Free tool Survey question generator for NPS, CSAT and CES
Most satisfaction surveys go wrong before anyone answers: a reworded NPS question, a scale with no labels, a follow-up that asks promoters what went wrong. Pick the metrics and the moment you are asking about, and get the questions with the standard wording, labelled scales, follow-ups that branch on the score and the variables your survey tool needs. Copy it, or export CSV or JSON to build the survey.
- Standard wording, with sources
- Follow-ups branch on the score
- Runs locally
Build your survey
Tick the metrics you want and the questions rewrite themselves. The table is a preview; the text, CSV and JSON below hold the same survey.
Changes the subject of the CSAT question, the CES task and the answer options of the “what influenced your rating” question.
Independent of this page’s language. Chinese wording is TapCub’s translation of the English originals.
Used in the NPS and CES questions. Leave it empty to write “us” and “the company”.
Branching asks a different open question to unhappy and happy respondents.
The 1–7 agreement statement is the common form today; the 1–5 effort question is the 2010 original.
{{first_name}} is the common merge-tag style; check what your survey or email tool expects.
Preview
Q1a / Q1b are follow-ups; the right column says who sees them.
| # | Question | Answer | Shown to |
|---|---|---|---|
Q1 | Overall, how satisfied were you with the help you received from our support team? | Rating 1–5 | Everyone |
Q1a | Sorry to hear that. What could we have done better? | Open text | If Q1 = 1–3 |
Q1b | Glad to hear it. What did we do well? | Open text | If Q1 = 4–5 |
Q2 | Northwind made it easy for me to handle my issue. | Rating 1–7 | Everyone |
Q2a | What made it hard? | Open text | If Q2 = 1–3 |
Q3 | Which of these influenced your rating most? (choose up to 3) | Multiple choice (7) | Everyone |
Q4 | May we contact you about your answers? | Single choice | Everyone |
【Intro】
Hi {{first_name}}, thanks for contacting Northwind support. {{agent_name}} helped you with request {{ticket_id}}. Could you answer a few quick questions? It takes about 2 minutes.
Q1. Overall, how satisfied were you with the help you received from our support team? *
1–5 (1 = Very dissatisfied · 2 = Dissatisfied · 3 = Neither satisfied nor dissatisfied · 4 = Satisfied · 5 = Very satisfied)
Q1a. (If Q1 = 1–3) Sorry to hear that. What could we have done better?
[Open text]
Q1b. (If Q1 = 4–5) Glad to hear it. What did we do well?
[Open text]
Q2. Northwind made it easy for me to handle my issue. *
1–7 (1 = Strongly disagree · 2 = Disagree · 3 = Somewhat disagree · 4 = Neither agree nor disagree · 5 = Somewhat agree · 6 = Agree · 7 = Strongly agree)
Q2a. (If Q2 = 1–3) What made it hard?
[Open text]
Q3. Which of these influenced your rating most? (choose up to 3)
☐ Speed of the first reply
☐ My issue was resolved
☐ Knowledge of the agent
☐ Friendliness
☐ Having to repeat myself
☐ Finding a way to contact you
☐ Other (please specify)
Q4. May we contact you about your answers?
☐ Yes
☐ No
【Thank you page】
Thank you. Your answers go straight to the team and help us improve.
————————————
* = required
How to score:
- CSAT = ratings of 4 and 5 ÷ all ratings × 100%. Calculator: tapcub.com/tools/csat-calculator
- CES (1–7) = average rating; also report the share of 5–7 (“found it easy”). Calculator: tapcub.com/tools/customer-effort-score-calculator
- Q1: Template wording: CSAT has no single originator.
- Q2: CES 2.0 wording: CEB, The Effortless Experience (2013), with the company name filled in. Higher is better.
Variables (replace in your survey tool):
- {{first_name}} — Customer’s first name
- {{agent_name}} — Name of the agent who handled the case
- {{ticket_id}} — Ticket or conversation numberQuestions, answer scales, the scoring rules and the variables, ready to paste into a document or a survey builder.
no,id,type,required,show_if,question_en,scale_min,scale_max,options_en,note
,intro,statement,no,,"Hi {{first_name}}, thanks for contacting Northwind support. {{agent_name}} helped you with request {{ticket_id}}. Could you answer a few quick questions? It takes about 2 minutes.",,,,
Q1,csat,rating,yes,,"Overall, how satisfied were you with the help you received from our support team?",1,5,1=Very dissatisfied | 2=Dissatisfied | 3=Neither satisfied nor dissatisfied | 4=Satisfied | 5=Very satisfied,Template wording: CSAT has no single originator.
Q1a,csat_low,text,no,csat=1-3,Sorry to hear that. What could we have done better?,,,,
Q1b,csat_high,text,no,csat=4-5,Glad to hear it. What did we do well?,,,,
Q2,ces,rating,yes,,Northwind made it easy for me to handle my issue.,1,7,1=Strongly disagree | 2=Disagree | 3=Somewhat disagree | 4=Neither agree nor disagree | 5=Somewhat agree | 6=Agree | 7=Strongly agree,"CES 2.0 wording: CEB, The Effortless Experience (2013), with the company name filled in. Higher is better."
Q2a,ces_hard,text,no,ces=1-3,What made it hard?,,,,
Q3,drivers,multiple_choice,no,,Which of these influenced your rating most? (choose up to 3),,,Speed of the first reply | My issue was resolved | Knowledge of the agent | Friendliness | Having to repeat myself | Finding a way to contact you | Other (please specify),
Q4,contact,single_choice,no,,May we contact you about your answers?,,,Yes | No,
,thanks,statement,no,,Thank you. Your answers go straight to the team and help us improve.,,,,The downloaded file starts with a UTF-8 byte order mark so Excel shows Chinese correctly.
{
"generator": "TapCub survey question generator (tapcub.com/tools/survey-question-generator)",
"language": "en",
"touchpoint": "support",
"intro": "Hi {{first_name}}, thanks for contacting Northwind support. {{agent_name}} helped you with request {{ticket_id}}. Could you answer a few quick questions? It takes about 2 minutes.",
"questions": [
{
"no": "Q1",
"id": "csat",
"type": "rating",
"metric": "csat",
"required": true,
"question": "Overall, how satisfied were you with the help you received from our support team?",
"scale": {
"min": 1,
"max": 5,
"labels": {
"1": "Very dissatisfied",
"2": "Dissatisfied",
"3": "Neither satisfied nor dissatisfied",
"4": "Satisfied",
"5": "Very satisfied"
}
},
"note": "Template wording: CSAT has no single originator."
},
{
"no": "Q1a",
"id": "csat_low",
"type": "text",
"required": false,
"question": "Sorry to hear that. What could we have done better?",
"showIf": {
"question": "csat",
"min": 1,
"max": 3
}
},
{
"no": "Q1b",
"id": "csat_high",
"type": "text",
"required": false,
"question": "Glad to hear it. What did we do well?",
"showIf": {
"question": "csat",
"min": 4,
"max": 5
}
},
{
"no": "Q2",
"id": "ces",
"type": "rating",
"metric": "ces",
"required": true,
"question": "Northwind made it easy for me to handle my issue.",
"scale": {
"min": 1,
"max": 7,
"labels": {
"1": "Strongly disagree",
"2": "Disagree",
"3": "Somewhat disagree",
"4": "Neither agree nor disagree",
"5": "Somewhat agree",
"6": "Agree",
"7": "Strongly agree"
}
},
"note": "CES 2.0 wording: CEB, The Effortless Experience (2013), with the company name filled in. Higher is better."
},
{
"no": "Q2a",
"id": "ces_hard",
"type": "text",
"required": false,
"question": "What made it hard?",
"showIf": {
"question": "ces",
"min": 1,
"max": 3
}
},
{
"no": "Q3",
"id": "drivers",
"type": "multiple_choice",
"required": false,
"question": "Which of these influenced your rating most? (choose up to 3)",
"options": [
"Speed of the first reply",
"My issue was resolved",
"Knowledge of the agent",
"Friendliness",
"Having to repeat myself",
"Finding a way to contact you",
"Other (please specify)"
],
"maxChoices": 3
},
{
"no": "Q4",
"id": "contact",
"type": "single_choice",
"required": false,
"question": "May we contact you about your answers?",
"options": [
"Yes",
"No"
]
}
],
"thanks": "Thank you. Your answers go straight to the team and help us improve.",
"variables": [
{
"token": "{{first_name}}",
"meaning": "Customer’s first name"
},
{
"token": "{{agent_name}}",
"meaning": "Name of the agent who handled the case"
},
{
"token": "{{ticket_id}}",
"meaning": "Ticket or conversation number"
}
]
}For importing into your own form or survey code. showIf holds the branching rules.
Version v1.0Updated Update history
What each part of the survey is for
A short survey with one clear rating question and one good open question usually tells you more than a long one.
The rating question
This is the number you will track. Keep the wording and the scale exactly the same from one survey to the next, or the trend line breaks. The standard NPS and CES questions are kept as their originators published them; only the company name is filled in.
Follow-ups that branch
Asking a 2-star customer “what did we do well?” wastes the moment. Branching shows “what could we have done better?” to low scores and “what did we do well?” to high ones. The cut-offs follow each metric: CSAT 1–3 vs 4–5, CES 1–3 on the 7-point scale, NPS 0–6 / 7–8 / 9–10.
“What influenced your rating”
Open comments are rich but slow to read. A short list of reasons that fit the touchpoint lets you count causes: if “having to repeat myself” is ticked by a third of unhappy chat customers, you know where to look.
Intro and variables
Mentioning the agent, ticket or order tells the customer which experience you mean. The variables are placeholders: your survey or email tool replaces them with each customer’s data.
Scores computed from these questions can be checked with the NPS, CSAT and CES calculators; the scoring rules are written at the end of the plain-text survey.
From questions to a running survey
The generator writes the questions; you send them with the tool you already use.
-
1
Choose the moment and the metric
CSAT or CES right after a support case or delivery; NPS on a schedule for the relationship. Tick only what you will act on.
-
2
Set language, follow-ups and variables
Pick the survey language, keep branching follow-ups on, and choose the variable format your tool expects.
-
3
Export and build
Paste the text into a form builder, or import the CSV / JSON. Keep the rating question’s wording and scale exactly as generated.
-
4
Close the loop
Read the low-score answers every week and contact the people who said yes to a follow-up. A survey nobody acts on teaches customers to ignore the next one.
Where the standard questions come from
NPS and CES each have a published original. CSAT has no single originator, so its wording here is a template.
- NPS: “How likely are you to recommend us to a friend or colleague?” from 0 to 10, as published on Bain & Company’s Net Promoter System site. NPS was introduced by Fred Reichheld of Bain in Harvard Business Review in 2003.
- CES 1.0: “How much effort did you personally have to put forth to handle your request?” from 1 (very low effort) to 5 (very high effort), Dixon, Freeman and Toman, HBR, July–August 2010 (CEB research).
- CES 2.0: agreement with “The company made it easy for me to handle my issue.” from 1 (strongly disagree) to 7 (strongly agree), from CEB’s book The Effortless Experience (2013). CEB is now part of Gartner.
- CSAT: “Overall, how satisfied were you with …?” on a labelled 1–5 scale is a common template, not a standard.
| Metric | Scale | Score | Follow-up cut-off |
|---|---|---|---|
| NPS | 0–10 | % 9–10 minus % 0–6 | 0–6 / 7–8 / 9–10 |
| CSAT | 1–5 | % of 4 and 5 | 1–3 vs 4–5 |
| CES 2.0 | 1–7 agreement | Average, or % of 5–7 | 1–3 = hard |
| CES 1.0 | 1–5 effort | Average (lower is better) | 4–5 = hard |
A post-chat survey for Northwind
The default settings: a support chat, CSAT and CES, branching follow-ups, reasons and a follow-up permission question.
| # | Question | Shown to |
|---|---|---|
| Q1 | Overall, how satisfied were you with the help you received from our support team? (1–5) | Everyone |
| Q1a | Sorry to hear that. What could we have done better? | Q1 = 1–3 |
| Q1b | Glad to hear it. What did we do well? | Q1 = 4–5 |
| Q2 | Northwind made it easy for me to handle my issue. (1–7) | Everyone |
| Q2a | What made it hard? | Q2 = 1–3 |
| Q3 | Which of these influenced your rating most? (choose up to 3) | Everyone |
| Q4 | May we contact you about your answers? | Everyone |
Each respondent sees at most six questions. If 120 customers answer Q1 and 96 choose 4 or 5, CSAT is 96 ÷ 120 = 80%.
Survey habits that distort the score
Small wording changes move satisfaction scores more than most service changes do.
Rewording the standard question
“Would you recommend us?” with yes / no is not NPS, and its result cannot be compared with anyone’s NPS. Keep the published wording and scale.
Unlabelled or flipped scales
Label at least both ends. Mixing a 1–5 effort scale (lower is better) with a 1–7 ease scale (higher is better) in one report makes trends meaningless.
Too many questions
Every extra question loses some respondents, and the ones who stay are not typical. One rating, one open question and one reasons question covers most needs.
Asking at the wrong time
CSAT and CES belong right after the interaction, while it is remembered. NPS after a single support case mostly measures that case.
Satisfaction ratings already built into TapCub live chat
For website chat you do not need a separate survey: TapCub asks for a rating when a conversation ends. Use this generator for NPS and email surveys.
Rating when a chat is resolved
When an agent marks a conversation resolved, the visitor sees a 1–5 star rating with an optional comment, once per conversation.
Learn moreAlerts for low ratings
A rating under 3 stars triggers an alert, alongside new and transferred conversations, so someone can follow up the same day.
Learn moreSatisfaction in the chat report
First response, resolution rate and satisfaction per agent and per day, next to the analytics for the same visitors.
Learn moreSurvey question FAQ
Still have a question?
Chat with the team behind TapCub — we usually reply within a few hours on business days.
What is the standard NPS question?
“How likely are you to recommend us to a friend or colleague?”, answered from 0 to 10. That is the wording on Bain & Company’s Net Promoter System site; you can replace “us” with your company or product name. Scores of 9–10 are promoters, 7–8 passives and 0–6 detractors, and NPS is the percentage of promoters minus the percentage of detractors.
What should a customer satisfaction survey include?
One rating question you will track over time, one open question asking why, and optionally a short list of reasons and a permission question for follow-up. Add a one-line intro that says which experience you are asking about. Longer surveys rarely add insight and lose respondents.
What follow-up question should I ask after NPS?
Ask detractors what you should improve, passives what would make them more likely to recommend you, and promoters what they like most. If your tool cannot branch, a single “What is the main reason for your score?” works for everyone.
Which CES question should I use?
The 1–7 agreement statement “The company made it easy for me to handle my issue” is the form most teams use today. The original 2010 question asked how much effort the customer had to put forth on a 1–5 scale, where lower is better. Pick one and do not mix them in one report.
Is there a standard CSAT question?
No single originator defined one. “Overall, how satisfied were you with …?” on a labelled 1–5 scale is the common form, and CSAT is usually reported as the share of 4 and 5 ratings. Whatever you pick, keep it unchanged.
Should I use NPS, CSAT or CES?
CSAT and CES suit a single interaction, such as a support case or a delivery, and are asked right after it. NPS describes the whole relationship and is usually sent on a schedule. Many teams run CSAT or CES after every case and NPS once or twice a year.
How do I import the questions into a survey tool?
Most survey and form builders take questions by copy and paste; the plain-text version keeps the numbering and the branching notes. The CSV has one question per row with scale and options columns, and the JSON includes the branching rules for your own form code.
Is what I type stored?
No. The questions are generated in your browser, and this tool has no share link.
Update history
New features, fixes and data refreshes for this tool, newest first.
-
· v1.0
LaunchedFirst release.
Missing something or found an error? Tell us — requests that come up often go into the next update. All tool updates
Ask for a rating where the conversation happens
TapCub live chat asks for a star rating when a chat is resolved and shows satisfaction next to first response and resolution rate.