What We Learned About Google’s Local Results From Analyzing 119,221 URLs

You probably already know that Google’s search results are highly personalized and based on your geographical location.

If you’re in Tucson and you search for “fish tacos,” you’ll get different results than if you were in Tampa and typed the same query.

Even if you’re not logged into Google, you’ll get different results, based on where you are in the world, in your country, in your state and even in your city!

Why? How? What ranking signals influence local SEO, and why do results differ so widely?

If we can figure this out, then we would know how to dominate local SEO.

I did exactly that. I set out to discover how Google’s local algorithm works. To do so, I turned to the only answer source I know:  Data.

What you’re reading is one of the first ever massive data research projects that seeks to uncover how Google’s local algorithm works and what it takes to gain rank in local SEO.

What factors influence local SEO?

Based on previous research, we already knew that the following factors were important in local SEO:

• Backlink profile
• Domain age
• Website’s technical SEO
• Name, address, and profile (NAP) on the website
• Depth and comprehensiveness of content
• Number and quality of citations
• Quality and quantity of user clicks on a local pack result
• Google My Business verification, reviews and comprehensive profile
• Distance of the business from the user’s search

But, how important are these factors relative to each other? And, what about ranking universally vs. ranking locally? Do certain techniques or ranking factors apply to one and not to the other?

Our goal in this data-driven study was to go beyond the basic local SEO assumptions and find out some of the more advanced aspects of Google’s local results.

For example, we wanted to find out how a small local business could outrank a large national chain in the local pack, how domain names influenced local pack search results and why universal search results differed based on query location.

The Data We Used

The data we pulled was straightforward, but crunching the numbers was incredibly complex.

In summary, here is the data we used:

• We used Google’s category names as our keyword list. Using this list would ensure that we got local packs in the SERPs.
• We queried from five IPs. Four were geographically specific (Chicago, Houston, Los Angeles, and New York City). The fifth was from an IP that Google could only identify as USA, in order to prevent the SERP from producing local or map results.
• Using SEMRush’s Position Tracker, we analyzed the top 100 results from each of the categories/queries in each of the five locations.
• We further analyzed each of these results to measure domain-specific and page-specific SEO factors, such as page speed, size, keyword presence, domain, authority, etc.

In total, we analyzed 119,221 URLs/SERP results for this study.

The local packs that we analyzed all contained maps, like this one:

But, we also looked at universal results, the “non-map” kind that you see in the SERPs from a typical informational query:

It’s important to note that there are two types of local results — the pack and the one-box. Google produces one or the other in response to a query.

Most of the time, you’ll see the pack, which is a listing of three local businesses.

If your query is specific enough, you will see a one-box, like this:

What we learned

We went into the study with an open mind. Our goal wasn’t to prove anything, but rather to observe how the algorithm responded, based on different locations and queries.

Major Finding #1:  Domains that contain location names tend to rank higher in the area where the search takes place.

You probably already know that Google changes SERP results based on where a user is located. What we discovered in the data, however, is that location names in a URL are highly significant in improving a website’s likelihood of appearing in the SERPs.

If a user is located in Chicago, for example, then Google favors results that contain part of the city name in the URL.

On one hand, this makes sense. Someone who is in Chicago is probably more interested in the website “ChicagoNow.com” than someone in Houston. As a result, the algorithm gives the Chicago-based user Chicago results.

The algorithm is all about providing users with relevant results, based on their geographical location and intent.

But, other results seem arbitrary. For example, why is PsychicScience.org more likely to appear in a Chicago-based SERP than in a Houston, L.A., or NYC-based SERP? (See the chart below.)

Look at the domain name again: PsychicScience.org. Do you see how the domain name contains a fragment of the word “Chicago”? Even though “psychic” is unrelated to “Chicago,” the domain contains part of the word Chicago. The algorithm interprets this as a partial city name, so Chicago searchers have more than the usual number of psychic readers in their universal SERPs.

Note: Psychicmediummichael.com and psychicscience.org do not appear to be Chicago-based businesses.

Major Finding #2: High-ranked universal results generally do not have local pack listings.

Many of the websites with high universal ranking do not have a placement in the local pack. Why not? Simple. These websites do not contain local addresses and/or geographical names in the URL.

This major finding has two takeaways:

1. If you want to rank in the local pack, your primary local SEO tactic will be your NAP (Name, Address, Phone) and citations.
2. If you want to rank universally (non local pack) for geographically specific keywords, then create lots of deep, comprehensive content on local subject matter.

Take, for example, a nightlife business that wants to rank in the local pack, but isn’t concerned about universal ranking. If I search for “nightlife” (incognito, but with a locally specific IP) the results in image below will appear in the 3 pack.

For Full House Sportzaria, this is good news. They’re at the top of the local pack. But, do they have a lot of deep, comprehensive and long form content? Nah, but it’s no big deal. Why not? Because of their NAP and citations. That’s enough to put them at the top of the local pack.

If your goal is to bring people to your brick-and-mortar business, then you should use your NAP on your website, build local citations and concentrate on getting high reviews.

Heat map studies demonstrate that top-ranked local results get more clicks. Here’s a heat map overlay of a law firm search. Notice how the majority of the clicks are for the top-ranked local result in the 3-pack.

If the SERP contains a local pack, then the local pack receives the majority of the clicks. Notice in the chart below how the local 3-pack gains 44% of the total clicks.

As you build local awareness around your brand, you may be able to gain a one-pack placement. As users hear more about your business, they may remember the name of the business and query it. The result will most likely be a one-pack.

The query “sportzaria” is specific and geographically identifiable, based on IP. Thus, this business gets the coveted one-pack sidebar and a full SERP of organic universal results when users search for it.

What about a business that wants universal rankings, but not a local pack placement? They don’t need to be concerned about NAP and citations. In fact, if they do work on building NAP citations, then they might lose universal result rankings to a local pack placement.

NYpost.com has high universal visibility for locally specific queries. Getting a placement in a local pack won’t help them, because their goal is web readership, not brick-and-mortar visits.

For my incognito query “new york news,” nypost.com had two organic universal results.

What does NYPost.com do in order to gain such high universal rankings in the SERPs? The answer: Content. It’s all about the deep, comprehensive content.

Colorado.gov is another good example of a site that gains high universal ranking for geographically specific terms.  They rank high in the universal results for Colorado-specific queries, even when the searches appear far from the state of Colorado.

The reason for their excellent results are obvious when you take a look at their website.  It contains useful information, thousands of pages and comprehensive content.

Major Finding #3:  Small local businesses can outrank large national brands in local pack placement.

For a long time, SEO favored the big boys — large, well-known brands with massive websites and strong link profiles.

In local SEO, the tables are turned. In fact, a small business doesn’t even need a website to get a placement in the local pack. Ranking signals, such as link profile and content quantity, have secondary significance for local pack placement.

Any small business can set up up their Google My Business, Facebook, Yelp and TripAdvisor profiles. Larger businesses with hundreds or thousands of locations have less agility to curate citations and build personal customer connections that lead to favorable local reviews.

Notice how an incognito query for “steak restaurant” brings up primarily local establishments in the local pack, rather than major chains.

Even though big chains, such as Ruth’s Chris, Logan’s Roadhouse, LongHorn Steakhouse, Outback Steakhouse and Texas Roadhouse all have one or more restaurants in this geographical IP, they do not appear in the local pack. Large businesses have an advantage in the universal results, not the local pack.

Here are some more advantages that help a small local business gain better local-pack positioning.

1. It’s easier for a local business to connect with customers and ask for reviews.
2. Local businesses are more likely to post pictures and videos.
3. Large chains may end up having multiple listings that create confusion.
4. It’s harder for large chains to address feedback in Google listings.
5. Small businesses are more likely to build strong citations through community mentions and involvement in local events.

Conclusion

In summary, we uncovered three major findings from this study:

• Major Finding #1:  Domains that contain location names tend to rank higher in the area where the search takes place. Choose a domain name that contains the name of the local area in which you want to rank. For example, if you are a Houston-based attorney, you may want to choose a domain name such as houstonattorney.com. If you already have a website without your city’s name, create a new website that has the city name in the domain and 301-redirect it to the existing website.
• Major Finding #2: High-ranked universal results generally do not have local pack listings. If you have a local business and you want to rank in the local pack, focus your efforts on NAP and citation building. Content is important, sure, but not as significant of a ranking factor for a local pack placement. If you have to choose between creating content for your website and gaining more citations, choose the citations.
• Major Finding #3:  Small local businesses can outrank large national brands in local pack placement. If you are a local business and you want to rank in the local pack, maintain an active presence on your local pages and keep a high level of engagement in the community. Participate in fairs, fundraisers, events and other local activities that will improve your chance of being mentioned and noticed.

As far as local SEO goes, there has never been a better time to be a local business. Getting a placement in the local pack takes effort, yes, but it’s not as time-consuming or expensive as creating a large content presence or building a big link profile.

What observations have you made regarding Google’s local results, and how does this influence your local SEO efforts?