10 Things to Know About Location Data in 2020 – Expert Guide

Mobile location data, also referred to as geo-location data or simply geo-data, encompasses GPS (Global Positioning System) data, WPS (WiFi Positioning System) data together with cell location data and has become increasingly well established as a source of information used in many industries. Location data is not new. Since the dawn of mobile in the 1970s it has been possible to estimate the location of a mobile device based on its relationship with its cellular network. Indeed, for a long time mobile network operators have offered services to locate mobile devices on their networks for a fee – known as Location Based Services, or LBS. 

What has changed in recent years is that the accuracy and availability of location data through other means has increased as mobile handsets became more powerful and feature-rich. This in turn has led companies such as Huq to realise the value of this data for a whole range of purposes that have become steadily more present in the tools and coverage we encounter. Huq has been a pioneer in the field of location data for research and academic purposes since 2014 and key figures within the company have specialised in this field since as early as 2002. 

Table of contents:

  1. What is mobile location data?
  2. What are the different types of mobile location data?
  3. What are the sources of mobile location data?
  4. What is location data used for?
  5. How accurate is location data?
  6. How is location data used in advertising?
  7. Can I see location data held on me?
  8. Is location data and compliant with regulations like GDPR and CCPA?
  9. How will changes in Android 14 affect the location data industry?
  10. What does 2021 hold for mobile location data?

In light of that, this article uses Huq’s deep expertise in this field to provide answers to some of the most common questions relating to the nature and use of mobile location data.

1. What is mobile location data?

Mobile location data is the latitude, longitude, timestamp and an identifier of a mobile device at a particular place in time. This dataset can also be obtained in expanded form with its estimated horizontal accuracy in meters. In certain cases, mobile location data vendors go on to link this simple dataset to the physical places visited at that time. This may include geographic address information deduced from the coordinate data – for example, state, city or street. Mobile location data vendors may also infer the names and industries of the shops, stores, transport links and other environments that the mobile device was present in at the time by making use of separate point-of-interest or POI data. For this you can think of as the place or business locations and descriptions you see on Google Maps.

2. What are the different types of mobile location data?

Mobile location data can be obtained in a number of ways, with varying degrees of accuracy, availability and mobile battery usage. 

  • Cell ID

The fastest, most abundant and often least accurate type of mobile location data is known as Cell ID data. This is usually obtained from a mobile network operator, and simply reports the location of the cell tower that the device was connected to at the time.

  • Cell Triangulation or Location-based Services

The method outlined above can be improved by reverse-triangulating the location of the device within the cell (or at least three of the cell towers) based on the differences in time at which a signal broadcast by the mobile device reaches each tower, or through other means such as by measuring the signal strength.

  • WiFi

Both methods above can be replicated on a smaller scale using the known, or estimated, locations of connected or available WiFi networks. Given the abundance of WiFi routers and their closer proximity to one another, it is often possible to achieve a more accurate estimate of the mobile device location. 

  • GPS

We’re used to opening a navigation app like Google Maps and being able to spot ourselves accurately on the right side of a road. Mobile location data in practice is rarely like this. This level of accuracy requires the use of a GPS chip on the mobile device, which consumes significantly more battery than either method above. It is also slower to determine the device position, traditionally resulting in a poorer user experience. This mode is generally reserved for when this level of accuracy is required by the user, such as when using an app for directions. Further, the GPS signal is easily interrupted by physical objects such as buildings or ceilings, making it unsuitable for indoor use.

  • Beacon, or BLe

Beacons, whether powered or unpowered (acting as a ‘reflector’) distributed throughout an indoor environment can offer similar opportunities to the methods outlined above, and often on an even smaller scale than WiFi. The challenge here is that like WiFi their physical locations are not declared to any database and can only be inferred over time though numerous ‘sightings’ by passing mobile devices. The fact that Beacons are small and often portable also complicates this fact. Their small and uneven distribution makes them better suited to the unique purposes of the Beacon owner, rather than to obtain mobile device locations at any great scale.

3. What are the sources of mobile location data?

The source of mobile location data will depend on its type. Mobile network operator derived data is sourced either directly from the network itself – often through specialised professional services teams – or through their agents. This data can be expensive and potentially out of each for large parts of the potential market. WiFi, GPS (and potentially Beacon or BLe) derived data are more readily available via the mobile operating system, ie. through apps. The use of these individual methods is often not specified by the app. Rather, apps request a location update from the mobile operating system which provides the best location estimate based on the circumstances at the time. Mobile location data vendors that source this data from across many app owners will receive the data via their own software component embedded within the app (an SDK), or via file transfer (known as server-to-server, or S2S).

4. What is location data used for?

For most mobile location data vendors, the core of their business comes from the sale of the basic set of location data properties – latitude, longitude, timestamp and device – together with an ‘Advertising ID’ that enables third parties to recognise the same handset across their various software platforms and thereby pass information around. The economics of this use-case prioritises large volumes of data that is often poorly qualified in terms of its technical source and accuracy. This kind of location data has more limited value for professional researchers, academics and analysts – but is suitable for use in proximity based advertising or retargeting campaigns.

There is a significant emerging market among researchers and analysts working in industries such as Finance, Real-estate, Retail and the Public Sector for whom mobile location data is also valuable. Common use-cases for this market include:

  • Store footfall and foot traffic used to predict trading performance
  • Analysing the composition of a store’s customers by high / low value segments 
  • Estimating productivity across different types of industrial settings
  • Assessing store performance based on changing catchment areas
  • Planning for new store locations based on catchments and footfall
  • Learning how to ease pedestrian or motorised traffic flows in cities

An important distinction to make between these two broad use-cases is that the application of one concerns the habits of a mobile device owner as an individual (advertising) versus applications for commercial or civic gains permitted by using the data in a collective sense. The debate continues.

5. How accurate is location data?

There are broad differences in the accuracy of mobile location data according to its type. These can range from up to 20km down to 5m. To help illustrate these differences, the chart below shows the distribution of location accuracy according to the many billions of location data points collected by Huq’s own platform via the WiFi and GPS method (known as assisted-GPS, or A-GPS) described above:

Location data accuracy
Source: Huq Industries Location Data Asset (2020)

This shows a modal average of roughly 20m accuracy, with a mean average of something close to 65m. In practical terms, that degree of accuracy is not enough to confidently place a mobile device within any particular shop, store or other physical environment – or indeed to determine contacts between handsets or to differentiate between different handsets without a mobile device ID. The typical location accuracy available through Cell ID methods without reverse-triangulation ranges between 500m to 2km in areas with good coverage.

6. How is location data used in advertising?

Mobile location data will most often be offered to advertisers as a means to reach audiences who have been or are frequently present across different neighbourhoods, cities, states or countries. Location data derived from the mobile operating system per se is not required for this, and will often be supported by an altogether different technique not discussed here known as IP to Location. A more traditional use-case for mobile location data is to target or re-target audiences that are or have been present at larger destinations such as train stations, airports, shopping malls or sports stadia with content that advertisers consider to be relevant to them on that basis. These applications of location data are known as ‘Proximity Campaigns‘. More sophisticated still are use-cases that attempt to organise or segment audiences according to their socio-economic or geo-demographic profiles based on the neighbourhoods in which those mobile devices appear to dwell, and cross-referencing that with commercial or census-based datasets to provide the usable output. 

While by its very nature advertising is designed to prompt a response from an individual to an ad, the size and scale of audiences required by the economics of the advertising industry means that the exposure of professionals handling location data in advertising is only ever in aggregate, across potentially many millions of mobile device IDs at any one time.

7. Can I see location data held on me?

Major mobile operating system vendors Apple (iOS) and Google (Android) have both taken steps to incorporate features that allow users to view and download their location histories as shared across all apps on their devices, and any location data observations made by the operating system itself. 

On iOS, navigate to:

Settings > Privacy > Location Services > System Services > Significant Locations

On Android, open the Google app and navigate to:

More button (3-dots) > Your name > Manage your Google Account > Data & personalisation > Activity controls > Location history > Manage Activity

At that point you will be able to browse the mobile location history data that your device holds for you.

It is also possible that apps that you have downloaded and that you use are also collecting and storing this location data for their own or for wider commercial purposes. You should read the Terms & Conditions of each app you download, and look for the section often named ‘Location services’ or ‘Location data’. These terms will list any third party companies or organisations that they have permitted to share your location data with. 

To obtain a record of the information that they, or any other third-party service holds on you, you can submit what is known in the EU under GDPR and in the US under the CCPA as a ‘Subject Access Request’. This requires you provide a copy of your mobile device identifier, typically the ‘Advertising ID’ (Android) or IDFA (iOS) together with certain other identity documents, on receipt of which the relevant company is required to respond with any records relating to you if a match is found.

8. Is location data and compliant with regulations like GDPR and CCPA?

Yes, but any organisations sourcing or handling raw location data need to take steps to ensure that the data has been collected with the relevant consent. There is no getting away from the fact that under GDPR, location data is personal data. The collection and use of location data complies with GDPR and CCPA so long as the data, purpose and company responsible have been explained to the end-user prior to collection, and that the user has provided their consent for this use and furthermore is provided with simple mechanisms to revise their consent and obtain copies of their data should they wish them. Consent management platforms (CMPs) that simply the consent process for the end user are a very good way to manage this process and in fact opt-in rates for data collection are often significantly higher for CMPs than those without.

9. How will changes in Android 14 affect the location data industry?

Both Apple and Google have introduced new measures designed to restrict the circumstances in which apps may collect mobile location data, and to make that data available to mobile location data companies. On iOS, apps must request access to location data while not actively in use (ie. in background mode) in a very specific way that involves multiple confirmations of end-user approval. If that app wishes to make location data available to third parties, its terms of use must request consent in the relevant way in order to be compliant with Apple’s own App Store Terms & Conditions.

Similarly, apps must also seek end-user consent in the appropriate way in order to be compliant and available for download on the Google Play Store. The latest edition of Google’s Android operating system policy also requires app developers to provide a suitable written justification for accessing location services. This submission is reviewed by the Play Store team who adjudicate on whether access should be provided. In most cases where there is a legitimate reason for accessing the mobile device location – such as for weather, fitness and navigation apps – it will be hard to deny access. For Sudoku apps that request fine location in the background however, time will soon be up. Further, in Android 14 the use of location services by apps will be further highlighted to users via iOS-like popups.

Edit: As of September 2020, Google have announced plans to roll-back the implementation of its use-case review policy to mid-2021.

10. What does 2021 hold for mobile location data?

As we look towards the year ahead, it is clear that the market for mobile location data beyond advertising has accelerated – not least as a result of the Covid-19 pandemic and the urgent need for businesses, governments, NGOs and retailers to understand and plan how best to navigate the disruption particularly to the physical domain that this has presented. 

What was considered a mid-priority, experimental dataset for many location data customers had rapidly become indispensable. The same crisis that has brought mobile location data for research to the fore has also seen advertising investment decrease sharply. Those location data vendors whose businesses are largely predicated on that marketplace will turn their attention to more analytical applications and help to drive maturity in the industry. 

At the same time, pressures from regulators and operating system vendors will continue to build and stress the supply of mobile location data for any of these purposes, creating a shortage at the very same time as demand is rising. How companies working within this space will adapt and compete will be an exciting and formative experience for the industry.