Privacy by design: from location data to location intelligence

Kjartan Slette · Co-founder, Pine59

Location intelligence helps businesses and organizations understand how physical places are used. How many people visit a shopping area? Where do visitors come from? How do movement patterns change over time?

Answering these questions requires mobility data. And working with mobility data comes with an important responsibility: protecting the privacy of the people behind the observations.

At Pine59, privacy is not something we add at the end of the process. It shapes the data we work with, how we process it, and what we make available to customers.

Privacy starts with the data source

Pine59's products are built on pre-aggregated first-party data, regardless of the underlying signal source. The aggregation we apply removes all personally identifiable information and is designed so the data can never be traced back to an individual.

Whether the original observations come from mobile applications, telecommunications networks, or other sources, individual signals are aggregated before becoming part of Pine59's data foundation.

We then apply quality checks, validation, and statistical modeling to produce estimates of activity across places, areas, and time periods.

This allows us to estimate visitation to a shopping center, compare activity between neighborhoods, or analyze movement patterns between geographic areas, without accessing individual location histories.

Our focus is not on understanding where individual people go, but on understanding how places and areas are used by populations.

Privacy protection in the data itself

Aggregation is the foundation, but it is not the only safeguard.

Depending on the country, data source, and product, Pine59 applies additional techniques to protect privacy while preserving the statistical value of the data.

Minimum thresholds and K filtering

Some statistics are only published when they represent a sufficiently large group of observations.

When a group is too small, the result is suppressed rather than published. This reduces the risk of individual contributions being isolated in small datasets.

Differential privacy

For certain datasets, carefully calibrated statistical noise is introduced to reduce the risk of information about individuals being inferred from published statistics.

The aim is to preserve meaningful patterns across populations while providing additional protection for individual privacy.

Rounding

Published statistics may also be rounded, providing another layer of protection against overly precise reporting.

These safeguards are applied according to the characteristics of each dataset and methodology⁠ (opens in a new tab). Not every technique is used in every product, but the underlying principle remains the same: useful statistical insights should not come at the expense of individual privacy.

What customers actually receive

Pine59 customers receive aggregated statistical insights through our dashboards, APIs, and data services.

Our products are designed to answer questions about places and populations, not to identify or follow individual people.

Whether customers explore data through Pine59 Insights, integrate it into their own systems, or use it in AI workflows, the same principle applies: the output is aggregated location intelligence, not individual mobility data.

Privacy as a product principle

Privacy is often discussed as a compliance requirement. For location intelligence, it is also a fundamental product-design decision.

We believe organizations can make better decisions about cities, retail locations, infrastructure, and public spaces without knowing the movements of individual people.

The value lies in understanding collective patterns.

That is why Pine59 is built around statistical insights about places and populations, with privacy considerations extending from the original data source to the final product.