Reading your results

Every question has a summary card on the Response Dashboard. To get there, see See your survey’s results. Each card shows a chart automatically. What’s underneath the chart depends on your plan and the question type.

The Data Table

Below each chart is a Data Table toggle. Click it, or press Enter on it, to expand the exact numbers. It’s on every plan, because it’s also how a screen reader reads the chart. On Team and above you can click an answer in the table to filter the dashboard to it; on other plans the table lists the answers as plain text.

What it shows depends on the question type:

Question type	Data Table shows
Multiple choice, image choice, Yes/No	Option, Count, Percentage
Rating	A Mean / Median / Std Dev / Min / Max / Count table, plus a Count/Percentage table per rating value
Slider	Mean / Median / Std Dev / Min / Max per statement, plus an overall average
NPS	Promoter/Passive/Detractor counts and the NPS Score — no mean or standard deviation, since NPS is reported by segment, not by average
Matrix	A count per row and scale point — no percentages or averages
Text input	The 20 most frequent words, with frequency counts
MaxDiff	Share of preference with its 95% interval, utility and standard error, the best − worst score, and how often each item was shown and picked. See MaxDiff questions

On a multi-select question, percentages don’t sum to 100%. Percentage is each option’s share of all selections, not of respondents, so someone who picked three options is counted three times.

Standard deviation is the population version (computed across everyone who answered, not a sample estimate), and Min/Max are the lowest and highest answers actually given, not the scale’s endpoints.

CSAT, CES and Top 2 box on rating questions

A rating question can report one extra figure beside its average: CSAT, CES or Top 2 box, shown as a percentage of the people who answered.

• CSAT and CES count every point on the positive side of the scale. On a 5-point scale that is 4 and 5.
• Top 2 box counts the top two points of the scale.

How SurveyRock picks one. By default your answer labels decide. A scale whose top point is labelled “Very satisfied” reports CSAT, and one whose top point is labelled “Very easy” reports CES. To choose yourself, open the question in the survey editor and set Report as under Results. The same section lets you set the point where counting starts.

What you see. The figure appears on the question’s card. Once there are enough responses, the dashboard adds a trend line under it. When a survey has exactly one CSAT or CES question, the dashboard also shows a KPI card for it. The figure, the trend, the KPI card and the Data Table are on every plan. The card’s CSV download and the dashboard filter bar need Team or above, so recalculating the figure for a filtered group is a Team feature.

Good to know.

• Labels are read in English only. If your labels are in another language, the question reports Top 2 box until you set Report as.
• The figure is calculated when you view it. If you change the scale or its labels later, old answers are read against the new setting.
• It is not in the raw exports, shared reports or the API. The CSV and Excel response exports hold each person’s answer, not the CSAT or CES figure. Calculate it from the answers there if you need it outside the dashboard.
• Only rating questions report it. Matrix, NPS and slider questions do not, and there is no survey-level score.

For what the metric means and how to read it, see What CSAT measures.

Your NPS against your industry

Under the NPS Score on a question’s card, a line from −100 to +100 shows the typical range in published industry studies for your industry, with your score placed on it.

• The shaded band is the spread of NPS figures that recent published studies report for your industry. Those studies cover companies surveying their own customers.
• The short mark to the right is where those studies put the top quarter of companies.
• The dot is your score, and the line through it is its margin of error at 95% confidence. With 200 responses the margin is typically 9 to 12 points either way, depending on how the answers split.

What the label beside your score means. Once your score is compared with a range, the label says where it sits:

Label	Means
Within the typical range	Your score is inside the band
Close to the typical range	Your score is outside the band, but its margin of error reaches it, so the difference may not be real
Below the typical range	Even the top of your margin of error is under the band
Above the typical range	Even the bottom of your margin of error is over the band
Above the typical range, at top-quarter level	As above, and your score is at or past the top-quarter mark

With fewer than 30 responses, the margin is too wide to say anything, so the card keeps its usual label (Excellent, Great, Good or Needs improvement) and tells you how many responses a comparison needs.

Which industry it uses. The range follows your account’s industry, set in Workspace Settings → Anonymous Benchmarks. If your account has no industry yet, the card asks you to choose one. To compare one survey with a different industry, pick it under Compare with on the card, or choose Don’t compare. This is useful when you survey a client’s customers. The choice applies to every NPS question in that survey. Anyone who can edit the survey can change it. Setting the account’s industry from the card needs an owner or admin.

Good to know.

• It’s a range from published studies, not a SurveyRock benchmark. Studies measure in different ways and disagree, so the band is wide in some industries (insurance runs from 36 to 65). Treat it as context, not a target. Your own trend over time is the stronger comparison: see Compare results over time.
• Some industries have no range yet. Where fewer than two studies cover an industry (hospitality, government, non-profits), the card shows the range across all industries and says so.
• Only customer and user surveys are compared. A survey set as an employee, event or market-research survey shows no range, because the studies don’t describe those.
• It follows your filters. Filter the dashboard and the score, its margin and the label are recalculated for the filtered group.
• It’s on every plan, and it goes wherever your NPS does. Shared report links show the same line and label (with no Compare with). The PDF report says it in your survey’s language, and Word and PowerPoint exports print it under the NPS score. Digests add it under the NPS figure, from the score across all your NPS questions. The AI summary and the assistant can quote it. The API’s results summary carries it in each NPS question’s nps.industryRange.

Exporting the numbers

The per-question Download button on each card (Team and above) exports the same statistics shown in that card’s Data Table. For raw response-level data instead of summaries, use Export on the Respondent List — see Results & Analysis for which export formats are on your plan.

Comparing across questions or segments

Cross-tabulation — comparing one question’s answers against another’s — is a separate tool from the per-question summary above. It’s on Growth and above; see Results & Analysis for what it covers.

Acting on what you find

When a result tells you something needs doing, you can raise an action from it, with an owner and a due date, so the finding turns into a follow-up. See Turn a theme, a driver, or an answer into an action.

Troubleshooting

The chart shows “Chart rendering pending” instead of an image. Downloading a chart as a PNG or SVG image doesn’t work yet. The numbers themselves are still accurate; use the Data Table or the CSV export instead of the image download.

A rating question’s header reads “/ 4” but I used a 5-point scale. That number reflects the highest answer anyone actually gave, not the scale’s maximum — if nobody has selected 5 yet, it won’t show.