Reading and Interpreting Survey Results – Prosper

Collecting survey responses is the easy part — the value comes from reading the results well and turning them into action. This guide explains how to interpret your survey results in Prosper: what the key metrics mean, how to spot the real story in the data, and how to avoid the common traps that lead to the wrong conclusions.

Who this is for: Administrators and Managers with survey access

Read time: 7 minutes

Where to find your results

Survey results are available in two main places:

Within the survey itself — navigate to Surveys & Happiness and select a survey in Active or Complete. Here you'll find overall scores, response rates, and a breakdown of how people responded.
In the Management Dashboard (Analytics) — for tracking scores over time, comparing groups, and using the Heatmap.

The key metrics, explained

Response rate

The percentage of invited recipients who completed the survey. Response rate is the first thing to check — it tells you how much confidence to place in everything else.

High response rate → results are representative and trustworthy.
Low response rate → results may be skewed (often by the most or least satisfied employees), so interpret with caution.

Tip: A glowing score from a 20% response rate is far less meaningful than a moderate score from an 80% response rate. Always read the score through the lens of the response rate.

Favourability score

Favourability is the percentage of respondents who answered positively — typically those scoring 4 or 5 on a 5-point scale. It's the headline measure of how positively a group feels about a given area.

Response distribution (Positive / Neutral / Negative)

The split of responses across Positive, Neutral, and Negative. This is often more revealing than the score alone — two groups can share the same favourability score but have very different distributions.

Example: A 60% favourability score could mean "60% positive, 35% neutral, 5% negative" (broadly content, some on the fence) or "60% positive, 5% neutral, 35% negative" (polarised, with a significant unhappy group). The same score, two completely different stories — and two completely different responses required.

eNPS

Where you've run an eNPS question, the score ranges from −100 to +100 and reflects advocacy specifically. See Understanding and Running eNPS Surveys for the full detail.

Measurement pillars

Questions drawn from the Question Bank or ready-made surveys are mapped to measurement pillars. Pillar scores let you see strengths and focus areas across themes like Learning & Development, Leadership, or Wellbeing, and track them over time in the Management Dashboard.

How to read your results: a sensible order

Work through results in this order to avoid jumping to conclusions:

Check the response rate. Decide how much confidence the data warrants.
Look at the overall favourability. Get the headline picture.
Examine the distribution. Is sentiment consistent or polarised?
Compare against benchmarks. Company average, previous surveys, and relevant groups.
Identify the outliers. The highest and lowest pillar scores point to your strengths and focus areas.
Read the comments. The "why" behind the numbers.
Decide on action. Pick a small number of focus areas to act on.

Comparison is everything

A single number in isolation tells you very little. The insight comes from comparison:

Against the company average — is this group above or below the rest of the organisation? The Heatmap is built for exactly this.
Against previous surveys — is sentiment improving or declining? The trend often matters more than the absolute score.
Against other groups — do certain teams, locations, or roles consistently score differently? That points to local factors worth investigating.
Against external benchmarks — where available, how do you compare to your sector?

Reading comments well

Comments are where the numbers come to life — but they need careful interpretation:

Look for themes, not anecdotes. One strong comment is a data point; the same theme appearing repeatedly is a signal. Focus on patterns.
Weight by frequency, not volume. The loudest comment isn't necessarily the most representative. Count how often a theme recurs.
Read positive and negative together. Comments that explain why people are happy are as valuable as complaints — they tell you what to protect.
Respect anonymity. Never try to identify who wrote a comment. See Understanding Survey Anonymity and Thresholds.

Common interpretation traps to avoid

Over-reading small samples. A dramatic-looking result from a handful of responses can be noise. Check the response count before reacting.
Chasing the absolute number. A score of 70% sounds fine until you see it's down from 85%. The direction of travel is often the real story.
Ignoring the neutrals. A large neutral group isn't "fine" — it often represents disengagement or people who haven't been won over. They're frequently your biggest opportunity.
Cherry-picking comments. Selecting the comments that confirm what you already believed isn't interpretation — it's confirmation bias. Look at the full picture.
Reacting to everything at once. A survey can surface a dozen issues. Trying to fix all of them dilutes your effort and signals nothing got real attention.
Mistaking correlation for cause. A low score in one area doesn't automatically explain a low score in another. Investigate before concluding.

From results to action

Interpretation only matters if it leads somewhere. A simple, effective approach:

Pick two or three focus areas. Choose the areas with the biggest gap to the company average, the steepest decline, or the clearest comment themes. Resist doing everything.
Confirm the "why". Use comments and, if needed, follow-up conversations to understand the root cause before acting.
Agree concrete actions. Each focus area should have a clear, owned action — not a vague intention.
Close the loop. Tell employees what you heard and what you're doing. This single step does more for future response rates than anything else.
Re-measure. Use the next survey to check whether your actions moved the needle.

Tip: Nothing erodes survey participation faster than feedback that disappears into silence. "You said, we did" — even for small changes — is the most powerful thing you can do to keep your survey programme healthy.

Tips and best practices

Always read the score through the response rate. Confidence in the data comes first.
Lead with the trend. Where you have multiple surveys, the direction of change is usually more actionable than any single score.
Use the distribution, not just the average. The Positive/Neutral/Negative split reveals stories the headline score hides.
Pair numbers with comments. Quantitative data tells you what; comments tell you why. You need both.
Focus ruthlessly. A few well-actioned focus areas beat a long list of noted concerns.
Involve Managers. Local Managers often understand the context behind their team's scores better than anyone — interpret with them, not just for them.

Troubleshooting

My score went down but I'm not sure why.
Start with the distribution and the comments. A falling score with rising negatives points to a specific emerging issue; a falling score with rising neutrals suggests growing disengagement. The comments usually reveal the cause.

A group's results aren't showing.
Groups below the anonymity threshold (fewer than 3 responses) are suppressed. Analyse at a higher level or encourage participation. See Understanding Survey Anonymity and Thresholds.

Two groups have the same score but feel very different.
Check the response distribution. Identical favourability scores can mask very different Positive/Neutral/Negative splits — and very different realities.

The comments contradict the scores.
This often happens when a vocal minority dominates the comments while the silent majority drove the score. Weight comments by how often a theme recurs, and read them alongside the quantitative data rather than instead of it.

Response rate is low — should I still act?
Treat low-response results as indicative rather than conclusive. Look for strong, consistent signals, but be cautious about major decisions. Focus first on lifting participation next time.

I can't compare to previous surveys.
Consistent comparison requires consistent questions. Where you've used Question Bank or ready-made questions mapped to pillars, the Management Dashboard tracks them over time. Custom questions that change between surveys won't trend cleanly.

Related to