Skip to main content
Loqate have a unique process that brings together and combines multiple expert data sources to create one best record in the address and location hierarchy. We call this process the Persistent Data Hierarchy, or PDH. A country dataset is improved by using the PDH process to enhance reliability, improve the speed and quality of the location data, creates consistency, adds normalization, and resolves data errors.

​
How it works

  1. Loqate will source and evaluate the best sources for address and location data
  2. Using our home-grown, expert system, the data is scanned and mapped into a standardized hierarchy
  3. The completed dataset is then exported for use across all Loqate products

​
Benefits of a single hierarchy

The PDH process benefits customers in two ways: completeness and coverage. By blending multiple sources, the improvement process will complete missing fields, resolve component-level conflicts, and bring consistency and alignment across the regional hierarchy. For example, one source may state a value as a city, whereas another will identify it as a Village or smaller regional boundary. The PDH software will resolve this, and also enrich or fill any gaps that may be unique to each source. Coverage is is measured by the number of delivery-point records, which is maximized by having more sources. This is dependent on the quality of data already available in the market for each country.

​
Coverage status

We have now updated the Data Coverage page to include if a country has been through our PDH process. Measuring improvement We internally measure the improvements made by the PDH process when we complete it for a particular country. You may see a record returning a level 2 (locality level) prior to PDH and a level 4 (premise level) after PDH; these types of changes demonstrate a positive improvement due to better alignment and cleaner structures in the reference data. In cases where a record does the opposite, the enhanced dataset may be resolving a previously incorrect or false result. These changes are dependent on the amount of sample data available.

​
Coverage & Completeness

For countries using existing data sources, these are unlikely to show any big changes to the total number of deliverable records. Many improvements in datasets will be based around field/value alignments and corrections to the geography hierarchy.