Should you build or buy POI location datasets?

Do you need location datasets?

  • Retail — Select sites to build brick-and-mortar shops/stores based on the population density in the specific location
  • Ecommerce — Enhance customer experience by delivering the goods at the right time and place
  • Healthcare — Measure health check accessibility in a locality and determine its impact on people’s health
  • Banking and Finance — Assess brands and industries that will have good returns in the future and analyze ATM positioning in the city
  • Government — Improve the city’s infrastructure and build smarter cities by analyzing the locations.
  • Telecom — Plan network set-up based on demand and competition in a location
  • Marketing — Execute targetted marketing campaigns based on the demographics of the location and target audience
  • Real estate — Choose profitable properties for investments that would yield better results when sold to buyers
  • Restaurant — Determine the foot traffic of a particular restaurant to identify the tastebuds of most people visiting that location

Building POI location datasets

  • Cost: Building location datasets involves a lot of costs, and businesses must ensure that they have a minimal budget to create quality datasets.
  • First, recruitment cost, which involves getting qualified resources onboard, such as GIS experts, AI/ML experts, and software engineers, to build datasets.
  • Second, building web scraping tools to scrape POI information from the public is an additional cost to the company.
  • Third, infrastructure costs, i.e., server, database, and storage costs.
    Other additional costs include partnerships, data maintenance, resource to manage data anomalies, and many more.
  • Data Quality — This factor is vital to your brand image because poor quality data can affect your business reputation and significantly impact the cost.
    To begin with, you can start by using POIs available online for free, such as from OpenStreetMaps. These data sources usually offer continual user-generated places. However, do note that the coverage of these sources may be limited. Only 7 out of 10 places can be extracted. Therefore, a web scraping tool must be built and automate bots to create custom POIs. While building POI datasets on your own, you should also ensure that the place exists in real-time. If not, this can cause serious problems in data analysis and modeling.
  • Data Maintenance — Data cleansing, refreshing, and maintaining are one of the most time-consuming tasks. However, it is important because shops or stores close, open, vacate, and shift from one place to another at any given moment. The Covid-19 pandemic is one latest example that forced businesses to shut down their outlets. In such cases, updating this information in the POI database is essential.
    Additionally, attributes of a few POIs may change, such as opening and closing hours, shop numbers, facilities, etc. Therefore, these must be updated before using them for location-based analysis.
  • Scalable platforms — POI data requirements are huge and unique. Almost every industry is looking for location-based data and insights to boost their business performance. Therefore, in addition to data scientists and engineers who develop these datasets, you also need a research team that can show you the POI needs of the future. When you build POI datasets in-house, future requirements need to be anticipated. Therefore, a scalable platform is important to cater to new geographies, industries, or attributes for future POIs.

Buying POI location datasets

  • Cost — All the overhead cost of recruiting skilled resources and building platforms, tools, and infrastructure gets canceled while opting for this method. It is noted that the cost of creating a POI dataset in-house is usually slightly more expensive than buying from the vendor. Therefore, validate if the amount, time, and effort spent in creating POI datasets are justified compared to buying, where the risk of failure is minimal.
  • Data licensing — Double-check the licensing terms while you leverage a POI data vendor. In rare scenarios, POI vendors do not allow buyers to use their POI data for commercial uses. Also, before licensing, it may be mandatory to ask the following questions –
  • How many industries or geographies have been covered?
  • What are their quality and coverage percentage?
  • What is the total volume of the records?
  • Flexible platforms — When you buy POI data from the vendor, ensure that they offer all flexible delivery formats that can help you ingest the data to your platforms. POI data can be delivered through various ways — excel files, data visualization models, or directly integrated into the company’s API. Also, ensure that on-demand data generation is possible and delivered on time. is a POI and Polygon data vendor that ensures 95% accuracy and 96% global coverage. They also help customers by fulfilling on-demand POI data requirements.
  • Authenticity & Refresh Cadence — Check the actual source of the data that the POI data vendor is providing you. The data licensed must be from reliable sources and not contain any faulty locations. Similarly, ensure that POI data is updated regularly to attain recency.
  • What is the source of data?
  • How often is the database updated and maintained?
  • Are obsolete data deleted immediately?




Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store

Get actionable data and meaningful insights with our data extraction and preparation solutions that cater to all your data needs.