Independent Research and Policy Advocacy

Using behavioural science to design effective consent: A case study of Account Aggregators

Consent is necessary for data protection (even if it is not sufficient in itself). Yet, the way consent artefacts are currently designed does not allow users to pay active attention to consent artefacts because of their design, the language used in them, and because of the users’ own biases and mental models. Therefore, the consent that financial service providers obtain from users when they process their data is far from the RBI standards of consent being informed and explicit. These concerns may be sharper for new-to-finance, digital immigrant users with low income and literacy (‘Constrained users’).

Dvara Research and Final Mile are undertaking a project to redesign consent artefacts so that users are able to (i) read and understand the terms and conditions of the use of their personal data, and (ii) provide informed, thoughtful consent that upholds their right to privacy and data protection. The first use case that we take is that of a constrained user taking a loan using the Account Aggregator channel.

This is a behavioural study unpacking how users make consent decisions and the factors that influence it. The insights from the study are used to create design templates to make consent artefacts more effective. The study comprises three broad stages:

Stakeholder immersion with AA and digital lending providers, and design, customer protection and behavioural science experts to better understand the behavioural underpinnings of users making consent decisions.
Behavioural field study with 60 constrained customers in Uttar Pradesh and Maharashtra using gamified simulations of the AA-loan transaction (the EthnoLab™️) in which users made consent decisions.
Piloting the learnings from the field study, including the design prototypes created based on learnings, with a live AA and lender.

All the outputs from the project are available below.

Using behavioural science to design effective consent: A case study of Account Aggregators