Service quality improvement means changing how a service is designed and delivered so it meets defined customer needs more dependably. The right method depends on the problem: unclear expectations, recurring errors, lost handoffs, unnecessary waiting or inconsistent execution.
These seven methods offer different starting points. Each includes a practical service example, a first artifact and a limitation. The examples are illustrative proposals, rather than reported company results or promises that a tool will improve satisfaction automatically.
Compare seven service-quality improvement methods
Begin with an observed problem and the decision you need to make. A survey can identify a perceived shortfall; a process investigation is still needed to establish why it happens. A map can expose a handoff; it does not establish that the proposed redesign works.
| Method | Question it supports | First useful artifact | Important requirement |
|---|---|---|---|
| SERVQUAL | Where do expectations and perceptions differ? | Defined survey instrument and dimension report | Appropriate wording, sampling and interpretation |
| Six Sigma | What drives a measurable service defect or variation? | DMAIC charter and trustworthy baseline | Reliable measurement and suitable analytical skills |
| Service blueprinting | Which delivery dependencies support the customer’s experience? | Frontstage, backstage and support-process map | Customer evidence and cross-team process knowledge |
| Customer journey mapping | Where does a customer struggle to achieve a goal? | Evidence-based journey for an actor and scenario | Research with relevant customers |
| Lean service management | Which work or waiting fails to support the intended value? | Observed current flow and improvement proposal | End-to-end observation and required-control review |
| Total Quality Management | How will quality improvement become sustained shared work? | Quality priorities, owners and review process | Leadership commitment, staff involvement and resources |
| Mystery shopping | What happens in a defined test scenario? | Scenario, observation criteria and sampling plan | Realistic evaluation, trained shoppers and evidence review |
The methods overlap and can be combined. Journey mapping can identify a customer problem; blueprinting can show the delivery dependencies; DMAIC can structure investigation and evaluation. Use the minimum combination that answers the question, with one clear improvement owner.
1. Use SERVQUAL to investigate expectations and perceptions
SERVQUAL examines service quality through paired expectation and perception questions. The Marketing Science Institute’s original working-paper summary identifies five dimensions: tangibles, reliability, responsiveness, assurance and empathy. Select and document the instrument version when implementing a formal study.
For a digital service, those dimensions can inspire investigation into the presentation of service materials, dependable delivery, help and response, confidence in advice, and attention to an individual’s circumstances. These are practical adaptations, rather than a reproduction of the validated questionnaire.
Use it when: Customers report that the service differs from what they expected, but a general satisfaction score does not identify the dimension to investigate. Keep the service, audience and timeframe specific, and check that participants understand the questions.
Illustrative example: A service team studies whether customers receive updates as expected. In a small local exercise using the same valid paired responses to seven-point questions with consistent anchors, the mean expectation rating is 6 and the mean perception rating is 4. A defined perception-minus-expectation difference is 4 − 6 = −2 scale points. It is a local indicator for investigation, not a percentage or an industry benchmark.
Review the paired responses, distribution, response coverage and comments before deciding what to change. An expectation shortfall might involve an unrealistic promise, missed updates or different interpretations of timing. Interview customers and inspect actual commitments to distinguish those explanations.
First artifact: A survey specification and dimension report, including wording, scale, sample, scoring and missing-response rules. An adapted few-question exercise should be labeled as such; it is not automatically equivalent to a formal SERVQUAL score or comparable across unrelated services.
2. Use Six Sigma for a defined recurring defect
ASQ describes Six Sigma as a disciplined approach to process capability and variation. In a service, define the customer requirement and the failure against it instead of calling every long conversation a defect.
Use it when: A recurring process has a measurable quality problem and enough reliable evidence to investigate variation. Agree the unit, eligible population, defect definition and measurement rules before comparing rates.
The DMAIC approach organizes the work into Define, Measure, Analyze, Improve and Control: scope the problem, establish a trustworthy baseline, investigate contributing causes, evaluate a change and maintain a reaction plan.
Illustrative example: A billing team reviews 100 eligible invoices issued during a defined month and confirms that 12 have an incorrect tax code. The affected-invoice rate is 12 ÷ 100 × 100 = 12%. Counting each erroneous line item would answer a different question. The rate alone does not identify why the code was wrong.
Trace successful and failed cases to examine inputs, mapping rules and exception handling. Test the suspected explanation before proposing a control. A pilot might revise validation at the point where the error enters, then examine incorrect codes, legitimate exceptions and extra processing work.
First artifact: A project charter with a verified baseline and investigation plan. Choose analytical support appropriate to the data; a small descriptive review is not proof of Six Sigma process capability. Keep an owner for monitoring and responding to recurrence.
3. Use service blueprinting to connect visible service with delivery
A service blueprint shows how the customer’s actions connect with visible employee actions, backstage work and supporting processes. Nielsen Norman Group’s definition ties those relationships to a specific customer journey.
Use it when: A customer-facing promise depends on several teams or systems, and ownership or information is lost between them. Map actual cases and exceptions with the people doing the work, rather than copying the ideal procedure.
Illustrative example: A customer requests a replacement. The frontstage action is the agent’s confirmation; backstage work includes checking eligibility and preparing a warehouse request; supporting processes include inventory and delivery updates. The blueprint reveals that the customer receives a dispatch promise before stock is confirmed.
A redesign proposal could delay that promise until confirmation and give an earlier, accurate status message. Review who confirms stock, how the agent receives it and what happens when stock is unavailable. A clearer diagram alone does not repair that dependency.
First artifact: A blueprint with evidence, dependencies, failure points and named owners. Separate a communication improvement from a fulfillment improvement, and evaluate both. Our journey-mapping method explains how customer research can supply the starting journey.
4. Use customer journey mapping to identify meaningful touchpoints
A journey map represents a particular person’s goal and experience over time. Nielsen Norman Group’s journey components include an actor, scenario, phases, actions and reported thoughts or emotions, plus opportunities and ownership. A journey map and service blueprint complement each other.
Use it when: Internal teams understand their tasks but do not know where the customer encounters difficulty across the overall experience. Define the scenario and recruit relevant participants, including people who did not complete the task.
Illustrative example: A first-time user wants to produce a usable report. They read a guide, attempt an import, encounter a permission limit, ask for help and wait for an administrator. Research might reveal that the guide does not explain the prerequisite role. Treat that as a finding only after observing it or hearing supported accounts.
Distinguish evidence from hypotheses. Do not draw an emotional low point because the team assumes everyone dislikes a stage. Preserve different paths and relevant exceptions rather than forcing every customer into awareness, purchase and loyalty.
First artifact: A scoped, evidence-based map with an opportunity assigned to an owner. Use the journey template to record the work. Check whether the changed guidance helps the intended users complete the task, not just whether the diagram is approved.
5. Use Lean service management to improve flow and remove avoidable work
The Lean Enterprise Institute’s principles connect customer value, the value stream, flow, pull and ongoing improvement. For service work, examine queues, repetition, unnecessary processing, avoidable transfers and rework.
Use it when: Customers wait between otherwise functional activities, or several teams repeatedly enter or request the same information. Observe end-to-end work before deciding which activity is unnecessary.
Illustrative example: An appointment request is copied from a form into two spreadsheets before confirmation. Record who uses each copy and why. A proposed single authoritative record may remove repetition, but it still needs access, ownership, exception handling and an appropriate retention process.
Record processing and waiting separately. Faster work in one team may only enlarge another team’s queue. Include error correction, customer effort and total completed work when evaluating the future flow.
Required checks can serve safety, privacy, access or financial controls even when customers do not see them. Preserve their purpose while considering a simpler design. Removing every check or maximizing employee utilization is not the same as improving service flow.
First artifact: An observed current-state flow and a bounded future-state proposal. Involve staff who perform the work and review whether the change shifts effort or risk elsewhere. The business process methods guide covers broader method selection.
6. Use Total Quality Management to sustain shared improvement
ASQ describes TQM as a customer-focused management approach involving employees, processes, data, communication and continual improvement. It concerns organizational practice, rather than a single survey or a one-time training event.
Use it when: Teams repeatedly correct individual cases while the same underlying service problem returns. Leadership needs to give improvement work priorities, resources and a reliable decision route across functions.
Illustrative example: Support keeps explaining a confusing renewal policy. Product, billing and support review the relevant cases, clarify the policy presentation and update guidance together. An owner checks later cases for the same misunderstanding and records whether the change actually helped.
Invite staff to raise recurring problems and give submissions a response. Distinguish individual mistakes from unclear policy, missing access and conflicting objectives. Involve customers through appropriate research and feedback; shared quality goals should reflect a real need.
First artifact: A quality-priority record with cross-team owners, actions, resources and a review cadence. Keep standards and guidance current after changes. Suggestion counts and workshops indicate activity; examine completed actions and customer outcomes. Use continuous improvement examples for smaller experiments within that operating approach.
7. Use mystery shopping to inspect a defined service scenario
Mystery shopping uses evaluators to experience a specified service scenario and report what happened. It can examine digital support, phone service or physical locations. It is designed research with selection and evaluator limits, rather than an inherently unbiased view of every customer’s experience.
The MSPA Europe/Africa guidelines address scenario design, shopper selection, sample and rotation, questionnaires, quality control and reporting. They recommend factual observations as the basis for scored criteria, with clearly defined subjective comments providing context.
Use it when: You need to check whether a documented service standard is delivered in a particular situation. Use a realistic task, suitable evaluators and a defined channel and timing plan. Establish the evaluation and recording arrangements through the responsible organizational process.
Illustrative example: An evaluator asks how to change an appointment under the actual policy. The worksheet records whether the agent explains the available options, checks the required information and confirms the next step. “Explained both available options” is an observable criterion; “seemed enthusiastic” is an interpretation needing context.
Train evaluators, pilot the worksheet and review contradictory or incomplete records. A narrow set of scenarios does not establish a failure rate for the whole service. Report which scenarios, times and channels were tested and which were absent.
First artifact: A scenario and evidence checklist with a sampling and reporting plan. Keep access to personal records appropriate, let relevant staff respond to disputed evidence and use multiple evidence sources for consequential performance decisions. The MSPA guidance specifically cautions against isolated mystery-shopping results as the basis for disciplinary decisions.
Run a focused pilot and maintain the improvement
Select one service problem and the method that supplies the missing evidence. Confirm the owner can act and identify dependencies before collecting a large dataset or launching another tool.
Service quality improvement brief Service, audience and customer requirement: Observed failure or gap, evidence and uncertainty: Selected method and reason: First artifact and responsible owner: Baseline unit, denominator, period and case mix: Proposed change, resources and required controls: Pilot scope and comparison approach: Customer outcome, quality and workload measures: Review date, decision and attribution limits: Updated standard, maintenance owner and reaction plan:
Pair the intended outcome with a quality safeguard. For a routing change, inspect correct ownership and repeat questions alongside time. For clearer instructions, check task completion and accuracy. For a survey finding, inspect the service behind the rating rather than only trying to raise the number.
Record the baseline definition, eligible population and comparison window. Account for changed issue mix, staffing, policy or product behavior. Before-and-after results alone do not establish that the selected method caused the difference. Keep missing evidence visible and decide whether to keep, revise or stop the change.
Support QA can review interaction standards and coaching needs alongside this work. It serves a different purpose from selecting a method to redesign service delivery.
Use technology for an identified evidence or communication need. Screendesk provides recording and live video workflows that may help explain a visible issue. Record only appropriate information and provide written context. Check the underlying cause separately from the visible behavior. Evaluate the proposed workflow instead of forecasting an unsupported improvement from installing a tool.
