Run a WhatsApp NPS Survey and Log Scores to Google Sheets
An n8n workflow that captures customers' WhatsApp replies to a 0–10 'how likely are you to recommend us' survey, logs the score and any comment to Google Sheets, and routes unhappy responders to your team — turning WhatsApp into a lightweight, high-response-rate NPS channel.
- 1
Capture the reply
Add a
Webhooknode subscribed to the WhatsAppmessagesevent. This fires when a customer replies to your NPS template. ACodenode extracts the sender number and message text. - 2
Parse the score
A
Codenode finds a 0–10 number in the reply and assigns an NPScategory: promoter (9–10), passive (7–8), detractor (0–6). Any non-numeric reply is kept as a free-text comment instead of a score. - 3
Log to Google Sheets
A
Google Sheetsnode appends the date, phone number, score, category and any comment. This growing table is your NPS dataset — chart the rolling score or export it whenever you report. - 4
Route detractors to the team
An
IFnode checks fordetractorand, if so, aSlacknode alerts your success team with the number and comment so they can reach out personally — recovering an unhappy customer before they churn or leave a public review. - 5
Send a thank-you reply
An
HTTP Requestsends a short WhatsApp acknowledgement ('Thanks for the feedback!'). Detractors can get a slightly different message promising a follow-up. Closing the loop makes customers feel heard and lifts future response rates.
Frequently asked questions
Why NPS over WhatsApp instead of email?
WhatsApp survey response rates dwarf email — a one-tap number reply feels effortless, so you collect far more scores. The trade-off is you must send the initial ask as an approved template; but the replies come back as free-form messages this workflow captures within the 24-hour window.
How does it classify promoters, passives and detractors?
Standard NPS buckets: 9–10 is a promoter, 7–8 passive, 0–6 a detractor. A Code node parses the number from the reply and applies these bands. Detractors trigger an immediate Slack alert so someone can reach out before a bad experience becomes a bad review.
What if someone replies with words instead of a number?
The parser looks for a 0–10 number first. If it finds none, the workflow logs the raw text as a comment against their last score (or sends a gentle 'please reply with a number 0–10'). This keeps the data clean while still capturing qualitative feedback people volunteer.
Brand Mention Tracker: know what the internet said about you today — in one evening digest
An n8n workflow that sweeps Google News RSS and Reddit for your brand name (and common misspellings) every few hours, has Claude judge sentiment and importance, and sends one evening Telegram digest — with an instant bypass alert if something negative starts moving.
Free recipe
Google Review Guardian: catch every new review and get an AI-drafted reply before breakfast
An n8n workflow that checks your Google Business Profile for new reviews every morning, has Claude draft a personalized owner reply for each (with extra care on 1-3 star reviews), logs everything to a sheet, and pings your Telegram so bad reviews never sit unanswered for days.
Free recipeTrustpilot Review Monitor: every new review answered within hours, drafted in your voice
An n8n workflow that checks your Trustpilot profile daily via an Apify actor, has Claude draft a reply for each new review — grateful for praise, accountable and concrete for criticism — and sends drafts to your Telegram for one-tap approval before you post them.