Good for the Many or Best for the Few? A Dilemma in the Design of Algorithmic Advice
This addresses a design problem for developers of algorithmic advice systems in domains such as route planning and well-being, though it is incremental in exploring user preferences.
The paper tackles the dilemma between offering goal-directed advice (optimal for individuals but less adopted) versus adoption-directed advice (more widely adopted but sub-optimal), finding a preference for goal-directed advice across domains like finance and health, with significant variation in results.
Applications in a range of domains, including route planning and well-being, offer advice based on the social information available in prior users' aggregated activity. When designing these applications, is it better to offer: a) advice that if strictly adhered to is more likely to result in an individual successfully achieving their goal, even if fewer users will choose to adopt it? or b) advice that is likely to be adopted by a larger number of users, but which is sub-optimal with regard to any particular individual achieving their goal? We identify this dilemma, characterized as Goal-Directed vs. Adoption-Directed advice, and investigate the design questions it raises through an online experiment undertaken in four advice domains (financial investment, making healthier lifestyle choices, route planning, training for a 5k run), with three user types, and across two levels of uncertainty. We report findings that suggest a preference for advice favoring individual goal attainment over higher user adoption rates, albeit with significant variation across advice domains; and discuss their design implications.