Local SERP tracking has become an SEO necessity. And yet, it’s anything but simple. Here, we look at seven local tracking strategies and explore why they are so critical for understanding how people search.

Pick a search location, then start tracking. Couldn’t be easier, right?

Well, if you already have comprehensive local SERP tracking in place, you know that things go much deeper than that. When you consider the interplay between location services, device types, and searcher intent, and then factor in all of the various ways that searchers move through the world, the complexity soon reveals itself.

In this guide, we’ll take a look at how local SERP tracking has become absolutely critical to competitive SEO. Then we’ll look at the seven key tracking strategies that allow SEOs to understand searchers and address the unique needs of different industries or verticals. Finally, we’ll show you how to implement these strategies within STAT.

1. Why everyone should track local
2. How are mobile and local connected?
3. Geo-location and geo-modification
4. Seven ways to do local search
   1. The armchair researcher
   2. The mobile time-killer
   3. The “I want it now”
   4. The long-distance dreamer
   5. The metropolitan
   6. The international
   7. The market baseline
5. How locations work in STAT

 

A. Why everyone should track local

In some industries, the importance of local SEO and local SERP tracking is immediately obvious. Typically, these are products or services that are inherently tied to location: things like brick-and-mortar retail, real estate, professional services, automobiles and heavy equipment, restaurants and hospitality, and so on. In these industries, if you are performing poorly on the relevant local SERPs, you are effectively invisible to local customers online.

But what about businesses that are not as closely tied to brick-and-mortar locations and local customers? The reality is that the majority of searches are now localized to varying degrees. Google and every other major search engine routinely modify desktop and mobile search results based on location — even when the searcher does not explicitly include a location in their search query.

“4 in 5 consumers use search engines to find local information.”
— Think with Google, May 2014

“74% of adult smartphone owners ages 18 and older say they use their phone to get directions or other information based on their current location.”
— PewRearchCenter, September 2013

“96% of desktop owners, 79% of smartphone owners and 81% of tablet owners use their device to search locally.”
— ComScore, April 2014

What this means, very simply, is that local search results now matter to every business. Even if you ship globally, only sell digital goods, or are trying to attract new app users, poor performance in critical, high-volume local SERPs will still result in lost traffic and lost conversions.

Luckily, the flipside of this big pitfall is a big opportunity. Most industries have leading brands that successfully dominate national or regional SERPs, making it seem impossible to break through. But when you shift your focus to localized SERP tracking and analytics, those same brands often show surprising weaknesses. We saw this effect clearly in our GEICO case study, which revealed a number of high-value local markets where the US insurance giant GEICO was underperforming.

For up-and-coming brands, these local weaknesses can be strategically exploited. For big, incumbent brands, knowing your own local weaknesses can be an opportunity to further cement your market-leading position.

 

B. How are mobile and local connected?

There isn’t an SEO professional left on the planet who really needs to be convinced about the importance of mobile search. And since mobile SERPs are very distinct from desktop SERPs — from UX to organic rankings to SERP feature types — the reality is that most online industries have to track and analyze both SERP types to stay competitive. But mobile SERPs do require one special consideration: the majority of mobile searches are also highly localized.

There are two key reasons why this is the case. First, location services are standard on the most recent generations of mobile devices. As a result, even if there is no explicit local intent on behalf of a mobile searcher, they will still end up seeing localized search results more often than not.

“Nearly 3 out of 4 mobile phone searches that end in a purchase bring customers into brick-and-mortar stores.”
— ComScore, April 2014

“81% of mobile searches are driven by speed and convenience.”
— Think with Google, March 2013

“50% of consumers who conducted a local search on their smartphone visited a store within a day, and 34% who searched on computer/tablet did the same.”
— Think with Google, May 2014

Second, mobile searchers themselves are different. Or, more accurately, search behaviours and usage patterns are different when people use smartphones or tablets versus laptops or desktop computers. Mobile searchers are less likely to be at home, more likely to be on-the-go, and are more likely to be seeking information with immediate local intent. Naturally, that includes the intent to make a purchase.

Yes, it’s true that mobile search is more important in some industries than in others, especially in the case of spontaneous, smaller-ticket purchases made exclusively at brick-and-mortar locations. (Consumer packaged goods and fast food restaurants would both fall into this category, for example.)

However, much like in the case of localized desktop searches, if your site is performing poorly for localized mobile searches, you will miss out on traffic from the SERPs even if your business is not tied to brick-and-mortar locations. That goes for everything from insurance, to software, to media, to online shopping.

 

C. Geo-location and geo-modification

For every individual search, there are two distinct factors that determine whether and how the search results are localized.

• Geo-modification is when the searcher manually includes geographical terms in the search query itself — for example, in the search [best beaches in NSW Australia]. (Google calls this “explicit location.”)
• Geo-location is when the searcher’s device automatically provides location data as a part of the search query — for example, in the search [best beaches] when performed within Sydney on a smartphone. (Google calls this “user location.”)

Geo-location is essentially synonymous with the familiar notion of “location services.” The search provider leverages data supplied by the device — including data related to GPS, cell towers, Wi-Fi nodes, and IP address — to help serve up relevant, local information without the user having to manually modify their search.

Just about anyone who uses a smartphone does these kinds of searches naturally. Say you want to find a nearby gas station. You simply search for [gas station], and trust that Google knows where you are thanks to your device’s location services. Remember, however, that other devices like tablets, laptops, and desktop computers are increasingly equipped with some flavour of location services as well.

Because geo-location happens automatically when location services are active, with no direct intervention by the user, it is not by itself a strong signal of a searcher’s local intent. Geo-modification, on the other hand, is usually a very strong signal of a searcher’s local intent. (Google sometimes refers to this somewhat awkwardly as “visit-in-person intent.”)

Think about when you’d use geo-modification for your own searches. Maybe you’re on a work computer that doesn’t have any location services, but you need the business hours of a restaurant nearby. Or, maybe you’re on a mobile device that does have location services, but you want to find info about another part of the city. In both cases, the deliberate modification of the search query gives us clear insight into what the searcher really wants.

Geo-modification is especially prevalent in two common search scenarios:

• Searching at a distance
  First, geo-modification is common whenever there is a mismatch between where the searcher is currently located and the place that they are searching about. This would be a characteristic search pattern for commuters, tourists, and trip-planners — people who are heading somewhere else.
• Searching for hyper-local results
  Second, in dense metropolitan areas, searchers will attempt to narrow local results using hyper-local search terms like the names of streets, avenues, districts, or boroughs.

In both of these scenarios, the searcher knows better than their device about the kind of local results that they want.

Of course, it is also very common for geo-modification and geo-location to work together on the same search. For example, a woman in Vancouver searches for [used cars Main St] on a mobile phone. There are countless “Main Streets” in the world, but the device geo-locates the search in Vancouver. Furthermore, Vancouver has many used car dealerships, but the searcher has intentionally geo-modified the search to get results around Main Street. Together, the two factors give the searcher a better chance of finding exactly what she’s looking for the first time around.

TL;DR

Geo-location	Geo-modification
Location provided by the device	Location specified by the searcher
More common on mobile (but also on desktop)	Common on desktop and mobile
Weak signal of searcher intent	Strong signal of searcher intent
For local searches only	For local searches or distance searches
Also called “user location”	Also called “explicit location”

 

D. Seven ways to do local search

There are a zillion different ways to search for local info, especially when you start looking at the interplay between device types, location services, and searcher intent. If you’re trying to track local SERPs, all of this variety can be intimidating.

Happily, there are common patterns to how people do local searches. Here, we take a close look at seven of the most important. We should be clear that we’re talking about both search patterns and SERP-tracking strategies in the same breath here. Searchers use certain tactics to find the info they want, and your SERP tracking tool needs to match those usage patterns in order to see exactly what searchers are seeing. Understanding how and why someone conducts a local search is the first step to building out a comprehensive SERP-tracking plan tailored specifically to your industry or vertical.

Each of the seven patterns covered here is differentiated by six factors:

Geo-location
Does this search pattern use geo-location data? The answer can be yes, no, or maybe. (Maybe indicates that geo-location data may be used, but it’s not a critical factor in this pattern.)

Geo-modification
Is the searcher using explicit location terms to geo-modify their query? The answer can be yes, no, or maybe. (Maybe indicates that geo-modification may be used, but it’s not a critical factor in this pattern.)

Device
Is this searcher on a mobile or desktop device? (In some cases, they could be on mobile or desktop — it doesn’t matter.)

Distance
Is the searcher looking for results about a place that is not the same as their current location? In other words, are they looking for something “over there” rather than “here”? The answer can be yes, no, or maybe.

Visiting
Is the searcher physically visiting a market that does not match their typical device preferences, language settings, or default regional search engine? (In other words, have they brought a device with them to a new market?)

Intent
How strong is the local intent signal? (In other words, how strongly is the search telling us that the searcher intends to buy or do something locally?)

 

D1. The armchair researcher

You’re at home on your laptop looking to buy a high-end road bike — checking reviews, comparing prices, and looking at inventory in your local area.

This is a typical search pattern for desktop users who don’t have location services but who are looking for local results, making them more likely to include a location term in their search query. (This pattern also captures other devices without location services.)

As a tracking strategy, this pattern is important for many industries or verticals in which desktop users are important. It’s especially true for business, professional, and personal services, as well as bigger-ticket items like real estate, automobiles, appliances, entertainment, restaurants, and so on — basically, anything that people research, compare, and make a point of seeking out rather than buy spontaneously.

Example use-case:

Your client is a high-end bicycle manufacturer who sells their bikes and accessories in corporate stores as well as through select local retailers across the country. Their buyers tend to be savvy, well informed, and prone to doing oodles of research before making purchasing decisions. In this category, buyers prefer to examine, touch, and test products in person rather than buying online.

You build out extensive keyword sets that cover everything from early research to price comparisons to post-purchase and maintenance. Then you track those sets for desktop devices in national markets where the brand is currently represented, plus a few other markets where they plan on expanding in the future. For example [Manchester custom bike fitting] tracked in GB-en, or [Shimano parts Auckland] tracked in NZ-en.

 

D2. The mobile time-killer

You have some downtime during your lunchbreak, so you’re on your phone doing a bit of aimless Googling for home improvement ideas.

This is a common default pattern for mobile users, who are less likely than desktop users to provide an explicit location in their query.

Because the device is doing the geo-location without the user’s explicit intervention, the intent signal in these types of searches is low. These searchers may or may not have any real, imminent plan to buy locally. We just can’t say for sure.

As a tracking strategy, this pattern is relevant to verticals where mobile users are significant. (Of course, that means just about every industry!) However, it’s particularly important for verticals that are only loosely tied to location — which can be anything from insurance to online marketplaces to social media to digital goods — because map results and other localized content can easily crowd out relevant non-local content on these kinds of SERPs.

Note that this pattern also captures desktop users who do have location services, but who aren’t sending out a clear intent signal because they are not geo-modifying their search.

Example use-case:

Your company operates a dozen English-language media properties, with a heavy focus on lifestyle, fitness, beauty, gardening, and home improvement. One critical challenge that you are facing on mobile SERPs is competition from local retail when people do general searches on home, garden, and beauty subjects while their location services are turned on.

To combat this, you put together an extensive list of relevant keywords that have no clear local intent, signalled by a lack of geo-modification. (For example, [affordable window treatments] or [best new hair products 2015].) In each of your national markets, you track your keyword list in all cities with a population over one million, representing your most critical urban markets.

With the SERP data in hand, you’re able to identify the highest-value keywords where map results and other local retailer organic results are crowding out your media properties. Equipped with that analysis, you can now start strategically building out locally relevant content designed to improve your visibility on these searches.

 

D3. The “I want it now”

You’ve broken your tooth on an olive pit late one night, and you need to find a trustworthy emergency dentist right away.

In contrast to “the mobile time-killer” above, searchers using this pattern are providing a clear signal of some kind of local intent by explicitly providing location terms in their search. These people want to buy, see, visit, or do something in the nearby area — maybe not right this second, maybe not today, but soon enough.

This tracking strategy is essential for verticals that rely heavily on local business, including brick-and-mortar retail, restaurants, entertainment, and personal services.

However, it’s good to keep in mind that these are not necessarily spontaneous purchases in all cases. These searchers could also be looking for big-ticket items like cars, real estate, or heavy equipment, or they could be looking for healthcare, insurance brokers, legal services, and so on.

Example use-case:

Your client is a large network of medical information sites that also offers real reviews and ratings of medical professionals all over the globe, in several languages. Your research shows that, even on mobile devices, people tend to search for medical professionals using geo-modifying terms to limit the results to locally relevant material.

You start with an extensive, multi-lingual list of core medical queries (e.g. [COPD treatments] or [inflamação das articulações]), a second list of place names corresponding to all of your major urban target markets (e.g. [London Ontario] or [Lisboa Baixa]), and a third list of doctor- and review-related terms (e.g. [best specialists] or [comentários de reumatologistas]).

Using a combinatorial approach, you spin out a massive list of potential search terms and start tracking them in all of the relevant markets and locations. After a week or two running this massive list, you start trimming back by eliminating all of the terms that show zero search volume. You now have an exhaustive baseline on which to build all of your local SEO experiments.

 

D4. The long-distance dreamer

You’ve been thinking about that vacation to Indonesia. Maybe it’s finally time to start planning it for real.

This is a common search pattern for people who are looking for information about a place or within a place that is different from their current location.

Think of industries where there is a mismatch between the searcher’s location and the target location. That can include travel, hospitality, tourism, shipping, logistics, media, news, and so on. In these cases, you could expect the search terms to remain more or less the same regardless of whether the user is on a desktop or a mobile device.

Note that this isn’t just about big-ticket global travel. It also covers everyday movement like a daytrip to a neighbouring town. In fact, the closer the place is to the searcher’s physical location, the stronger the intent signal becomes. Why? Because people may daydream about a faraway place for many years without taking action, but they are much less likely to do the same for a nearby city.

Example use-case:

You work for a global travel company that builds local interest content and monitors its visibility in markets all over the globe. People who are looking for this type of information use geo-modification terms that reflect the place that the content is about, and not the physical location of the searcher.

For example, the content team recently created a suite of editorial pieces in French, English, and German all about shopping and night markets in Jakarta. You want to monitor the performance of this content wherever your target travellers live, so you spin out a list of related search queries and start tracking them in all of the markets that have shown high search volume in those three languages (e.g. GB-en, US-en, AU-en, CA-en, CA-fr, DE-de, AT-de, CH-de, CH-fr).

Of course, this is hardly the only content that you are monitoring, so you group these keywords into a number of different custom segments alongside similar content (e.g. segments for “destination: Indonesia” and “shopping queries: Asia” and “language: German”). This makes it easier to understand broader trends affecting your SERP visibility.

 

D5. The metropolitan

You’d prefer not to walk more than five blocks if possible, so you’re always very specific with your Google searches.

The key feature of this search pattern is geo-modification using hyper-local terms like neighbourhoods, boroughs, intersections, and addresses.

Because of the high degree of specificity, this search pattern is a very strong signal of local intent. Depending on the vertical, it can also be a strong signal of immediacy — a sign that somebody is looking to do something right now. This is particularly the case on mobile devices: these searchers are looking for convenience and they’re ready to convert.

As a tracking strategy, this is super important for industries or verticals that focus on large metropolitan markets and are highly location-dependent, including spontaneous or small-ticket purchases, urgent or convenience services, and events.

Note that this search pattern can also involve short-distance searches, such as in the case of commuters who are actively searching as they move through a large city.

Example use-case:

You represent a global coffee chain that typically has dozens of locations in large metropolitan markets. When searching for coffee shops in branded and non-branded searches, many of your urban customers will include a specific part of the city, a street name, or an intersection in their search to limit the results for convenience.

You turn to publicly available resources to put together an exhaustive list of neighbourhoods, street names, and intersections that are relevant to your existing locations. You combine these with a list of 5,000 core search terms related to coffee, tea, baking, convenience food, and so on. (For example, [flavoured syrups for coffee] or [who makes the best London Fog].)

This equips you with 100,000–300,000 search terms per metropolitan market. After tracking these for a day or two, you can immediately start culling your list based on actual search volume numbers, leaving you with a more manageable list of keywords that people are actually searching.

 

D6. The international

You’re on a business trip, equipped with your laptop and phone. Time to start investigating the local cuisine.

This pattern is a hallmark of international travellers, and the standout feature is a mismatch between the search location and the device’s default location, language, or regional search engine.

Note that the intent signal is mixed here, because this pattern includes everything from low-intent informational searches [what’s in Thai food?] to high-intent searches for immediate wants [best Thai restaurant 4th ave].

Unfortunately, it can also be problematic to track, in part because Google’s treatment of these kinds of queries is unpredictable. In some cases, the searcher remains on their “home” version of Google throughout the process, while in other cases the searcher is redirected to the local version of Google but is served results only in the language that was used in the search query. Let’s call these non-native queries versus redirected queries.

These two scenarios can result in drastically different results, especially if local content is in fact available in the target language. For example, a Taiwanese visitor is looking for Hong Kong-style bakeries in Vancouver. She will find plenty of relevant, local results in Mandarin on Google.ca (a redirected query), and these SERPs will be significantly different from the SERPs for the exact same query on Google.com.tw (a non-native query).

Currently, it’s not clear if any one factor causes a query to be treated as either redirected or non-native. We informally polled a number of industry professionals and frequent travellers, and they reported seeing both types with neither type clearly dominating.

We suspect that this search experience differs based on some combination of factors:

• what device type you are using;
• whether you’re logged in to a user account;
• whether you’re searching on a browser homepage or in an address bar;
• whether you’re accepting cookies;
• whether you have location services on or off;
• whether you’re on Wi-Fi or on a mobile network.

Redirected queries are straightforward to track at scale within STAT. Tracking non-native queries at scale, on the other hand, is generally far more technically complex and may be cost-prohibitive.

Example use-case:

Your client owns a chain of beachfront restaurants throughout Mexico and Central America, aimed primarily at tourists from Japan, Europe, the US, and Canada. Most of these visitors will be using their own devices to find local cuisine. That means they’ll start the search using regional search engines from all over the world, and they’ll search in languages that may not match the local tongue.

Your client has determined that their strategic priority is travellers from the UK, US, Germany, and Japan; they are already producing content in English, German, and Japanese in addition to Spanish. You build out a large keyword list in each of the four languages, including both geo-modified and non-geo-modified terms, and then track those in all of your destination markets and locations. For example, [best beach patios Puerto Vallarta] and [gute Restaurants für Kinder] both in Google.com.mx with the location set to Puerto Vallarta, Jalisco.

While this will only capture SERPs served by redirected queries, it will still provide a relevant measurement of your client’s performance in local SERPs, especially at large scale.

 

D7. The market baseline

Maybe you don’t want local results. Or maybe you’re paranoid about Google knowing where you are. In any case, local search just isn’t something you do.

Ok, this one is a bit of a cheat, since it’s not technically a local search at all. So why is it still an essential pattern for getting a grip on local SEO?

In order to understand how you are doing in various local markets, you need to have a baseline to compare against. That’s where this pattern comes in handy. It provides cost-effective coverage of your keywords across an entire national market, which you can use as an accurate, objective performance baseline to compare the other local search patterns against.

As such, this strategy is an important one for basically any industry or vertical, as it creates a foundation for the rest of your SERP tracking. Depending on budgets, it can also be used to fill in gaps whenever it isn’t possible to track every relevant keyword for every relevant local market or search pattern.

Example use-case:

You’re part of a team that’s working on a new local SEO plan to make your software more visible to English-speaking and Spanish-speaking users. While your service does not depend on location, you’ve realized how important your performance on local SERPs can be, especially in large urban markets.

To measure the on-going success of your new plan, and to set strategic priorities as it evolves, you’ll need to establish national baselines. And hey — some people do search without geo-location or geo-modification, so let’s not miss them.

You begin by tracking your core keyword list in all national markets with significant numbers of English or Spanish speakers, as well as some major cities within those countries. Right away, you can see which cities need the most attention, since they’re performing below their national baselines.

 

E. How locations work in STAT

In STAT, we use two basic terms to describe how keywords are tracked with respect to geographical places:

• Regular keywords are tracked within a top-level market, identified only by country and language. This essentially returns the SERPs that would be seen by someone who does not have location services.
• Local keywords are tracked in a specified city, state, province, ZIP code, postal code, or other location within any top-level market. This returns the SERPs that would be seen by someone who has location services turned on.

Of course, local keywords are requisite for understanding both mobile and local search results. However, regular keywords are also indispensable, since they allow broad and cost-effective coverage, and are useful for establishing a market baseline to compare local performance.

It should be carefully noted that both types of keywords can be geo-modified. As an example [Ontario best litigators] could be tracked in the Canada-English market as a regular keyword, or as a local keyword in the Canada-English market with Toronto specified as the location.

STAT supports every market, location, and language that is supported by Google.

There are many valid ways to specify a location in STAT. As a general rule, if your location terms are understood properly in a manual search in Google and Bing, then they will also work in STAT.

Here are some ways that you can specify a location:

• Cities and towns
  As well as metropolitan regions, villages, townships, and major neighbourhoods or boroughs.
  [women’s clothes Chelsea]
  [luxury Maui hotels]
  [things to do Helsinki]
• States and provinces
  As well as territories, regions, prefectures, and in some cases counties. (If the nation uses standard abbreviations for states or provinces, then you should be able to use those in place of the full name.)
  [custom cabinets TX]
  [Queensland beaches]
  [Ontario camping]
• Streets and avenues
  Including intersections and in some cases directions or quadrants.
  [Main St. West restaurants]
  [coffee Yonge and Bloor]
  [best Times Square hotel]
• Postal codes
  i.e. ZIP code, Postcode, Eircode, CAP, CEP, PIN code, PLZ, and so on.
  [babysitters 90210]
  [post office V7N 1J9]
  [Wanzleben 39164]
• Combinations
  You can combine any of the above, especially to help reduce ambiguity with common names.
  [Springfield, MA schools]
  [High St Canterbury hospital]
  [San Francisco, Mexico beach houses]

Finally, a few quick words about languages. STAT tracks in every language that is served by Google globally, both in terms of query language and search engine UI language.

The search engine UI language is specified as a part of the market designation, like so:

• US-en = United States, English
• US-es = United States, Spanish
• PH-tg = Philippines, Tagalog

In the majority of searches, the country, search engine language, query language, and location all match in a predictable way; e.g. [pet stores downtown] tracked in US-en with Los Angeles specified as the location.

In special cases where you are tracking redirected searches, the query language will not match the designated search engine language. For example, [pet obchody v centru Los Angeles] may be tracked in US-en with the location set to Los Angeles.

This pattern is especially common in industries that deal with travellers and tourists, who search in their native languages on their own devices, but are frequently redirected to the appropriate regional search engine. In these cases, we recommend just specifying the dominant default language for a given market.

TL;DR

Regular keywords	Local keywords
Only market is specified	Market and location specified
Location services disabled or unavailable	Location services enabled
Provides broad, cost-effective coverage	Allows comparison of local performance
May be geo-modified	May be geo-modified
Mobile and desktop	Especially mobile