularly noteworthy.

The decreased emphasis on citations compared to the increased emphasis on link and review factors, is reflective of shifting focus, but as I’ll discuss below, citations are still crucial to laying down a proper foundation in local search. We’re just getting smarter about how far you need to go with them.

The importance of proximity

For the past two years, Physical Address in City of Search has been the #1 local pack/finder ranking factor. This makes sense. It’s tough to rank in the local pack of a city that you’re not physically located in.

Well, as of this year’s survey, the new #1 factor is… drumroll please…

Proximity of Address to the Point of Search

This factor has been climbing from position #8 in 2014, to position #4 in 2015, to claim the #1 spot in 2017. I’ve been seeing this factor’s increased importance for at least the past year, and clearly others have noticed as well. As I note in my recent post on proximity, this leads to poor results in most categories. I’m looking for the best lawyer in town, not the closest one. Hopefully we see the dial get turned down on this in the near future.

While Proximity of Address to the Point of Search is playing a stronger role than ever in the rankings, it’s certainly not the only factor impacting rankings. Businesses with higher relevancy and prominence will rank in a wider radius around their business and take a larger percentage of the local search pie. There’s still plenty to be gained from investing in local search strategies.

Here’s how the proximity factors changed from 2015 to 2017:

Proximity Factors

2015

2017

Change

Proximity of Address to the Point of Search

#4

#1

+3

Proximity of Address to Centroid of Other Businesses in Industry

#20

#30

-10

Proximity of Address to Centroid

#16

#50

-34

While we can see that Proximity to the Point of Search has seen a significant boost to become the new #1 factor, the other proximity factors which we once thought were extremely important have seen a major drop.

I’d caution people against ignoring Proximity of Address to Centroid, though. There is a situation where I think it still plays a role in local rankings. When you’re searching from outside of a city for a key phrase that contains the city name (Ex: Denver plumbers), then I believe Google geo-locates the search to the centroid and Proximity of Address to Centroid impacts rankings. This is important for business categories that are trying to attract searchers from outside of their city, such as attractions and hotels.

Local SEOs love links

Looking through the results and the comments, a clear theme emerges: Local SEOs are all about the links these days.

In this year’s survey results, we’re seeing significant increases for link-related factors across the board:

Local Pack/Finder Link Factors

2015

2017

Change

Quality/Authority of Inbound Links to Domain

#12

#4

+8

Domain Authority of Website

#6

#6

-

Diversity of Inbound Links to Domain

#27

#16

+11

Quality/Authority of Inbound Links to GMB Landing Page URL

#15

#11

+4

Quantity of Inbound Links to Domain

#34

#17

+17

Quantity of Inbound Links to Domain from Locally Relevant Domains

#31

#20

+11

Page Authority of GMB Landing Page URL

#24

#22

+2

Quantity of Inbound Links to Domain from Industry-Relevant Domains

#41

#28

+13

Product/Service Keywords in Anchor Text of Inbound Links to Domain

-

#33

+17

Location Keywords in Anchor Text of Inbound Links to Domain

#45

#38

+7

Diversity of Inbound Links to GMB Landing Page URL

-

#39

+11

Quantity of Inbound Links to GMB Landing Page URL from LocallyRelevant Domains

-

#48

+2

Google is still leaning heavily on links as a primary measure of a business’ authority and prominence, and the local search practitioners that invest time and resources to secure quality links for their clients are reaping the ranking rewards.

Fun fact: "links" appears 76 times in the commentary.

By comparison, "citations" were mentioned 32 times, and "reviews" were mentioned 45 times.

Shifting priorities with citations

At first glance at all the declining factors in the table below, you might think that yes, citations have declined in importance, but the situation is more nuanced than that.

Local Pack/Finder Citation Factors

2015

2017

Change

Consistency of Citations on The Primary Data Sources

n/a

#5

n/a

Quality/Authority of Structured Citations

#5

#8

-3

Consistency of Citations on Tier 1 Citation Sources

n/a

#9

n/a

Quality/Authority of Unstructured Citations (Newspaper Articles, Blog Posts, Gov Sites, Industry Associations)

#18

#21

-3

Quantity of Citations from Locally Relevant Domains

#21

#29

-8

Prominence on Key Industry-Relevant Domains

n/a

#37

n/a

Quantity of Citations from Industry-Relevant Domains

#19

#40

-21

Enhancement/Completeness of Citations

n/a

#44

n/a

Proper Category Associations on Aggregators and Tier 1 Citation Sources

n/a

#45

n/a

Quantity of Structured Citations (IYPs, Data Aggregators)

#14

#47

-33

Consistency of Structured Citations

#2

n/a

n/a

Quantity of Unstructured Citations (Newspaper Articles, Blog Posts)

#39

-

-11

You’ll notice that there are many “n/a” cells on this table. This is because I made some changes to the citation factors. I elaborate on this in the survey results, but for your quick reference here:

  1. To reflect the reality that you don’t need to clean up your citations on hundreds of sites, Consistency of Structured Citations has been broken down into 4 new factors:
    1. Consistency of Citations on The Primary Data Sources
    2. Consistency of Citations on Tier 1 Citation Sources
    3. Consistency of Citations on Tier 2 Citation Sources
    4. Consistency of Citations on Tier 3 Citation Sources
  2. I added these new citation factors:
    1. Enhancement/Completeness of Citations
    2. Presence of Business on Expert-Curated "Best of" and Similar Lists
    3. Prominence on Key Industry-Relevant Domains
    4. Proper Category Associations on Aggregators and Top Tier Citation Sources

Note that there are now more citation factors showing up, so some of the scores given to citation factors in 2015 are now being split across multiple factors in 2017:

  • In 2015, there were 7 citation factors in the top 50
  • In 2017, there are 10 citation factors in the top 50

That said, overall, I do think that the emphasis on citations has seen some decline (certainly in favor of links), and rightly so. In particular, there is an increasing focus on quality over quantity.

I was disappointed to see that Presence of Business on Expert-Curated "Best of" and Similar Lists didn’t make the top 50. I think this factor can provide a significant boost to a business’ local prominence and, in turn, their rankings. Granted, it’s a challenging factor to directly influence, but I would love to see an agency make a concerted effort to outreach to get their clients listed on these, measure the impact, and do a case study. Any takers?

GMB factors

There is no longer an editable description on your GMB listing, so any factors related to the GMB description field were removed from the survey. This is a good thing, since the field was typically poorly used, or abused, in the past. Google is on record saying that they didn’t use it for ranking, so stuffing it with keywords has always been more likely to get you penalized than to help you rank.

Here are the changes in GMB factors:

GMB Factors

2015

2017

Change

Proper GMB Category Associations

#3

#3

-

Product/Service Keyword in GMB Business Title

#7

#7

-

Location Keyword in GMB Business Title

#17

#12

+5

Verified GMB Listing

#13

#13

-

GMB Primary Category Matches a Broader Category of the Search Category (e.g. primary category=restaurant & search=pizza)

#22

#15

+7

Age of GMB Listing

#23

#25

-2

Local Area Code on GMB Listing

#33

#32

+1

Association of Photos with GMB Listing

-

#36

+14

Matching Google Account Domain to GMB Landing Page Domain

#36

-

-14

While we did see some upward movement in the Location Keyword in GMB Business Title factor, I’m shocked to see that Product/Service Keyword in GMB Business Title did not also go up this year. It is hands-down one of the strongest factors in local pack/finder rankings. Maybe THE strongest, after Proximity of Address to the Point of Search. It seems to me that everyone and their dog is complaining about how effective this is for spammers.

Be warned: if you decide to stuff your business title with keywords, international spam hunter Joy Hawkins will probably hunt your listing down and get you penalized. :)

Also, remember what happened back when everyone was spamming links with private blog networks, and then got slapped by the Penguin Update? Google has a complete history of changes to your GMB listing, and they could decide at any time to roll out an update that will retroactively penalize your listing. Is it really worth the risk?

Age of GMB Listing might have dropped two spots, but it was ranked extremely high by Joy Hawkins and Colan Neilsen. They’re both top contributors at the Google My Business forum, and I’m not saying they know something we don’t know, but uh, maybe they know something we don’t know.

Association of Photos with GMB Listing is a factor that I’ve heard some chatter about lately. It didn’t make the top 50 in 2015, but now it’s coming in at #36. Apparently, some Google support people have said it can help your rankings. I suppose it makes sense as a quality consideration. Listings with photos might indicate a more engaged business owner. I wonder if it matters whether the photos are uploaded by the business owner, or if it’s a steady stream of incoming photo uploads from the general public to the listing. I can imagine that a business that’s regularly getting photo uploads from users might be a signal of a popular and important business.

While this factor came in as somewhat benign in the Negative Factors section (#26), No Hours of Operation on GMB Listing might be something to pay attention to, as well. Nick Neels noted in the comments:

Our data showed listings that were incomplete and missing hours of operation were highly likely to be filtered out of the results and lose visibility. As a result, we worked with our clients to gather hours for any listings missing them. Once the hours of operation were uploaded, the listings no longer were filtered.

Behavioral factors

Here are the numbers:

GMB Factors

2015

2017

Change

Clicks to Call Business

#38

#35

+3

Driving Directions to Business Clicks

#29

#43

-14

Not very exciting, but these numbers do NOT reflect the serious impact that behavioral factors are having on local search rankings and the increased impact they will have in the future. In fact, we’re never going to get numbers that truly reflect the value of behavioral factors, because many of the factors that Google has access to are inaccessible and unmeasurable by SEOs. The best place to get a sense of the impact of these factors is in the comments. When asked about what he’s seeing driving rankings this year, Phil Rozek notes:

There seem to be more “black box” ranking scenarios, which to me suggests that behavioral factors have grown in importance. What terms do people type in before clicking on you? Where do those people search from? How many customers click on you rather than on the competitor one spot above you? If Google moves you up or down in the rankings, will many people still click? I think we’re somewhere past the beginning of the era of mushy ranking factors.

Mike Blumenthal also talks about behavioral factors in his comments:

Google is in a transition period from a web-based linking approach to a knowledge graph semantic approach. As we move towards a mobile-first index, the lack of linking as a common mobile practice, voice search, and single-response answers, Google needs to and has been developing ranking factors that are not link-dependent. Content, actual in-store visitations, on-page verifiable truth, third-party validation, and news-worthiness are all becoming increasingly important.
But Google never throws anything away. Citations and links as we have known them will continue to play a part in the ranking algo, but they will be less and less important as Google increases their understanding of entity prominence and the real world.

And David Mihm says:

It’s a very difficult concept to survey about, but the overriding ranking factor in local — across both pack and organic results — is entity authority. Ask yourself, “If I were Google, how would I define a local entity, and once I did, how would I rank it relative to others?” and you’ll have the underlying algorithmic logic for at least the next decade.
  • How widely known is the entity? Especially locally, but oh man, if it’s nationally known, searchers should REALLY know about it.
  • What are people saying about the entity? (It should probably rank for similar phrases)
  • What is the engagement with the entity? Do people recognize it when they see it in search results? How many Gmail users read its newsletter? How many call or visit it after seeing it in search results? How many visit its location?

David touches on this topic in the survey response above, and then goes full BEAST MODE on the future of local rankings in his must-read post on Tidings, The Difference-Making Local Ranking Factor of 2020. (David, thank you for letting me do the Local Search Ranking Factors, but please, don’t ever leave us.)

The thing is, Google has access to so much additional data now through Chrome, Android, Maps, Ads, and Search. They’d be crazy to not use this data to help them understand which businesses are favored by real, live humans, and then rank those businesses accordingly. You can’t game this stuff, folks. In the future, my ranking advice might just be: "Be an awesome business that people like and that people interact with." Fortunately, David thinks we have until 2020 before this really sets in, so we have a few years left of keyword-stuffing business titles and building anchor text-optimized links. Phew.

To survey or to study? That is not the question

I’m a fan of Andrew Shotland’s and Dan Leibson’s Local SEO Ranking Factors Study. I think that the yearly Local Search Ranking Factors Survey and the yearly (hopefully) Local SEO Ranking Factors Study nicely complement each other. It’s great to see some hard data on what factors correlate with rankings. It confirms a lot of what the contributors to this survey are intuitively seeing impact rankings for their clients.

There are some factors that you just can’t get data for, though, and the number of these “black box” factors will continue to grow over the coming years. Factors such as:

  • Behavioral factors and entity authority, as described above. I don’t think Google is going to give SEOs this data anytime soon.
  • Relevancy. It’s tough to measure a general relevancy score for a business from all the different sources Google could be pulling this data from.
  • Even citation consistency is hard to measure. You can get a general sense of this from tools like Moz Local or Yext, but there is no single citation consistency metric you can use to score businesses by. The ecosystem is too large, too complicated, and too nuanced to get a value for consistency across all the location data that Google has access to.

The survey, on the other hand, aggregates opinions from the people that are practicing and studying local search day in and day out. They do work for clients, test things, and can see what had a positive impact on rankings and what didn’t. They can see that when they built out all of the service pages for a local home renovations company, their rankings across the board went up through increased relevancy for those terms. You can’t analyze these kinds of impacts with a quantitative study like the Local SEO Ranking Factors Study. It takes some amount of intuition and insight, and while the survey approach certainly has its flaws, it does a good job of surfacing those insights.

Going forward, I think there is great value in both the survey to get the general sense of what’s impacting rankings, and the study to back up any of our theories with data — or to potentially refute them, as they may have done with city names in webpage title tags. Andrew and Dan’s empirical study gives us more clues than we had before, so I’m looking forward to seeing what other data sources they can pull in for future editions.

Possum’s impact has been negligible

Other than Proper GMB Category Associations, which is definitely seeing a boost because of Possum, you can look at the results in this section more from the perspective of "this is what people are focusing on more IN GENERAL." Possum hasn’t made much of an impact on what we do to rank businesses in local. It has simply added another point of failure in cases where a business gets filtered.

One question that’s still outstanding in my mind is: what do you do if you are filtered? Why is one business filtered and not the other? Can you do some work to make your business rank and demote the competitor to the filter? Is it more links? More relevancy? Hopefully someone puts out some case studies soon on how to defeat the dreaded Possum filter (paging Joy Hawkins).


Focusing on More Since Possum

#1

Proximity of Address to the Point of Search

#2

Proper GMB Category Associations

#3

Quality/Authority of Inbound Links to Domain

#4

Quantity of Inbound Links to Domain from Locally Relevant Domains

#5

Click-Through Rate from Search Results


Focusing on Less Since Possum

#1

Proximity of Address to Centroid

#2

Physical Address in City of Search

#3

Proximity of Address to Centroid of Other Businesses in Industry

#4

Quantity of Structured Citations (IYPs, Data Aggregators)

#5

Consistency of Citations on Tier 3 Citation Sources


Foundational factors vs. competitive difference-makers

There are many factors in this survey that I’d consider table stakes. To get a seat at the rankings table, you must at least have these factors in order. Then there are the factors which I’d consider competitive difference-makers. These are the factors that, once you have a seat at the table, will move your rankings beyond your competitors. It’s important to note that you need BOTH. You probably won’t rank with only the foundation unless you’re in an extremely low-competition market, and you definitely won’t rank if you’re missing that foundation, no matter how many links you have.

This year I added a section to try to get a sense of what the local search experts consider foundational factors and what they consider to be competitive difference-makers. Here are the top 5 in these two categories:


Foundational

Competitive Difference Makers

#1

Proper GMB Category Associations

Quality/Authority of Inbound Links to Domain

#2

Consistency of Citations on the Primary Data Sources

Quantity of Inbound Links to Domain from Industry-Relevant Domains

#3

Physical Address in City of Search

Quality/Authority of Inbound Links to GMB Landing Page URL

#4

Proximity of Address to the Point of Search (Searcher-Business Distance)

Quantity of Inbound Links to Domain from Locally Relevant Domains

#5

Consistency of Citations on Tier 1 Citation Sources

Quantity of Native Google Reviews (with text)

I love how you can look at just these 10 factors and pretty much extract the basics of how to rank in local:

“You need to have a physical location in the city you’re trying to rank in, and it’s helpful for it to be close to the searcher. Then, make sure to have the proper categories associated with your listing, and get your citations built out and consistent on the most important sites. Now, to really move the needle, focus on getting links and reviews.”

This is the much over-simplified version, of course, so I suggest you dive into the full survey results for all the juicy details. The amount of commentary from participants is double what it was in 2015, and it’s jam-packed with nuggets of wisdom. Well worth your time.

Got your coffee? Ready to dive in?

Take a look at the full results


Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don't have time to hunt down but want to read!

No comments:

Post a Comment