Expectation-Confirmation Model (ECM) – Bhattacherjee (2001)

Why was the model made?

Bhattacherjee developed the Expectation-Confirmation Model of Information Systems Continuance (ECM-ISC) in response to a critical gap in information systems research. While substantial research had examined what causes individuals to initially accept and use information systems (the adoption stage), very limited research addressed what determines whether users continue using systems after initial acceptance. This distinction is crucial because long-term viability of information systems depends on continued use rather than first-time adoption. The motivation for creating ECM-ISC emerged from recognizing that system discontinuance—users stopping use after initial adoption—represents a significant problem in business-to-consumer electronic commerce and information systems implementation. For Internet service providers (ISPs), online retailers, online banks, online travel agencies, and other information system providers, acquiring new customers may cost as much as five times more than retaining existing customers through continued use.

This economic reality made understanding continuance decisions critical for organizational success. The impetus was also theoretical. Bhattacherjee noted that existing acceptance models concentrated on pre-consumption (initial acceptance) variables and could not explain why users discontinue systems after initial use. The expectation-confirmation framework from consumer behavior literature provided a valuable foundation, having demonstrated strong predictive ability for understanding post-purchase repurchase decisions in consumer product contexts. However, this framework had not been systematically applied to information systems continuance. The paper addresses substantive differences between acceptance and continuance behaviors. Initial acceptance involves forming judgments about an unfamiliar technology based on limited information, whereas continuance involves actual use experience, confirmed expectations, and refined perceptions. Bhattacherjee recognized that acceptance and continuance behaviors employ different psychological processes and require distinct theoretical explanations.

Many individuals who initially accept a system may discontinue use if their expectations are not confirmed through actual use. The research emerged from understanding that information systems continuance is not merely the inverse of discontinuance. Rather, continuance is an active, decision-making process where users evaluate their experiences, compare outcomes to expectations, and decide whether to continue using systems. This continued use decision parallels consumer decisions about repurchasing products or services after initial trial.

How was the model’s internal validity tested?

Bhattacherjee employed a rigorous empirical research methodology to test ECM-ISC’s internal validity. The study involved a cross-sectional field survey of online banking users, providing data about their information systems continuance intentions and the hypothesized antecedents of those intentions. The sample consisted of 1,000 online customers randomly selected from the customer database of a large national bank in the United States. These customers were solicited to participate in an online survey about online banking practices. The sample included established online banking users (customers who had online accounts for two months to three years, with mean of eight months) and balanced representation across diverse household income levels and professions. This sampling strategy ensured data collection from actual users with meaningful experience using the system.

The research instrument measured four key constructs identified in the theoretical model: IS continuance intention (users’ intention to continue using the online banking system), satisfaction (users’ affective response to prior online banking use), perceived usefulness (users’ perception of expected benefits from continued banking system use), and confirmation (users’ perception of congruence between expectations and actual system performance). Measurement development involved careful operationalization of constructs using established scales from information systems and consumer behavior literature. IS continuance intention was measured using items adapted from Mathieson’s (1991) behavioral intention scale, modified to capture users’ intentions to continue using online banking rather than initially accepting it. Satisfaction was measured using Spreng et al.’s (1996) overall satisfaction scale, employing seven-point scales with semantic differential items (very dissatisfied/very satisfied, disappointed/pleased, etc.).

This scale measured affective responses to online banking use. Perceived usefulness was adapted from Davis et al.’s (1989) perceived usefulness scale, originally developed in information technology acceptance research. Items assessed users’ perceptions of expected benefits from continued online banking use, including performance, productivity, and effectiveness. Confirmation was operationalized using a new scale developed specifically for this study, measuring users’ perceptions of congruence between pre-consumption expectations and post-consumption actual performance. Items assessed whether online banking performed well, lived up to expectations, and provided the features and functions expected. The measurement scales demonstrated acceptable reliability and validity properties. Construct validity was assessed, confirming that measurement scales appropriately captured the underlying constructs. Scale items showed consistency in measuring their respective constructs, and constructs demonstrated appropriate relationships with behavioral intentions.

The research tested hypothesized relationships among constructs through path analysis and structural equation modeling. Specifically, the model tested whether confirmation directly influences both satisfaction and perceived usefulness (post-adoption expectations), whether perceived usefulness influences satisfaction through a different pathway, and whether both satisfaction and perceived usefulness influence continuance intentions. The empirical analysis examined whether these hypothesized relationships were statistically significant and in the predicted directions.

How was the model’s external validity tested?

Bhattacherjee addressed external validity through several approaches. First, the choice of online banking as the research context provided ecological validity. Unlike laboratory settings or purely hypothetical scenarios, the study examined actual users making real continuance decisions about information systems they genuinely used. Online banking represented an appropriate context for studying continuance because users make active decisions about whether to continue using online banking services, and discontinuance has meaningful consequences (reverting to traditional banking methods). Second, the sample composition enhanced external validity by including diverse household income levels, professions, and demographics. Rather than restricting the sample to technology enthusiasts or early adopters, the study included mainstream online banking users representing broader population segments. This diversity increases confidence that findings generalize beyond narrow user populations.

Third, the theoretical framework itself contributes to external validity. Expectation-Confirmation Theory has demonstrated strong predictive ability across diverse consumer product and service contexts, from automobile purchases to restaurant services to professional services. By demonstrating that ECT applies to information systems continuance, the model suggests that underlying psychological mechanisms governing post-purchase decisions in consumer contexts also operate in information systems contexts. This theoretical consistency across domains enhances confidence in generalizability. Fourth, Bhattacherjee notes that findings in the online banking context may apply to other information-intensive business-to-consumer electronic commerce settings. While online banking provided the specific test context, the model’s logical structure suggests applicability to other electronic services, online retailers, online travel services, and similar contexts where users make decisions about continued engagement with information systems.

How is the model intended to be used in practice?

The ECM-ISC model provides critical guidance for information systems practitioners and managers responsible for maintaining user engagement and preventing discontinuance: For system design and enhancement, the model indicates that perceived usefulness of continued use is critical for continuance intentions. Practitioners should focus on ensuring that information systems provide clear, tangible benefits aligned with user expectations. System functionality should address user needs effectively and should be designed to deliver demonstrable value in users’ primary activities. For online banking, this might mean ensuring systems provide efficient account management, clear access to financial information, and convenient transaction capabilities. For other information systems, this translates to designing features and functionality that users recognize as valuable for their purposes. For expectation management and user communication, the model emphasizes the critical role of confirmation (whether actual performance matches prior expectations).

Practitioners should set realistic expectations about system capabilities and performance. Marketing communications and system introductions should communicate clearly what systems will and will not do, avoiding overpromising capabilities. System documentation and training should prepare users realistically for their experiences with systems. Bhattacherjee notes that disappointment resulting from unconfirmed expectations directly undermines continuance intentions, making expectation management critical. For user satisfaction focus, the model identifies satisfaction as a central driver of continuance intentions. Practitioners should actively work to enhance user satisfaction through system design, support, and service quality. Creating positive user experiences matters not only for initial adoption but critically for sustained use. This suggests that organizational investment in user support, training, and system responsiveness to user concerns pays dividends in reduced discontinuance.

For understanding user retention, the model provides a diagnostic framework for identifying why users discontinue systems. If discontinuance is occurring, practitioners can assess whether the problem lies with confirmation (system performance not meeting expectations), satisfaction (negative user experiences), or perceived usefulness (users not seeing value in continued use). This diagnostic understanding can guide corrective actions. For market segmentation and targeted retention efforts, the model suggests that retention strategies should address confirmation and satisfaction. Users experiencing disappointment due to unconfirmed expectations should receive additional support, training, or system modifications to address gaps between expectations and actual performance. Users with low satisfaction should be targeted with service improvements or enhanced support. Users doubting perceived usefulness might benefit from training focused on identifying and leveraging valuable system capabilities.

For understanding the relationship between acceptance and continuance, the model illustrates that initial acceptance does not guarantee continuance. Organizations implementing new systems should anticipate that some initial accepters will discontinue use and should design retention strategies accordingly. This is particularly important in consumer-focused electronic commerce contexts where discontinuance rates can be substantial.

What does the model measure?

ECM-ISC measures four primary constructs representing the psychological and affective processes underlying information systems continuance decisions: Expectation, measured at the pre-consumption stage (t1), represents users’ prior beliefs about system capabilities, expected performance, and anticipated benefits from system use. This construct captures the baseline against which users later evaluate actual system performance. Perceived performance (also called perceived usefulness in the model) represents users’ post-consumption evaluation of system benefits and capabilities. This post-consumption (ex post) variable assesses whether the system delivers the value and functionality users expected. Items measure whether the system performs well, provides useful capabilities, and contributes to productivity or effectiveness. Confirmation represents the critical linkage between expectations and post- consumption evaluation. This construct measures the extent to which users perceive congruence between their expectations and the system’s actual performance.

High confirmation exists when performance meets or exceeds expectations, whereas low or negative confirmation occurs when performance falls short of expectations. Satisfaction measures users’ affective response to system use and their overall evaluative judgments about the experience. Rather than capturing utilitarian judgments about system functionality, satisfaction captures emotional responses—whether the experience was pleasant, whether outcomes were satisfactory, and whether the user feels satisfied with the system overall. IS continuance intention measures users’ behavioral intentions regarding sustained use of the system. This construct captures whether users intend to continue using the system in the future, predict they will continue use, and expect to maintain their relationship with the system over time. This post-continuance behavioral intention represents the model’s primary dependent variable.

The model also captures theoretically derived relationships among these constructs. The structural relationships reveal how expectations shape perceptions of confirmation, how confirmation influences both satisfaction and perceived usefulness, how perceived usefulness directly influences satisfaction, and how both satisfaction and perceived usefulness influence continuance intentions.

What are the main strengths of the model?

ECM-ISC demonstrates several substantial strengths: First, the model addresses a critical practical problem neglected in prior research. By focusing on continuance rather than merely adoption, the model acknowledges the economic and strategic reality that retaining users is as important as acquiring them. This practical relevance distinguishes ECM-ISC as addressing a genuine business problem affecting many organizations. Second, the model demonstrates theoretical rigor by adapting a well- established consumer behavior theory (Expectation-Confirmation Theory) to information systems contexts. By grounding the model in expectation- confirmation principles with demonstrated empirical support across consumer product domains, Bhattacherjee provides strong theoretical justification for the model’s predictions. The approach of adapting established theory rather than proposing entirely novel mechanisms strengthens the theoretical foundation. Third, the model’s structure elegantly captures the psychological processes distinguishing acceptance from continuance.

By explicitly incorporating expectation, confirmation, and post-consumption satisfaction, the model articulates how actual use experience reshapes initial acceptance judgments. This progression from expectation through confirmation to satisfaction represents the user’s psychological journey as they accumulate actual experience. Fourth, the model provides a clear diagnostic framework for practitioners. By identifying confirmation, satisfaction, and perceived usefulness as key continuance drivers, the model guides practitioners toward specific improvement strategies. Organizations can diagnose whether discontinuance problems stem from unconfirmed expectations, low satisfaction, or low perceived usefulness and can target interventions accordingly. Fifth, the empirical validation through a field study of actual online banking users demonstrates practical applicability. Unlike laboratory studies, this approach examines real users making genuine continuance decisions about systems they actively use.

This methodological choice strengthens confidence in the model’s real-world relevance. Sixth, the model successfully integrates both cognitive (perceived usefulness, confirmation) and affective (satisfaction) variables in predicting continuance. This integration recognizes that information systems decisions involve both rational evaluation and emotional response, providing a more complete understanding of continuance determinants than models focusing solely on utility considerations.

What are the main weaknesses of the model?

ECM-ISC also exhibits limitations affecting its scope and applicability: First, the model’s reliance on a single context (online banking) limits generalizability. While expectation-confirmation principles should apply broadly, specific confirmatory evidence comes from one industry and one technology platform. Discontinuance drivers may differ for other types of information systems (such as enterprise systems, social media, or productivity software) where use patterns, expectations, and satisfaction drivers differ from online banking. Second, the model may not adequately capture the complexity of post- adoption decision-making. The cross-sectional research design measures intentions at a single point in time rather than observing actual continuance decisions over longer periods. Users’ actual continuation or discontinuation decisions, measured prospectively, might reveal different patterns than stated intentions measured retrospectively. Third, the model does not examine potential moderating factors that might affect the relationships among constructs.

For example, the relationship between confirmation and satisfaction might be moderated by user experience level, perceived system importance, or availability of alternative systems. The model treats relationships as universal rather than examining contingencies. Fourth, the model may underspecify how perceived usefulness influences continuance intentions. While TAM research established that perceived usefulness is a primary technology acceptance driver, ECM-ISC’s finding that satisfaction mediated the effect of perceived usefulness suggests indirect effects. However, in continuance contexts, direct effects of perceived usefulness (recognizing ongoing benefits) might also matter, particularly for system users who are pragmatically oriented. Fifth, the measurement of confirmation through belief-based items (system performed well, met expectations) may not fully capture users’ actual performance experiences. Objective measures of system performance or user outcomes might reveal different patterns than subjective perceptions of confirmation.

Sixth, the model does not examine switching costs, lock-in effects, or behavioral inertia, which may influence continuance decisions independently of satisfaction and perceived usefulness. Users might continue using systems despite low satisfaction or perceived usefulness if switching costs are high or if inertia (habitual continuation) operates psychologically. Seventh, the model may not adequately address how users’ continuance decisions involve anticipated future use rather than merely past experience. Users might continue using systems based on expected future usefulness or anticipated satisfaction even if past experiences were disappointing.

How does this model differ from older models?

ECM-ISC represents an important theoretical shift in several dimensions: First, ECM-ISC addresses a different decision stage than prior acceptance models. TAM and related models focused on initial adoption—the decision to begin using systems. ECM-ISC examines continuance—the decision to maintain use after initial adoption. This temporal shift acknowledges that different psychological processes govern initial adoption versus sustained use. Second, ECM-ISC incorporates actual use experience as fundamental, whereas acceptance models operated with limited experience information. Acceptance models predict initial use intentions based on beliefs formed with minimal hands-on experience. ECM-ISC assumes users have actually used systems, accumulated experience, observed actual performance, and refined their perceptions based on this experience. This incorporation of actual performance against expectations represents a fundamental shift. Third, ECM-ISC explicitly integrates expectation and confirmation processes absent from TAM.

Rather than treating perceived usefulness and ease of use as independent judgment dimensions, ECM-ISC roots these judgments in whether actual performance confirms prior expectations. This framing acknowledges that users’ post-adoption perceptions are shaped by prior expectations and the degree to which actual experience matches those expectations. Fourth, ECM-ISC prioritizes satisfaction as a critical mediating variable between performance confirmation and continuance intentions. While acceptance models recognize satisfaction, ECM-ISC identifies satisfaction as the primary psychological driver of continuance decisions. This reflects understanding that affective responses (whether users feel satisfied, pleased, and content with systems) determine sustained engagement more strongly than purely utilitarian judgments. Fifth, ECM-ISC draws explicitly from consumer behavior literature and expectation-confirmation theory rather than purely from technology acceptance research. This interdisciplinary foundation enriches the model by importing decades of empirical research from consumer satisfaction and post-purchase behavior studies.

Sixth, ECM-ISC’s focus on continuation (rather than initial adoption) implies different intervention strategies. Acceptance-focused interventions emphasize early impressions, initial training, and reducing adoption barriers. Continuance-focused interventions emphasize confirming expectations, maintaining satisfaction, and delivering sustained value. 6. Barriers Identification Section:

What Barriers to Technology Adoption does the model identify?

While ECM-ISC’s primary focus is information systems continuance rather than initial adoption, the model implicitly identifies barriers to sustained use (continuance barriers) that prevent continued adoption: Unconfirmed expectations represent a fundamental barrier to system continuance. When information systems do not perform as expected, when actual capabilities fall short of prior beliefs, or when anticipated benefits fail to materialize, users experience disconfirmation. This unmet expectation gap creates dissatisfaction and undermines continuance intentions. The barrier operates through disappointment—users who expected strong performance but encounter mediocre functionality, who expected easy systems but find them complex, or who expected substantial benefits but observe minimal improvement will discontinue use. The model’s empirical evidence demonstrates that confirmation is a significant predictor of both satisfaction and continued use intentions, indicating that unconfirmed expectations constitute a powerful barrier to continuance.

Low satisfaction and negative user experiences represent direct barriers to continued system use. Even when expectations are confirmed, users who have negative emotional responses to systems—who experience frustration, dissatisfaction, or irritation—will be inclined to discontinue. The model identifies satisfaction as a central driver of continuance intentions, suggesting that dissatisfactory user experiences constitute a significant barrier. Dissatisfaction might stem from poor interface design, inadequate system reliability, insufficient responsiveness to user needs, poor customer support, or generally frustrating interactions with systems. Perceived lack of usefulness represents another significant barrier. Systems users do not perceive as useful—that do not deliver tangible benefits, do not improve task performance, do not contribute to productivity, and do not provide clear value—face discontinuance risk. The model identifies perceived usefulness as an independent predictor of continuance intentions, suggesting that users who fail to see ongoing benefits from system use will discontinue.

This barrier is particularly significant when systems require sustained effort or time commitment to use; without perceived usefulness justifying this investment, users rationally discontinue. Availability of superior alternatives represents an implicit barrier in the continuance framework. While not explicitly measured in the model, the decision to continue using one system involves implicit comparison to alternatives. If users become aware of superior alternatives that would better meet their needs, are easier to use, or are less expensive, they face incentive to switch. The barrier operates through relative comparison—even systems providing confirmed expectations and adequate satisfaction may be abandoned if better alternatives become available. Inadequate expectation management represents an indirect barrier to continuance. Systems with marketing communications or pre-use descriptions promising capabilities that actual systems do not deliver create disappointment and disconfirmation.

Users expecting certain features or functionality might discontinue when they discover actual systems differ from expectations. This barrier stems not from system inadequacy per se but from misalignment between what systems promise and what they deliver. Declining perceived usefulness over time represents a barrier capturing the reality that users’ perceptions of system benefits may decline as they become familiar with systems. Initially, systems may seem remarkably useful; with continued use, novelty wears off and perceived usefulness may stabilize at lower levels. Users experiencing this perceived usefulness decline may discontinue if perceived benefits fall below psychologically acceptable thresholds.

What does the model instruct leaders to do in order to reduce these barriers?

ECM-ISC provides clear guidance for organizational leaders seeking to promote information systems continuance and reduce discontinuance barriers: To address unconfirmed expectations and expectation misalignment, leaders should implement careful expectation management strategies. Marketing communications, system descriptions, and user orientation should set realistic expectations about system capabilities, performance, and benefits. Documentation should clearly communicate what systems will and will not do. User training should prepare users realistically for their interactions with systems, avoiding overpromising or suggesting capabilities systems do not provide. As Bhattacherjee notes, managing expectations is not merely a marketing concern; it directly affects continuance decisions. Leaders should ensure consistency between expectations created through marketing communications and actual system performance. When capabilities or performance improvements occur, leaders should update user expectations to maintain confirmation and alignment.

To address low satisfaction and negative user experiences, leaders should prioritize creating positive, satisfactory user experiences. This involves investments in system design emphasizing usability, reliability, and responsiveness to user needs. Customer support should be readily available and responsive to user issues. System interfaces should be designed to be intuitive and pleasant to use. Leaders should actively monitor user satisfaction through surveys, feedback mechanisms, and usage analytics, identifying pain points and areas where negative experiences are occurring. When dissatisfaction is identified, leaders should take corrective action through system improvements, enhanced support, training, or service modifications. To address perceived lack of usefulness, leaders should focus on demonstrating and delivering clear value from system use. This involves several complementary strategies. First, system functionality should be designed to address genuine user needs and deliver demonstrable benefits.

Systems should solve real problems users face, improve task performance, or enable activities users value. Second, user training should help users recognize and leverage valuable capabilities they might otherwise overlook. Many systems provide underutilized features providing significant value if users understand how to use them. Training focusing on value-delivery helps users perceive usefulness they might not recognize independently. Third, organizational communication should emphasize and reinforce the benefits users derive from system use. Highlighting success stories, demonstrating ROI, and showcasing how systems contribute to work effectiveness or life improvement helps sustain users’ perceived usefulness. Fourth, system improvements should focus on increasing the value users derive, whether through new features, improved functionality, or integration with complementary tools. To address declining perceived usefulness over time, leaders should implement ongoing improvement and enhancement strategies.

Rather than assuming systems remain static and adequate once deployed, organizations should continuously enhance systems based on user needs and changing circumstances. Regular updates introducing valuable new functionality help maintain users’ perception that systems continue providing increasing value. Communication about improvements and enhancements reminds users of the ongoing value systems provide. To reduce the risk that superior alternatives will lure away users, leaders should work to maintain competitive advantage and value proposition. This involves monitoring competitive offerings, understanding emerging alternatives, and ensuring their systems remain competitive in terms of functionality, ease of use, customer support, and overall value. Building switching costs (such as data lock-in or integration with complementary systems) or developing strong customer relationships can also reduce discontinuance risk when alternatives become available.

To leverage the confirmation-satisfaction linkage, leaders should actively work to ensure actual system performance confirms and preferably exceeds user expectations. This requires ongoing attention to system reliability, functionality, and performance. Issues that undermine users’ perception that systems perform as expected should be addressed promptly. Leaders should communicate honestly about system limitations, ensuring users understand realistic performance parameters. When systems cannot meet certain expectations, leaders should acknowledge this transparently rather than allowing disappointment to arise when users discover limitations through use. The model suggests that information systems continuance is not guaranteed by initial adoption. Rather, organizations must actively manage user experience, maintain realistic expectations, deliver satisfaction, and demonstrate ongoing value to sustain continued system use. Practitioners should recognize that post-adoption management is as important as adoption management in determining system success. 7.

• Following Models or Theories: Following Models: Extended ECM models incorporating additional variables (such as switching costs, habit, social influences) Models examining continuance across diverse information systems (social media, productivity software, enterprise systems) Technology Abandonment models Digital Engagement and Sustained Use models Following Theories: Research on information systems habit and behavioral inertia Studies of customer loyalty in digital contexts Discontinuance and switching behavior research Service continuance models in electronic commerce Series Navigation This article is part of a Technology Adoption literature review series: 1
• A Model of Adoption of Technology in Households: Brown and Venkatesh, 2005 2
• Understanding Information Systems Continuance: An Expectation- Confirmation Model (Bhattacherjee, 2001) 3. Status Quo Bias in Decision Making (Samuelson and Zeckhauser, 1988) References 1.Ajzen, I. “The Theory of Planned Behavior.” Organizational Behavior and Human Decision Processes 50, no. 2 (1991): 179-211. 2.Bhattacherjee, A. “Understanding Information Systems Continuance: An Expectation-Confirmation Model.” MIS Quarterly 25, no. 3 (2001): 351- 370. 3.Davis, F. D. “Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology.” MIS Quarterly 13, no. 3 (1989): 319-340. 4.Davis, F. D., Bagozzi, R. P., and Warshaw, P. R. “User Acceptance of Computer Technology: A Comparison of Two Theoretical Models.” Management Science 35, no. 8 (1989): 982-1003. 5.Oliver, R. L. “A Cognitive Model of the Antecedents and Consequences of Satisfaction Decisions.” Journal of Marketing Research 17, no. 4 (1980): 460-469. 6.Mathieson, K. “Predicting User Intentions: Comparing the Technology Acceptance Model with the Theory of Planned Behavior.” Information Systems Research 2, no. 3 (1991): 173-191