Joint Border Analytics

Customs uses data analytics to help identify risks to New Zealand.


Customs’ main purpose is to stop any dangers, hazards, and threats entering New Zealand, as well as protecting and contributing to NZ’s economy by ensuring that the appropriate duties are paid and collected.

Information is an important factor in how Customs achieves this purpose and we use advanced intelligence-gathering techniques to uncover illegal activity. Data analytics is one way of gaining better value from information and intelligence to help control risks at the border.

The Joint Border Analytics (JBA) team was established in late 2016 and consists of staff and contractors from Customs, the Ministry for Primary Industries and the Ministry of Business, Innovation and Employment (Immigration NZ). Data modellers, data scientists, data wranglers and business analysts work together in the JBA team, with input from subject matter experts, to gain new insights into border risk through the use of analytics software and data sharing.

JBA uses data from across the cargo, passenger, and mail streams as well as open source data. The team has looked at a range of Customs risks and issues including methamphetamine importation, short-paid revenue, changing passenger volumes, and the Darknet drugs markets.

To ensure that data and analytical techniques are appropriately and safely used by the Joint Border Analytics team, processes and controls are in place including seeking legal advice on information sharing and use, developing privacy impact assessments, seeking approval for analytics initiatives from the Joint Border Analytics Governance Group and operating in a secure offline environment.

In 2018, a cross-government review was undertaken of how government uses algorithms to improve the lives of New Zealanders. The review aimed to ensure New Zealanders are informed, and have confidence in, how the government uses algorithms. Customs responded to this survey. A report on this review was issued in June 2018 along with a one page summary.

Before JBA start analysing data for a business unit, they are required to undertake a Privacy Impact Assessment (see an explanation). As PIAs are cleared for publishing we publish them here.

In addition where appropriate, we will publish output from this unit’s work.

One of these outputs was research into the Darknet. It was undertaken as a proof of concept in regards to the capability of the software being used. This output is Welcome to the insights of the DarkNet. Please note that the data used is from 2014-16 meaning this output is dated.