A three platform experiment points to mobile and web, but the more useful finding is how the team measured it.
Most companies do not choose a platform for their accounting system. They inherit one. The system arrives on the desktops in the office, people learn it, and the question is never asked again.
A study from Universitas Brawijaya put that question to the test. The team built the same accounting system for desktop, mobile and web, then timed seven employees at a spare parts distributor in Surabaya as they worked through the same tasks on each.
Time to complete a task, measured in seconds, varied with the platform. The authors recommend building for mobile or web rather than desktop. They are careful not to crown a single winner: web had the edge in some scenarios and mobile in many others.
That split is the useful part. Typing a long invoice at a desk and checking a stock figure on the warehouse floor are different jobs, and they do not share a best platform.
The cost of building three versions is real. The team needed about three months to develop all three, which is the strongest argument for testing before committing to a platform strategy rather than after.
Diana et al. (2023). The Impact of Platform Differences on System Use in Accounting Information System. Proceedings of the 8th International Conference on Sustainable Information Engineering and Technology, 439-445.
