Models and Theories in Human-Computer Interaction/TAM - a useful predictor of user adoption
In my work experience building consumer software products that provide utility, I have found TAM to be a useful predictor of user adoption, in that the attitude of the user is influenced by perceived usefulness and perceived ease of use. Specifically, I have found that the perceived ease of use has a direct influence on perceived usefulness and that system design characteristics can influence both (Chuttur, 2009).
In my work I undertook extensive customer discovery sessions for activity trackers. Variations of prototypes were used, with varying amounts of utility, from simple step counters, through to more robust prototypes with trackers for exercise, caffeine consumption, water consumption, weight, heart rate, caloric intake and sleep patterns.
Users indicated their relative intent to use each of the prototypes and the perception of usefulness or utility it would provide in their lives. Each prototype had basic and consistent UI elements and good usability, so as not to introduce any bias. Users then had the prototype they were exposed to installed on their smartphones and were asked to use them over the course of a few weeks. Their usage was tracked and analyzed, which overcame one of the criticisms of TAM research, in that self-reported data is subjective and unreliable (Chuttur, 2009).
Consistently we found that the prototypes with multiple utilities, were perceived as the most useful to users when first exposed to them, stating they would have the most beneficial impact on their lives. However the greater the utility, the greater the amount of user input required to receive the benefit. The prototype with the least utility, the step counter, was initially perceived as less useful and required no user input at all. It operated in the background as a user moved throughout their day.
We found that users inputting data in the heavy utility prototypes, waned very quickly, and by the end of the test they weren’t inputting any data at all, nor were they visiting the application daily, as they had been in the beginning. Conversely the prototypes with less utility that were perceived as only moderately useful, were still being used on a daily basis by the end of the test.
This indicated to me that the level of effort, or ease of use, in the first example, was too high and ultimately impacted the perception of usefulness and user adoption. Whereas in the second example, the application became more useful by its ease of use or low level of effort. In both instances the system design characteristic of data input, directly impacted both perceived ease of use and ultimately, perceived usefulness.
Theoretically then, design characteristics that offer benefit to the user without taxing them with a large amount of input investment, will increase user adoption.
This experience leads me to agree that TAM is a simple model that can be applied to any system, providing very broad information about perceived usefulness and perceived ease of use. I also accept some of the limitations around mandatory use, as my own work involves voluntary behavior.
Chuttur M.Y. (2009). Overview of the Technology Acceptance Model: Origins, Developments and Future Directions. Indiana University, USA . Sprouts: Working Papers on Information Systems, 9(37). http://sprouts.aisnet.org/9-37