Perceived usefulness and perceived ease of use jointly determine whether a user adopts a new technology

What it is

The Technology Acceptance Model (TAM) holds that whether a user adopts a new technology is jointly determined by two subjective beliefs: perceived usefulness (will this let me do the task better?) and perceived ease of use (how much effort will it take to learn and use?). The key point is that both are the user's own anticipatory beliefs, not the technology's objective feature list and not a usability-test score — the same feature can receive completely different subjective ratings from users with different backgrounds.

Why it happens

Neither variable is dispensable, because they influence adoption intention through different paths: perceived ease of use acts partly directly on intention (effort savings has value in itself) and partly by raising perceived usefulness — the easier something is to learn, the more readily a user believes they can actually make it work for them, and so judges it useful. The two also interact rather than acting independently: when perceived usefulness is high enough, users tolerate lower ease of use (professional software gets adopted by professionals even when it's hard to learn, because they have no alternative); but when usefulness is still unclear, the weight on ease of use grows, because the user is at that moment deciding whether it's worth the effort to find out, and ease of use is itself the entry threshold for that decision.

Studying it

The standard approach measures both variables with self-report scales: several items gauge perceived usefulness ("using this system would improve my efficiency") and perceived ease of use ("learning to use this system would be easy for me"), collected before and after adoption or use, with regression or structural equation modeling testing each variable's predictive power over adoption intention or actual use behavior. This paradigm has been replicated across many technology categories over decades and has spawned extended versions adding variables like social influence or facilitating conditions, but the core two-variable structure has held up. A methodological caveat: both variables measure pre-adoption expectation, not post-adoption objective experience — the gap between expectation and actual experience itself shapes continued-use intention, so a scale score taken at the moment of adoption cannot predict long-term retention on its own.

Where it stops holding

The model's explanatory power is strongest in voluntary-use settings, where the user can choose whether to use the technology at all; in mandatory-use settings (a job requires a given system), the two beliefs' power to predict "adoption" drops sharply, because the user has no choice and adoption is no longer an intention-driven behavior — the two beliefs then mainly shape use quality and satisfaction rather than adoption itself. The scale also captures a static belief at one point in time and has little to say about how a given user's beliefs evolve over long-term use, which requires pairing it with longitudinal measurement.

Applying it

• Before launching a new feature or product, measure target users' perceived-usefulness and perceived-ease-of-use expectations with a short scale, not just a usability test — many products test well on objective usability yet users simply don't use them, and the reason is often that perceived usefulness was never established: users don't understand or believe the feature matters to them.
• When perceived usefulness is inherently low (users aren't aware they have the need the feature addresses), put resources first into helping users understand what problem this solves for them, rather than continuing to polish ease of use — ease-of-use improvements have limited marginal return when the usefulness perception is missing in the first place.
• Verification: insert a short survey step into the onboarding flow for a new feature ("do you think this is useful to you?" / "do you think it would be easy to learn?"), then correlate the answers with that cohort's actual usage frequency over the following thirty days to determine which belief better predicts retention, and direct subsequent optimization toward that side.

Related

• Same group: A11.12.2 Self-efficacy is a user's subjective confidence that they can learn and use a given technology well · A11.12.3 Users with low self-efficacy overestimate task difficulty and give up before trying · A11.12.4 Success or failure in early use experiences directly shapes subsequent self-efficacy · A11.12.5 Raising self-efficacy drives adoption more than simplifying features alone
• Adjacent: A11.13.3 A design that assumes users have mainstream digital literacy systematically excludes low-literacy groups
• Search terms: Technology Acceptance Model · perceived usefulness · perceived ease of use