How grocery shopping data is unlocking financial inclusion

Access to affordable credit is fundamental to personal resilience and economic advancement. It helps fund housing, education, small businesses, and insurance to protect against financial shocks.

Globally, 1.4 billion adults have no access to formal financial services because they lack a credit history, which is only acquired once someone has been granted credit. This catch-22 means millions of people are financially excluded.

This is not only a problem in emerging and developing markets, but also in developed markets like the US and the UK where millions remain underserved: approximately 45 million Americans are either credit invisible or have unscorable credit files, and around 5 million UK residents lack a mainstream credit history.

For financial institutions, this represents not just a moral imperative, but also a major opportunity to unlock a new and largely untapped market through innovative and ethical data use.

Grocery shopping data is emerging as one of the most powerful alternative data sources for understanding the financial behaviour of “credit invisibles”.

These four key characteristics highlight why grocery data is so insightful for credit scoring people with no credit history:

Everyone buys groceries. Grocery shopping is a universal necessity that cuts across socioeconomic, geographic, and demographic boundaries. This makes grocery data uniquely representative of the broader population, which is a rare attribute among alternative data sources.

Unlike many traditional data sources, grocery data is continually refreshed. Most consumers shop for groceries weekly, if not more often. This regularity offers a real-time view into consumer behaviour, enabling financial institutions to assess an individual’s current financial situation with striking accuracy.

Grocery shopping data captures detailed behavioural signals. For example, consistent purchasing of staple goods at the same time each month can indicate budgeting discipline. Price sensitivity and use of discounts may suggest cautious financial management. A high-percentage of healthy food items and lack of junk food can be an indicator of financial responsibility.

The high frequency of grocery shopping offers a dense timeline of behavioural data, allowing models to detect consistent financial habits, patterns, and anomalies. Unlike once-off data points like loan applications, grocery data builds a behavioural track record over time.

Research by scholars at Rice University, the University of Notre Dame, and Northwestern University, Using Grocery Data for Credit Decisions, found that variables such as shopping frequency, consistency in spending, choice of products, and use of discount programs correlate strongly with credit risk profiles. Importantly, it demonstrated that these behavioural patterns could significantly improve the predictive power of credit models, particularly for consumers without formal credit histories.

While this remained theoretical for many markets, South Africa became the first to put it into practice at scale.

In South Africa, the theoretical promise of grocery data was realised through a groundbreaking collaboration between the continent’s largest grocery retailer and its leading banks. The objective: to determine whether shopping behaviour could serve as a reliable proxy for creditworthiness.

Crucially, this partnership was designed with consumer privacy at its core. Rather than sharing or exchanging consumer data, the banks and retailer worked within a privacy-preserving data collaboration platform powered by Privacy-Enhancing Technologies (PETs).

These technologies enabled each party to contribute anonymised data and extract shared insights without ever exposing personally identifiable information. This approach ensured full compliance with stringent data privacy regulations while unlocking valuable insights.

Using the platform’s built-in AI and machine learning capabilities, anonymised datasets from both parties were securely overlapped and analysed to uncover behavioural patterns. By identifying shopping traits common among individuals with strong credit histories, the banks were able to build and train predictive models capable of scoring previously unscored individuals without ever revealing personal identities.

The results were transformative:

What began as an experiment in South Africa is now a replicable blueprint for driving financial inclusion on a global scale. From Africa to Latin America, and from the US to Europe, millions of people remain outside of the margins of the financial system.

Grocery shopping data is recent, frequent, universal, and rich in behavioural insights. Coupled with banking data within a privacy-preserving data collaboration environment, it's opening the path to financial inclusion and protection for millions.

Financial inclusion has remained out of reach for far too many, for far too long. Grocery data, used responsibly and collaboratively, may be the innovation that changes that at scale.