Customer satisfaction survey: complete guide for small businesses and freelancers (2026)

Are your customers truly satisfied — or just silent? A well-designed satisfaction survey answers that question and turns feedback into concrete action: improve your service, spot unhappy customers before they post a negative review, and retain those who recommend your business. This guide, linked to our Google reviews and local visibility pillar, gives you the complete method to create, send and act on a satisfaction survey — even with no budget or technical skills.

Why a customer satisfaction survey is essential in 2026

The numbers speak for themselves:

• 68% of customers leave a business because they feel ignored — not because of price or quality (source: Bain & Company).
• Only 1 in 26 unhappy customers complains directly. The other 25 leave silently — or post a 1-star Google review.
• Businesses that actively measure satisfaction have a 25–30% higher retention rate than those that don't.

A satisfaction survey is an early warning system. It catches weak signals — a wait time that's too long, a cold reception, a service that didn't match expectations — before they become public problems on your Google Business Profile.

The 3 types of satisfaction surveys (and which to choose)

1. NPS — Net Promoter Score

One question: "On a scale of 0 to 10, would you recommend our business to a friend?" Respondents fall into 3 groups:

Formula: NPS = % of promoters − % of detractors. An NPS of 50+ is excellent. A negative NPS means you're losing customers faster than you're gaining them.

2. CSAT — Customer Satisfaction Score

Typical question: "How would you rate your experience?" (1 to 5 stars or emojis). CSAT measures immediate satisfaction on a specific point: response speed, service quality, phone reception. Simple, fast, ideal after every interaction.

3. CES — Customer Effort Score

Typical question: "How much effort did you have to put in to get what you wanted?" (scale 1 to 7). CES measures ease of interaction. For a tradesperson, this is crucial: a customer who had to call 3 times to book an appointment gives a high CES (= bad) even if they're happy with the work itself.

15 question examples for your survey

Here are the most effective questions, sorted by category. Pick those that match your activity — a hair salon doesn't have the same priorities as a car garage.

Overall satisfaction questions

1. On a scale of 0 to 10, would you recommend our business? (NPS — mandatory)
2. How would you rate your overall experience? (1 to 5 stars)
3. Did you get what you expected? (Yes / Partially / No)

Service-specific questions

4. How would you rate our response speed? (1 to 5)
5. Does the price seem fair for the service received? (1 to 5)
6. Was the reception professional and friendly? (1 to 5)
7. Did the service/product meet your quality expectations? (1 to 5)
8. Were you kept informed at each stage? (Yes / No)

Retention and improvement questions

9. What could we improve? (Open answer — the gold mine)
10. Would you come back to us? (Yes / Maybe / No)
11. How did you hear about us? (Google, word-of-mouth, social media, etc.)
12. Would you like an automatic reminder for your next appointment? (Yes / No)

Industry-specific questions

13. For a restaurant: "How would you rate the wait time between ordering and being served?" — a major irritant that a restaurant owner can fix quickly.
14. For a physiotherapist: "Were the prescribed exercises clear and easy to do at home?" — a differentiator for a physiotherapy practice.
15. For a car garage: "Did you receive a clear quote before the work started?" — the #1 trust factor in automotive.

How to send your survey (and actually get responses)

The best survey in the world is useless if nobody fills it in. The delivery channel makes all the difference:

WhatsApp wins because the message arrives where the customer already spends their time. No need to click a link, open an email or find a form. They respond directly in the conversation.

Automate sending with an AI bot

The manual approach (copy-pasting a survey after each appointment) works at 5 customers a week. Beyond that, it's unmanageable. Reepli.ai automates the entire process:

1. 2 hours after the appointment — the bot automatically sends a WhatsApp message: "Hi [First name], thank you for your visit! Could you rate your experience out of 5?"
2. The customer replies with a number or emoji — the bot records the score and asks a follow-up question if needed.
3. If the score is 4 or 5 — the bot offers to leave a Google review with a direct link. Result: satisfied customers become active promoters.
4. If the score is 1 to 3 — the bot asks "What could we improve?" and alerts the business owner to intervene before a negative review goes public.

How to act on your survey results

Collecting responses isn't enough. Here's the 4-step process to turn data into action:

1. Calculate your metrics monthly — NPS, average CSAT, Google rating. Compare month-over-month to spot trends. A dropping CSAT on "response speed" means you need to work on your reply time, not your service quality.
2. Identify the 3 most frequent irritants — read the open-ended answers (question 9: "What could we improve?"). Rank by frequency. If 40% of comments mention phone wait times, that's your #1 priority.
3. Fix one irritant per month — not everything at once. One measurable change (e.g., moving initial contact to WhatsApp to reduce phone wait times) can be verified in the next survey.
4. Redirect promoters to Google — customers who score 9 or 10 on NPS are your best advocates. Automatically send them a link to your Google Business Profile to leave a public review.

"Since we started sending the survey by WhatsApp after every haircut, our response rate jumped from 8% (email) to 52%. We caught a Saturday morning wait time issue and added a slot. Within 3 months, our Google rating went from 4.1 to 4.6." — Sophie, hairdresser in Lyon

5 mistakes that ruin your satisfaction survey

1. Too many questions — beyond 10 questions, the drop-off rate exceeds 50%. Each added question reduces response reliability.
2. Delayed sending — a survey sent 1 week after the service gets 3 times fewer responses than one sent within 2 hours. Memory fades fast.
3. Leading questions — "Was our service excellent?" isn't a question. Use "How would you rate our service?" with a neutral scale instead.
4. No open-ended question — scores show trends, but only free text tells you WHY. Always include at least one open-ended question.
5. Not acting on results — the worst: collecting feedback and changing nothing. Customers who take the time to respond expect their input to be taken into account.

Tools for creating your survey (free and paid)

The key difference: Google Forms or Typeform create the form, but it's up to you to send the link and chase non-responders. Reepli.ai automatically sends the survey via WhatsApp after every appointment, collects responses, redirects satisfied customers to your Google review link, and alerts you about unhappy ones — all without manual intervention.

Frequently asked questions about customer satisfaction surveys