Track recommendation intent, understand score drivers, and route follow-up by promoter, passive, or detractor.
How likely are you to recommend us to someone with similar needs?
Start with the complete working structure, then adjust the language, timing, and incentive to fit your customers and brand.
Best used when
Your team has a specific decision or growth objective and needs a focused artifact to move from planning to launch.
How likely are you to recommend us to someone with similar needs?
NPSWhy ask it: Recommendation intent looks forward at the relationship rather than back at one order, which makes it the number worth trending across quarters instead of weeks. Split the follow-up by score: invite the highest scorers to leave a review, and put the lowest scorers in front of a person rather than another email.
What is the main reason for your answer?
open textWhy ask it: The score records what happened and this records why, which is why it belongs immediately after the rating while the customer is still holding the thought. Read it grouped by score: the reasons behind low answers are your fix list, and the reasons behind high answers are the next line of your product page.
What is the one thing we could improve?
open textWhy ask it: Asking for one thing instead of general feedback forces a priority, so the answers arrive already ranked by what the customer cares about most. Count how often each theme repeats rather than reacting to the most articulate response, and carry the top three into the next planning cycle.
Is there anything else you would like us to know?
open textWhy ask it: The last question is where the unprompted things arrive: a packaging complaint, a size you do not stock, one support agent thanked by name. Keep it optional and last so it costs nothing in completion, and lean on AI analysis to group the themes once you pass a few hundred responses.
Generate a focused variation while keeping the expert-written version available as your baseline.
Add your goal and audience. If AI is unavailable, the complete template on this page still works as a practical starting point.
Your inputs are sent only to generate this result and are not stored by UserLoop.
Three specifics worth checking against your own store before you launch.
Start with the focused version, publish, and let the first answers steer the next iteration.
Send it by email two to four weeks after delivery, never on the confirmation page, so the score covers the product rather than the checkout
Treat the open-text follow-up as mandatory in your own reporting: a score with no reason attached cannot be acted on
Route promoters toward reviews and referrals, and get a person in front of detractors within a day
Strong execution usually comes from removing ambiguity and keeping every action tied to the original objective.
Running NPS as a quarterly board slide and never reading the comments underneath it
Surveying first-time buyers before the order has arrived, which measures your checkout rather than your brand
Comparing your number with published industry averages collected on other scales at other moments
Calling a one or two point move a result when the sample changed underneath it
Subtract the percentage of detractors, who score 0 to 6, from the percentage of promoters, who score 9 or 10. Passives at 7 and 8 sit in the base but not in the result, which is why the number swings hard on small samples. Publish the response count next to the score every time, or it gets read as far more precise than it is.
As one input, yes. As your only customer metric, no. Its strength is that a single forward-looking question is easy to trend and everyone in the company already understands it. Its weakness is that it compresses an entire relationship into one digit, so pair it with a satisfaction score on specific moments and read the open text that explains both.
Satisfaction is the better operating metric, because it attaches to something you can fix this week: this delivery, this ticket, this product. Recommendation intent is the better relationship metric, asked less often and only of customers who have had time to form a view. Most stores are best served running satisfaction continuously and recommendation intent two or three times a year.
Two to four weeks after delivery, and only to customers who actually received the order. Asked at checkout it measures optimism about a purchase rather than experience of one. If you run it more than quarterly, exclude anyone who answered recently so the same engaged customers do not end up carrying your score.
Collect useful customer insight with Surveys and grow your audience with Contests.
