Employee Engagement Survey Questions: How to Design a Survey You Can Act On
January 19, 2024 · Human Resources · 9 min read
An employee engagement survey can be useful, but only if the organisation is clear about what it is measuring, protects employee trust, uses appropriate questions and follows the results with credible action.
A weak survey collects opinions without a measurement model. A stronger survey distinguishes between validated engagement instruments, custom diagnostic questions and broader employee-experience questions.
This guide explains how to design a useful engagement survey without pretending that any list of homemade questions is a scientifically validated engagement scale.
First decide what you are trying to measure
Organisations often combine several different topics under the label “engagement”. These can include:
- engagement;
- job satisfaction;
- manager effectiveness;
- workload;
- leadership trust;
- career development;
- recognition;
- wellbeing;
- hybrid work;
- intention to stay;
- organisational culture.
These topics are related, but they are not interchangeable. Decide which outcomes and workplace conditions matter before writing questions.
Validated instruments and custom questions serve different purposes
Gallup’s Q12 is an example of a validated engagement instrument developed through decades of research and psychometric testing. Gallup distinguishes between actionable workplace conditions and broader attitudinal outcomes.
Source: Gallup Q12 Meta-Analysis, 11th Edition
Validated instruments can provide stronger measurement properties and, where licensing permits, benchmark comparisons.
Custom questions can still be valuable for diagnosing organisation-specific issues, but a custom questionnaire should not automatically be described as a validated engagement scale.
Important: proprietary survey instruments such as Gallup Q12 are protected intellectual property. Do not copy or administer proprietary question sets without the appropriate permission or licence.
Use a clear survey architecture
A practical engagement survey can contain four layers:
- Core engagement measure: a validated instrument or clearly defined internal measure.
- Driver questions: workplace conditions managers or leaders can influence.
- Context questions: current organisational issues such as change, workload or systems.
- Open comments: space for employees to explain what the numbers may not show.
This structure helps separate measurement from diagnosis.
Useful domains for diagnostic questions
If you are writing organisation-specific questions, common domains include:
- role clarity;
- resources and tools;
- manager support;
- feedback;
- recognition;
- voice and participation;
- teamwork;
- trust;
- learning and development;
- career opportunity;
- purpose;
- workload;
- change communication;
- fairness;
- systems and processes.
Gallup’s engagement research also emphasises actionable conditions such as clarity, resources, recognition, development and purpose. Use the underlying management concepts, not proprietary wording you do not have permission to reproduce.
Source: Gallup, Employee Engagement
Sample custom questions for internal diagnosis
The following are illustrative custom questions. They are not presented as a validated engagement scale.
Role clarity
- I understand the most important outcomes expected from my role.
- When priorities conflict, I know which work should come first.
Resources and systems
- I have access to the information and tools required to perform my work effectively.
- Our processes allow me to complete important work without unnecessary obstacles.
Manager support
- My manager gives me useful feedback that helps me improve.
- I can raise a work problem with my manager without expecting an unfair response.
Recognition
- Good work is acknowledged in a way that feels specific and credible.
- Recognition in my team is distributed fairly.
Development
- I understand which skills or experiences would help me progress.
- I have realistic opportunities to develop capabilities relevant to my role or career.
Voice
- Employees can raise concerns or ideas through channels that leaders take seriously.
- When employees give feedback, the organisation explains what action will or will not follow.
Workload
- My workload is generally achievable within reasonable working arrangements.
- When capacity is insufficient, priorities are adjusted rather than simply added.
Change
- I understand how current organisational changes affect my work.
- Leaders explain the reasoning behind important changes clearly enough for me to act.
Write one idea per question
Avoid double-barrelled questions such as:
“My manager communicates clearly and gives me enough development opportunities.”
An employee may agree with one half and disagree with the other. Split them into separate items.
Avoid leading questions
Do not write questions that tell employees what the organisation wants to hear.
Weak:
“How helpful is our excellent new wellbeing programme?”
Better:
“To what extent has the new wellbeing programme improved your access to useful support?”
Use a consistent response scale
Agreement scales are commonly used because they are easy to interpret, but consistency matters.
For example, an internal survey might use:
- Strongly disagree
- Disagree
- Neither agree nor disagree
- Agree
- Strongly agree
- Not applicable
Do not switch between 1-5, 1-10 and yes/no without a clear reason. Inconsistent scales create avoidable analysis problems.
Include “not applicable” when the question may genuinely not apply
Forcing employees to rate something they have not experienced can create noise.
Examples:
- employees with no direct manager during a restructuring;
- employees who have not used a particular benefit;
- new hires who have not yet experienced a promotion process.
Keep demographic questions purposeful
Demographic data can reveal whether experiences differ across locations, job levels or groups. It can also create privacy risk.
Ask only for data you have a clear use for, and consider whether combinations of fields could identify individuals in small teams.
Anonymity and confidentiality are different
Anonymous means responses cannot reasonably be connected back to an individual. Confidential means identifiable data may exist but access is restricted.
Do not promise anonymity if raw data, open comments, login identifiers or small-group reporting could reveal the respondent.
CIPD’s employee-voice examples emphasise secure handling of survey data, limited access and transparency about who can see results.
Set minimum group sizes for reporting
Team-level reporting can become identifying when only a handful of employees are in the group.
Choose a minimum reporting threshold appropriate to the organisation, survey platform and sensitivity of the questions. Open-text comments often require additional care because writing style and specific incidents can identify people even when names are removed.
Do not force participation if you call the survey confidential or voluntary
Pressure to participate can undermine trust. Explain:
- why the survey is being run;
- how results will be used;
- who can access raw data;
- whether participation is voluntary;
- how privacy is protected;
- when employees will hear what happened next.
Survey length should match the decision you need to make
A long annual survey may be justified when the organisation needs a broad diagnostic. A short pulse survey may be better for tracking a specific issue.
Do not ask 80 questions because the platform allows it. Every item should connect to an analysis or decision.
When to use a pulse survey
Pulse surveys are useful for:
- checking whether a change is landing as intended;
- tracking one or two priority issues;
- following up after an action plan;
- monitoring a fast-moving employee-experience issue.
A pulse survey should not simply repeat the full engagement survey every month.
Do not over-survey employees
Survey fatigue is not only about frequency. Employees become fatigued when they repeatedly provide feedback and see no credible response.
The strongest antidote is action and communication.
Use open-ended questions selectively
Open text can reveal issues that fixed-response items miss, but large volumes of comments require careful analysis and privacy handling.
Useful open questions include:
- What is one thing that helps you do your best work here?
- What creates the most unnecessary friction in your work?
- What is one change that would most improve your team’s effectiveness?
- What should the organisation keep doing?
Avoid asking several overlapping open questions if the organisation has no capacity to analyse the responses.
Do not treat an average score as the whole story
A company average can hide meaningful variation.
Where sample sizes and privacy permit, examine patterns by:
- business unit;
- team;
- job level;
- location;
- tenure;
- work arrangement;
- other relevant employee groups.
Look for patterns, not isolated differences that may reflect very small sample sizes.
Trend data can be more useful than one benchmark
External benchmarks can provide context, but your own trend over time may be more actionable.
Ask:
- Which areas improved?
- Which declined?
- Where did action occur?
- Which teams show a different pattern?
- Did operating conditions change between survey periods?
Do not imply causation from a survey alone
If teams with high engagement also have higher performance, the survey alone does not prove that one caused the other in your organisation.
Use survey data alongside operating data, qualitative evidence and appropriately designed analysis.
Manager results should lead to conversation, not punishment
If managers believe every low score will be used as a performance penalty, they may focus on influencing the score rather than improving the workplace.
Results should be used to understand conditions, identify priorities and support action. Serious management problems still require accountability, but survey scores need interpretation.
A practical engagement survey workflow
- Define the objective. What decision or diagnosis is the survey meant to support?
- Choose the measurement approach. Validated instrument, internal diagnostic, or both.
- Select domains. Include only topics relevant to the objective.
- Draft and review questions. Remove ambiguity, double questions and leading language.
- Test the survey. Pilot wording and technical flow with a small group.
- Define privacy rules. Minimum group sizes, access and comment handling.
- Launch with clear communication.
- Analyse patterns. Quantitative and qualitative.
- Diagnose before acting. Follow up where results need explanation.
- Create an action plan.
- Report back to employees.
- Track whether conditions improve.
Turn results into action
The survey is not the end product. It is an input into management action.
MATSH’s Employee Engagement Action Plan guide explains how to prioritise findings, investigate causes, assign owners and communicate progress.
Frequently asked questions
How many questions should an employee engagement survey have?
There is no universal ideal number. Use enough questions to measure the intended domains reliably without collecting data you will not analyse or act on.
Should engagement surveys be anonymous?
Anonymous surveys can encourage candour, but anonymity must be genuine. If the organisation can identify respondents through metadata, small groups or comments, describe the process as confidential rather than anonymous.
Can I copy Gallup Q12 questions into my own survey?
Gallup states that Q12 items are proprietary and protected. Use them only under the appropriate terms or licence. Do not reproduce proprietary instruments simply because the questions can be found online.
Are custom questions valid?
Custom questions can be useful for diagnosis, but they are not automatically a validated engagement scale. Be precise about what the survey measures.
How often should employee surveys run?
Frequency should reflect the purpose. A broad survey may run annually or periodically, while shorter pulse surveys can track specific issues more frequently. Avoid collecting feedback faster than the organisation can respond.
What should happen after the survey?
Share the main findings, investigate causes, choose a small number of priorities, assign actions and report progress. Repeated surveying without visible follow-through can damage trust.
Sources
Need In-House Training?
We run all our courses as private programmes for organisations across the GCC and Africa.
Request In-House →