Wiki Categories

Model Evaluation

Point of Interest Data

What is point of interest data?

Point of interest (POI) data provides intelligence on real-world public places, such as retail stores, restaurants, parks, monuments, and other sites of convenience or tourist attractions.

A POI location can be temporary, such as the location of a public event or a small shop. It can also be permanent, such as a monument or a national park. POI data provides helpful information about a place offering unique products, services, or experiences. People visit POIs for those interests and use mobile devices to locate them.

Public sector institutions and private businesses use POI data to learn more about specific locations and the people visiting them. This intelligence provides insights into why and how people use the POI, helping improve the experience.

A point to note is that POI data is a category of geospatial data, while POS (Point of Sale) data provides information about sale transactions and is a category of retail data.

Where does point of interest data come from?

The primary data provider for point of interest data is onsite data collection. Other point of interest databases include government sources, media platforms, geocoding, and 3rd party data.

Onsite data provides first-hand information that is high-quality, comprehensive, and accurate. Field representatives collect location intelligence data and gather specific information as required. First-party POI databases often use this method, though it turns out to be time-consuming, resource-heavy, and expensive. As some existing POIs can have very high visitor traffic, onsite data may prove to be difficult to update quickly. Another point to consider is the opportunity to collect data round the clock in the case of some 24/7 points of interest.

Most government sources offer official records of points of interest in their area, city, region, or country.  Part of this information may have restricted access due to privacy regulations.  Government sources often do not cover some POI categories such as new, upcoming, or temporary points of interest. Data for these types of locations needs to have other reliable sources.

Media platforms are increasingly becoming a major source of POI data. Data vendors use machine learning algorithms and other tools and APIs to identify locations, which have high traction from media and social media platforms. Sourcing from these platforms is efficient, cost-effective, and easily scalable for obtaining frequently updated data. As websites, microsites, and social media channels are easy to update, many small or temporary points of interest try to provide their information through these platforms. The challenge of point of interest data from media platforms is verifying its accuracy.

Geocoding is another source of this type of data, which provides location information. Data sourced from geocoding is useful for an overview of all the POIs in a specific area, aggregating their information. Data collected from geocoding needs to be augmented with other POI data to provide comprehensive information. Map services (such as digital maps like openstreetmap and navigation systems) often deliver this type of extensive information, though its accuracy and recency may need to be verified.

What types of attributes should I expect when working with POI data?

There are four main types of data attributes:

  • Location: Latitude and longitude coordinates, address, or both
  • Contact information: Name, address, website, phone number.and contact details
  • Function: Type of place - store, restaurant, natural landmark, atm, or other, commercial or public place
  • Franchise information

Other common attributes include images, hours of operation, ratings and reviews, foot traffic data, environmental information in some cases, and current updates issued by the authorities about the POI. Additional attributes may describe how the data is collected, depending on the type of location.

How should I test the quality of the data?

The most critical factors for the validation of the quality of POI data are accuracy, recency, and scale of coverage. The methods used to collect data impact these three factors. Some methods deliver highly accurate data, though it may not get frequently updated. On the other hand, some may provide large-scale data that lacks accuracy or consistency.

To test the quality of the data:

  • Ascertain the sources of data, and verify their credibility.
  • Determine the coverage and diversity of information about a single POI and if it matches your requirements.
  • Test the data for accuracy, consistency, and completeness.
  • Confirm the frequency of the dataset update and its most recent update.
  • Verify the frequency of dataset vetting for accuracy.

Who uses this type of data?

POI data is used by government, semi-governmental organizations, with many public or private businesses as end-users.  They mostly use it to assess a location, study the behavior of visitors, derive insights, and initiate actions.

Local governments use POI data for infrastructure planning and identifying gaps in public service establishments such as hospitals, schools, parks, or public libraries. They also leverage this data for better distribution of resources to support local tourism and the local economy. It is also used to understand neighborhood movements, to see if anyone needs help, and initiate the required changes. 

Retail businesses use this data to track the performance of their own brick-and-mortar stores as well as those of their competitors. Measuring store visits and foot traffic, they test the store locations for efficiency and profit. Accordingly, they decide on their store location strategy and expansion plans. POI data can also help in improving their distribution planning and employee engagement.

Marketing and advertising companies create geofences around points of interest to create location-based audiences. They leverage such audiences for focused campaigns to increase footfall traffic and conversions. POI data, when augmented by location data from smartphones, offers extra context about the demographics of customers and their needs. Marketers leverage this context to personalize messages and offers to existing as well as potential customers.

When people buy or rent, they consider nearby amenities, parks, schools, hospitals, shopping facilities, and local attractions. POI data offers this information, and real estate investors leverage it for actionable insights. For commercial properties, information about local business opportunities and potential local competitors is vital. The data provides these details, and businesses use it for planning manufacturing or retail establishments. Real estate investors power their analytics with POI data to predict the pricing and real-time trends in the real estate market.

Transportation companies leverage POI data to determine the most efficient routes and fleet sizes.

The manufacturing industry also uses this data when setting up a plant in a specific area to see if the local resources can support it.  

What are the common challenges when buying POI data?

The common challenges when buying the data are accuracy, timeliness, and standardization. A very particular challenge is the ambiguity about what makes a location a point of interest. 

  • Data accuracy: Data provided by different sources may not be complete or accurate. Inaccurate data can lead to imprecise analysis, affecting the decisions and long-term planning. Validating data from one source with other sources can help to improve the accuracy of the data.
  • Data timeliness: Some attributes of the data are transient, depending on specific occasions, holidays, seasons, reclassification of categories, and several other reasons (like the pandemic).  If they are not frequently updated, they can lead to wrong decisions. Data timeliness is a huge challenge when buying this type of data, as some sources do not review their listings regularly.
  • Standardization: POI data does not have legal or empirical standards for identifiers, data formats, or models. The datasets from different vendors can be incompatible with your model or with each other.  Integration needs additional efforts and skills to ensure that you can derive significant value out of the data. Ensuring that the attributes and their formats match the requirements is essential before making any buying decision.
  • Ambiguity and inconsistency: Points of interest are entities whose significance and information evolve over a period of time. At any given time, some sources may declare a particular location as a point of interest, while other sources may not identify it as a significant entity. The significance of a point of interest also varies according to the definitions and priorities of the source. For example, a manufacturing plant may not appear in the tourist attraction listings but will appear in the business listings. There may be inconsistencies in naming and categorizing a location. If not updated continuously, other attributes such as address or hours of operation may also be inconsistent across datasets.

What are similar data types?

POI data is similar to Location Data, GIS data, Map Data, and other geospatial data categories commonly used for diverse use cases.

You can find a variety of examples of geospatial data in the Explorium Data Gallery.

Sign up for Explorium’s 14-day free trial to access the data available on the platform.                

Access POI Data wit Explorium

What are the most common use cases?

The key use cases are geofencing, geo-targeted advertising, marketing campaign strategy, and traffic management.

  • Geofencing: It is a location-based service, which triggers predefined actions when a device enters or exits a virtual boundary known as a geofence. Geofencing uses GPS, RFID, Wi-Fi, or cellular data to identify devices. Geofences are set around popular geographical locations or POIs to leverage the presence of a device. POI data helps marketers to decide on their choices of geofences.
  • Geo-targeted advertising: This type of advertising leverages location and behavior data to deliver personalized customer communication. It uses POI data to target visitors within a defined area around the location of the POI. For example, recommending a specific restaurant when the customers are close to the premises with promotional offers encourages them to visit it. Geo-targeting helps advertisers target a smaller group with more personalized content and relevant promotions, for higher conversion and a better ROI.
  • Marketing campaign strategy: Marketers leverage a wide range of data categories to plan their marketing campaign strategy.  POI data helps them identify areas where they can create opportunities for their products and services. It also delivers insights into the demographic profile of visitors and helps create targeted promotional campaigns.
  • Traffic Management: Efficient management of the traffic flow helps in reducing congestion and accidents. Traffic management systems use data from radars, cameras, and other sensors to arrive at the best guidance for traffic movement. It also leverages weather data and various categories of demographic data to predict rush hours and traffic jams.  POI data indicates the visitor movement patterns in popular areas, helping plan routes, diversions, and traffic directions in the surrounding areas.

Which industries commonly use this type of data?

Most governmental, semi-governmental, and private industries use this data. They include manufacturing, transportation, logistics, healthcare,  retail, CPG, travel, hospitality, leisure, entertainment, financial institutes, insurance providers, banks, and hi-tech companies.

How can you judge the quality of your vendors?

The quality of vendors is illustrated by customer reviews, testimonials, and case studies, usually available on vendor websites. Some vendors also provide recorded demos or arrange a live demo for assessing the quality and suitability. A personal discussion with vendor reps can resolve your queries quickly and ensure integration with your systems.

  • Customer reviews, testimonials, and case studies: Reviews and ratings are a measure of customer confidence, while testimonials discuss the strengths of the vendor. Case studies describe successful projects, highlighting vendor engagement and commitment to the challenges.
  • Demo: A demo is valuable in examining the range of datasets offered, integration capability, and turnaround time. Many vendors provide samples to help you assess if their datasets match your requirements.    
  • Interacting with vendor reps:  This approach always helps understand each other, which can establish a long-term relationship. Discussing requirements, customization, integration, and other issues directly with vendor reps accelerate the process of vendor selection.

 

Additional Resources:

Explorium delivers the end-game of every data science process - from raw, disconnected data to game-changing insights, features, and predictive models. Better than any human can.
Request a demo
We're Hiring! Join our global family of passionate and talented professionals as we define the future of data science. Learn More