SERVQUAL Service Quality Scale

SERVPERF (Cronin & Taylor, 1992) measures only perceived service performance without the expectation component, simplifying data collection but sacrificing the gap model insight. SERVQUAL's dual-administration reveals where expectations exceed perceptions, whereas SERVPERF yields only an overall performance score. SERVQUAL is preferred when understanding expectation-reality mismatches is strategic; SERVPERF when survey brevity is critical.

Yes. Retain the five-dimension framework and gap model logic, but revise the 22 items to reflect your service's specific touchpoints. For example, higher education uses HEdPERF (Srikanthan & Dalrymple, 2003), which adapts SERVQUAL items to academic contexts. Pre-test adapted items with customer focus groups to ensure they remain valid and relevant.

This indicates that, while overall quality meets expectations, that specific dimension is a weakness relative to customer expectations. For example, a hotel might have positive overall SERVQUAL but negative gaps in Responsiveness, signaling that speed of service lags customer expectations even if other dimensions excel. Target that dimension for improvement to prevent it from becoming a competitive vulnerability.

A minimum of 100–150 respondents per service location or segment is recommended to achieve stable dimension-level estimates. For multi-location benchmarking, 200–300 per site ensures reliable gap comparisons. Larger samples (500+) allow segmentation analysis (e.g., by customer tenure or usage frequency) and increase power for detecting meaningful changes over time.

Frequency depends on service volatility and organizational goals. Monthly or quarterly administrations capture short-term fluctuations and staff training impacts; annual surveys suffice for stable, mature services. After major service redesigns, conduct SERVQUAL at baseline, 3 months, and 6 months post-implementation to track convergence toward improved quality expectations.