Employee Survey Best Practices
How to design, run, and act on employee surveys that actually make a difference.
Before you survey: Planning
Define your purpose
Every survey should have a clear purpose. Ask yourself: What decisions will this survey inform? If you can't answer that, you're not ready to survey.
Common purposes include:
- Baseline measurement of engagement levels
- Tracking progress on specific initiatives
- Identifying priority areas for improvement
- Understanding employee experience during change
- Gathering feedback on specific policies or programmes
Get leadership commitment
Before launching a survey, ensure leadership is committed to:
- Actively promoting participation
- Reviewing results with openness
- Taking visible action on findings
- Communicating back to employees
A survey without leadership commitment is worse than no survey—it erodes trust.
Choose your timing
Avoid surveying during:
- Major holidays or leave periods
- Year-end crunch times
- Immediately after traumatic events (retrenchments, etc.)
- During system changes that might affect access
For South African organisations, December-January is typically a poor time. February-March or August-September often work well.
Survey design
Keep it focused
The most common mistake is asking too many questions. Every question should:
- Relate to something you can actually influence
- Provide information you'll act on
- Be clearly worded and easy to answer
Engagement surveys: 40-60 questions is typically sufficient
Pulse surveys: 5-10 questions maximum
Use validated questions
Don't reinvent the wheel. Use questions that have been tested and validated to measure what they're supposed to measure. This also enables benchmarking.
Resist the temptation to add "one more question" from every stakeholder—this is how surveys become unwieldy.
Balance quantitative and qualitative
Rating questions give you measurable data. Open-ended questions give you context and depth. Include both.
Recommended: 2-3 open-ended questions for engagement surveys, 1 for pulse surveys.
Consider your workforce
In South Africa, consider:
- Language — Offer surveys in multiple languages if your workforce isn't all English-fluent
- Access — How will frontline workers without computers complete the survey? (SMS, kiosks, paper options)
- Literacy — Keep language simple and avoid jargon
- Context — Are there site-specific or department-specific issues that need addressing?
Maximising participation
Why participation matters
Low participation (<60%) creates two problems: statistically unreliable results, and the perception that the survey isn't representative. Aim for 70%+ overall, with no group below 60%.
Communicate the purpose
Employees need to understand why they should participate. Communicate:
- What the survey is about
- How results will be used
- What happened from the last survey (if applicable)
- How anonymity is protected
Leadership endorsement
A message from the CEO or MD explaining why the survey matters has significant impact. Follow up with messages from line managers closer to the ground.
Make it easy
- Mobile-friendly surveys that work on any device
- Clear instructions and estimated completion time
- Save-and-continue functionality for longer surveys
- Multiple reminder channels (email, SMS, internal comms)
Address anonymity concerns
Fear of identification is the biggest barrier to honest feedback. Be transparent about:
- Minimum group sizes for reporting (typically 5+)
- What demographic data is collected and why
- Who has access to responses
- How open-ended comments are handled
Using an external provider like EngagePulse adds credibility to anonymity claims.
After the survey: Taking action
Share results quickly
Don't sit on results for months. Communicate key findings within 2-3 weeks of survey close. Employees are waiting to see if their participation was worthwhile.
Be transparent
Share both strengths and areas for improvement. Employees know things aren't perfect—pretending otherwise destroys credibility. Acknowledge challenges and explain what you're doing about them.
Focus on priorities
You can't fix everything at once. Select 2-3 priority areas based on:
- Impact on overall engagement
- Feasibility of improvement
- Alignment with business strategy
- Employee expectations (what did they emphasise in comments?)
Involve managers
Equip managers to discuss results with their teams and develop local action plans. Provide:
- Team-level reports (where sample sizes allow)
- Guidance on how to facilitate conversations
- Templates for action planning
- Support from HR or OD if needed
Close the loop
The biggest predictor of future participation is whether employees saw action from the last survey. Communicate regularly about progress on action items. Even when progress is slow, communicate that work is ongoing.
Common mistakes to avoid
Survey fatigue
Problem: Too many surveys, too often, with no visible outcomes
Solution: Consolidate surveys, coordinate timing across the organisation, and always act on results
Analysis paralysis
Problem: Spending months analysing data before taking any action
Solution: Identify quick wins you can act on immediately while deeper analysis continues
Blaming the messenger
Problem: Getting defensive about negative feedback or trying to explain it away
Solution: Treat all feedback as data. Perception is reality—if employees perceive a problem, there's a problem
One-size-fits-all responses
Problem: Implementing the same actions everywhere regardless of local context
Solution: Allow for local customisation of responses while maintaining organisation-wide consistency on key issues
Ignoring the positive
Problem: Focusing only on what's wrong and ignoring what's working
Solution: Celebrate strengths and understand what's driving them so you can replicate success
Want to run better employee surveys?
EngagePulse brings 15+ years of survey expertise to help you avoid common pitfalls and get results that drive action.
Book a DemoRelated resources
- The Complete Guide to Employee Engagement
- How to Measure Employee Engagement
- Our Engagement Survey Solution
- Our Pulse Survey Solution