Loading tutorials…
Loading tutorials…
Most customer feedback surveys collect data that nobody uses. This walks through structuring NPS, CSAT, and qualitative feedback in Google Forms so the data drives decisions.
Who this is forOperators running customer-feedback loops. Founders trying to do NPS the first time. CX leads building post-purchase, post-onboarding, or churn surveys.
What you'll need
Step 1
NPS = ongoing relationship strength. CSAT = transactional satisfaction. CES = effort/friction. Pick based on what you need to learn.
NPS (Net Promoter Score): "How likely are you to recommend [X] to a friend or colleague?" 0-10 scale. Use for ongoing relationship health. Send quarterly or annually.
CSAT (Customer Satisfaction): "How satisfied were you with [specific interaction]?" 1-5 scale. Use post-purchase, post-support-ticket, post-onboarding.
CES (Customer Effort Score): "How easy was it to [task]?" 1-7 scale. Use to find friction in specific workflows.
Pick ONE per survey. Mixing them dilutes signal.
Step 2
NPS: 0-10 scale, 11 options. CSAT: 1-5 scale. Use the standard wording — comparing against benchmarks requires it.
NPS exact wording: "How likely are you to recommend [Company] to a friend or colleague?" Scale: 0 (Not at all likely) to 10 (Extremely likely). Use Linear Scale question.
CSAT exact wording: "How satisfied were you with [specific touchpoint]?" Scale: 1 (Very dissatisfied) to 5 (Very satisfied). Linear Scale.
CES exact wording: "How easy was it to [action]?" Scale: 1 (Very difficult) to 7 (Very easy). Linear Scale.
Industry benchmarks exist for these exact wordings. Custom wording = no benchmarks = no context for your number.
Step 3
NPS by itself is a number. The qualitative follow-up is where you learn what to do about it.
After the NPS/CSAT/CES question, add: "What's the main reason for your score?" (Paragraph type, optional).
Optional matters — required open-text drops completion by 20-40%.
The text answers are your roadmap. The number is just the dashboard.
Optional: add "Anything else you'd like to share?" as a second open-text. Captures volunteered feedback.
Step 4
Use section-based branching: if NPS score is 0-6 (detractor), branch to a follow-up section asking "What would we need to do to make this a 9 or 10?"
Convert the NPS Linear Scale question to a Multiple Choice with options 0-10 (so branching works).
For each option, set branching: 0-6 → "Detractor Follow-up" section. 7-8 → "Passive Follow-up." 9-10 → "Promoter Follow-up."
In Detractor section: "We'd love to fix this. What would we need to do to make your next experience a 9 or 10?"
In Promoter section: "Glad you love it! Would you be willing to share a quick review at [link]?"
This turns the survey from a measurement into an action loop.
Step 5
In the linked Sheet, create an Analysis tab. Calculate NPS, count Promoters/Passives/Detractors, prepare for theme-tagging.
Linked Sheet → new tab "Analysis."
NPS formula: `=(COUNTIF('Form Responses 1'!B:B,">=9") - COUNTIF('Form Responses 1'!B:B,"<=6")) / COUNTA('Form Responses 1'!B:B) * 100`.
Add columns for theme-tagging: "Theme 1," "Theme 2." Manually tag each open-text response with 1-2 themes (price, product, support, onboarding, etc.).
After 50-100 responses, the theme distribution shows you where to focus.
Step 6
Send to a real customer list. Every detractor gets a personal email or call within 48 hours.
Send via your email tool (Mailchimp, ActiveCampaign) with the form URL. Personal subject line wins: "[Name], 2-minute favor?"
Monitor responses daily in the Sheet.
For every detractor (score 0-6): send a personal email within 48 hours acknowledging the score and offering to chat.
The follow-up is what turns NPS from vanity metric into retention tool.
Common mistakes
Custom-wording the NPS question
What goes wrong: You change 'How likely to recommend' to 'How happy are you.' Your number is no longer comparable to industry benchmarks. You cannot tell if a 35 is good or bad for your category.
How to avoid: Use the exact standard wording for NPS, CSAT, CES. Benchmark comparability is the point.
Required open-text question
What goes wrong: You mark the 'why' question required. Completion drops 30%. You get fewer detractor follow-ups, which are the most valuable responses.
How to avoid: Open-text questions = always optional. The detractors who want to explain will.
Surveying too often
What goes wrong: You send NPS quarterly. After 12 months, customers stop responding because they have seen the same survey 4 times. Response rate drops to 5%.
How to avoid: NPS annually for most B2B. Quarterly only if you have a clear improvement program tied to the data.
Not following up on detractors
What goes wrong: Detractor scores 2/10 on NPS. No follow-up. They churn 3 months later. Survey was an early-warning signal you ignored.
How to avoid: Every 0-6 score = personal outreach within 48 hours. Treat as a save opportunity, not a data point.
No theme-tagging of open-text answers
What goes wrong: You have 200 open-text answers. You read 10 of them. The patterns are invisible. Survey was for nothing.
How to avoid: After every 25-50 responses, tag themes manually. Use 5-7 standard themes. Distribution shows priorities.
Recap
Done — what's next
How to set up a Google Forms survey the right way
Read the next tutorial
Hand it off
Running a feedback survey is a project. Running a customer-feedback program with action loops, theme analysis, and quarterly reporting is a job. A vetted specialist will run the system. From $14-16/hr.
See specialist rates
Annually for B2B SaaS with relationship-based customers. Quarterly only if you have a clear improvement program. Avoid more often than quarterly — fatigue kills response rate.
Varies by industry. SaaS: 30-50 is typical, 50+ is great. Consumer brands: lower benchmarks. The number matters less than the trend — your score increasing year-over-year is the win.
Generally no for NPS — incentives skew responses positive. For CSAT after a transaction, small incentive (5% discount) can be fine. Long surveys: yes, incentive ($5 gift card) for completion.
Send NPS at most quarterly. Vary the format (NPS vs CSAT vs open-ended) so customers do not see the same survey repeatedly.
Google Forms works fine for NPS up to ~5,000 customers. Above that, paid tools (Delighted, Wootric, Refiner) automate distribution, segmentation, and follow-up. Worth $50-200/mo at scale.
Google Forms
Google Forms is free, fast, and forgiving — which is why most surveys collect garbage data. This walks through the setup path that actually returns reliable answers.
Google Forms
Google Forms → Sheets is the simplest sync in the SaaS ecosystem — and the easiest to mess up by adding formulas in the wrong place. Here is the setup that survives 10,000 responses.
Google Forms
Google Forms has weak native notifications, fine Sheets-based notifications, and powerful Apps Script notifications. This walks through each tier and when to upgrade.
Google Forms
DIY surveys are great — until the data stops driving decisions, or your forms convert at 18%. Here is the honest framework: when the cost of self-managing exceeds the cost of hiring.