Explaining consumers' channel-switching behavior in the post-COVID-19 pandemic era

In December 2019, COVID-19 was reported in Wuhan, China, and subsequently spread globally. With non-pharmaceutical interventions implemented widely, consumers spent more time online due to lockdowns. Southeast Asia, including Vietnam, experienced large shifts to digital consumption; Vietnam’s internet economy grew 36% in 2020 and e-commerce revenue reached about US$11.8 billion (~5.5% of retail sales). Online shopping, enabled by ICT, expanded alongside changes in consumer behaviour accelerated by the pandemic. Although prior studies examined pandemic-era behaviours, few investigated consumers’ channel-switching behaviour in the post-COVID-19 era. This study explores channel-switching behaviour resulting from the outbreak, using over 2,640 respondents to identify determinants of changing shopping channel choices. Factors including marital status, price, quality, convenience, and satisfaction with existing and new channels influence switching decisions; service quality, price, and convenience are salient, and satisfaction with an existing channel affects whether consumers switch. The study also contrasts determinants for online versus traditional channels post-COVID-19, addressing: (i) which factors significantly affect channel switching after the pandemic outbreak, and (ii) which factors significantly affect prioritization of online versus conventional channels during the pandemic. The paper proceeds with literature synthesis, methodology and data, findings, and conclusions.

Theoretical framework: Multiple behavioural theories explain individual behaviour change: belief-attitude theories (protection motivation theory, health belief model, TRA, TPB), control-based theories (self-determination theory), decision-making theories (stages-of-change), and competence-based theories (social cognitive theory). This study primarily applies belief-attitude theories to explain channel switching because purchasing behaviours are formed from beliefs and environmental stimuli. Theory of Planned Behaviour (TPB): Extends TRA with perceived behavioural control (PBC)—the belief in one’s capability to perform a behaviour—improving prediction of actual behaviour. TPB has been widely validated for understanding behaviours. Expectation Confirmation Theory (ECT): A cognitive theory (Oliver, 1977) for consumer satisfaction and post-purchase behaviour. Consumers form expectations pre-purchase; after consumption they assess perceived performance and compare it to expectations (confirmation/disconfirmation). Positive confirmation leads to higher satisfaction and repurchase intentions. Unified framework (planned and confirmation theory): The study proposes a two-stage framework integrating TPB and ECT. Stage 1: Intention to use a channel is shaped by PBC (e.g., convenience, time, price, product quality) and subjective norms (e.g., information abundance, safety), while expectations influence attitudes (e.g., trust). Stage 2: Decision to reuse is driven by satisfaction via confirmation of expectations. Exogenous factors (consumer traits, situational factors, product characteristics, prior online experience, trust) also influence attitudes and intentions. Research questions: (1) After COVID-19’s outbreak, which factors significantly affected consumers’ channel-switching behaviour? (2) During the pandemic, which factors significantly affected prioritization of online purchasing channels? (3) During the pandemic, which factors significantly affected prioritization of traditional channels (markets/supermarkets)?

Design and data: A structured questionnaire (five-point Likert scale) was developed from the planned and confirmation theory framework to capture behavioural variables (e.g., convenience, information abundance, availability, freshness, price level, price vs. other channels, product quality assessment, service quality, time spending, trust, tech competency) and demographics (gender, age, education, income, marital status). Data were collected in Vietnam via email and follow-up phone calls from December 2021 to March 2022. Over 5,000 individuals were contacted using stratified random sampling to ensure representation across purchasing channels. After data cleaning, 2,647 valid responses remained (~56% response rate). The sample demographics broadly reflect Vietnam’s population structure (ages largely 22–50; incomes mainly 10–25 million VND per month). Analytic approach: Descriptive statistics explored changes in channel choices pre- and post-outbreak. Logistic regression models were then estimated due to binary construction of outcomes: (a) channel-switching behaviour (switch vs. not), and (b) channel choices post-outbreak (online vs. others; traditional market & supermarket vs. others). Explanatory variables included: (1) respondent profile (education, age, income, gender, marital status, technology competence), (2) perceived channel benefits/characteristics (convenience, price, service quality, overall satisfaction), and (3) channel selection barriers (time, trust/safety, selection range/availability, information, technology, freshness). Ordinal predictors were treated as equally spaced continuous variables. Model form: logit(p/(1−p)) = β0 + β1·(respondent profile) + β2·(channel benefits/characteristics) + β3·(barriers). Descriptive findings showed a marked shift from traditional/supermarkets toward online channels after the outbreak.

Descriptive shifts in channel choice: - Before COVID-19: Traditional/small markets were dominant (49%, 1,296/2,647), supermarkets 24% (637), convenience/grocery stores 14% (370), online 13% (344). - After outbreak: Online became most popular (31%, 827), traditional markets 27% (703), supermarkets 16% (430), convenience/grocery stores 11% (294). Determinants of channel-switching behaviour (Table 4, logit; N=2,647; Pseudo R2=0.158): - Overall satisfaction with the new channel: coef 0.948, marginal effect 0.188, p<0.01 (strongest driver, positive). - Convenience: 0.305, 0.060, p<0.01 (positive). - Service quality: 0.250, 0.050, p<0.01 (positive). - Price level (perceived higher price): −0.225, −0.045, p<0.01 (higher prices deter switching to the new channel). - Tech competency: 0.098, 0.019, p<0.01 (positive). - Married: 0.196, 0.039, p<0.05 (married consumers more likely to switch). Non-significant: safety, product quality assessment, gender, education, income. Determinants of post-outbreak channel choices (Table 5; N=2,647): - Online channel (Pseudo R2=0.158): • Convenience: 0.215, 0.043, p<0.01 (+) • Service quality: 0.157, 0.031, p<0.01 (+) • Product quality assessment: −0.259, −0.051, p<0.01 (inability to assess quality reduces online choice) • Price level: −0.316, −0.063, p<0.01 (higher perceived price reduces online choice) • Price vs. other channels: 0.246, 0.049, p<0.01 (relative price advantage increases online choice) • Abundant information: 0.123, 0.024, p<0.01 (+) • Freshness of goods: 0.089, 0.018, p<0.10 (+) • Tech competency: 0.108, 0.021, p<0.05 (+) • Education level: 0.225, 0.045, p<0.10 (+) • Income: 0.074, 0.015, p<0.05 (+) • Non-significant: trust, time spending, availability of goods, gender, marital status. - Traditional market & supermarket channels (Pseudo R2=0.115): • Convenience: −0.082, −0.018, p<0.05 (higher convenience preference lowers traditional choice) • Service quality: 0.171, 0.038, p<0.01 (+) • Time spending: −0.155, −0.035, p<0.01 (more time required lowers traditional choice) • Availability of goods: −0.071, −0.016, p<0.10 (−) • Price level: −0.196, −0.044, p<0.01 (higher perceived price lowers choice) • Price vs. other channels: 0.126, 0.028, p<0.01 (+) • Freshness of goods: 0.109, 0.024, p<0.05 (+) • Non-significant: trust, product quality assessment, tech competency, gender, education; marital status shows a small negative marginal effect (−0.031, p<0.10); income non-significant. Overall: Switching is driven by higher perceived service quality and convenience, lower prices, greater satisfaction with the new channel, higher tech competence, and being married. Online choice is promoted by convenience, service quality, information abundance, relative price advantage, freshness, tech competence, higher education and income; it is hindered by higher prices and the inability to assess product quality. Traditional choice benefits from service quality and freshness but is deterred by higher time cost, higher prices, and, for consumers emphasizing convenience, by that preference. Trust was not a significant determinant for either channel choice.

The findings address the research questions by evidencing that, post-outbreak, switching behaviour is strongly influenced by channel satisfaction, service quality, convenience, price perceptions, technology competence, and marital status. The unified planned and confirmation theory framework helps explain intentions (via perceived control, norms, and expectations) and repeat behaviour (via satisfaction/confirmation). Determinants of channel prioritization differ across channels: online behaviour is enhanced by convenience, abundant information, tech competency, and competitive pricing but constrained by challenges in assessing product quality and higher price levels; traditional channels benefit from perceived service quality and product freshness yet suffer from higher time costs and reduced convenience. Trust, often central in e-commerce literature, did not significantly influence channel choice in this context. Managerially, operators should enhance service quality, streamline convenience, manage total price (including shipping), provide rich product information, support product quality assessment (e.g., clear specs, reviews, easy returns), and improve user experience to accommodate varying levels of technological competence. Freshness remains a key lever for traditional food/agricultural purchases.

The study documents a substantial shift toward online purchasing after COVID-19’s outbreak and quantifies drivers of both channel-switching and channel-selection behaviours. Switching is influenced by service quality, convenience, price levels, overall satisfaction with the new channel, technological proficiency, and marital status, with satisfaction being particularly pivotal. Post-outbreak channel choices are shaped by convenience, service quality, product quality assessment ability, time spent, price levels and relative prices, information abundance, freshness, technological competency, and education level; trust was not a significant determinant. The results provide actionable insights for e-commerce and traditional retailers to bolster purchase intentions and tailor strategies in a post-pandemic environment.