What should I ask shoppers in a survey?
Last updated
One goal per survey. The set is built for it, in the order to ask.
Optional. It rewords the questions and answer options for your products.
Pick what you want to learn and a question set appears, ready to copy, with its rating question already on a live survey popup.
- What you want to learnNeeded
- What you sell, to reword the questionsOptional
The parts of a survey question that decide whether it gets answered
An on-site survey question is one short question a shopper can answer in a tap while still on the store, so a store unsure which ad channel works asks first-time customers How did you first hear about us, with seven channel types to choose from.
The questions are written by Nudgesmart from survey practice; no response-rate or satisfaction figure is used or implied. Only 1 to 5 rating questions preview here, because the survey template is a rating; the other question types are copy-only on this page and are built in the Nudgesmart app. To decide who sees a survey, plan your targeting rules.
Each set above is built for one goal, in the order the questions should be asked, with a note on when each one earns its place. These are the parts every question is built from.
- Goal
- The one decision the answers will feed: which channel to spend on, what to fix on a product page, why carts stall. A survey without a goal collects opinions nobody acts on.
- Question type
- A 1 to 5 rating is one tap and gets answered most. Multiple choice is one tap too, and tells you which reason. Open text is slow to answer and the richest to read, so it goes last and stays optional.
- Answer options
- The choices on a multiple-choice question. Five or six, mutually exclusive, with a way out such as Something else, so nobody has to pick an answer that is not true.
- Scale anchors
- The words at each end of a rating: 1 is not happy, 5 is delighted. Without them a 3 means something different to every shopper who picks it.
- When to ask
- Where and to whom the question appears. A post-purchase question on a first visit, or an exit question to a loyal customer, collects answers from the wrong people.
Nudgesmart is available on the Shopify App Store. Browse the template library or see what it costs.
Seven store problems and the survey that answers each
Illustrative situations, not store data. Each row is resolved by the same function the generator above runs: the goal it calls for, the question to lead with, and how many of the set can be previewed here.
| Situation | Goal | Lead question | Type | Where | Preview here |
|---|---|---|---|---|---|
| Apparel store with a lot of size returns | Product feedback | How would you rate the quality? | Rating, 1 to 5 | Home and account pages, on a later visit, once they have had time to use it | 2 of 4 |
| Clothing store with busy product pages and few checkouts | Why they are not buying | What stopped you from buying today? | Multiple choice | Product, collection and cart pages | 1 of 4 |
| Paying for three ad channels, unsure which works | How they heard about you | How did you first hear about us? | Multiple choice | Home and account pages, on the visit after a first order | 1 of 4 |
| Subscription pet food, first box just delivered | Post-purchase satisfaction | How happy are you with your last order? | Rating, 1 to 5 | Home and account pages, on their next visit after ordering | 2 of 4 |
| Wants one loyalty number to track every quarter | Would they recommend you | How likely are you to recommend us to a friend? | Scale, 0 to 10 | Home and account pages, on a return visit rather than straight after checkout | 1 of 4 |
| Carts left at the shipping step | Why they are not buying | What stopped you from buying today? | Multiple choice | Product, collection and cart pages | 1 of 4 |
| New range launched, first reorders due | Product feedback | How would you rate the quality? | Rating, 1 to 5 | Home and account pages, on a later visit, once they have had time to use it | 2 of 4 |
What a survey answer cannot tell you
The people who answer are not everyone
A survey hears from shoppers willing to answer a popup, which is not a random sample. Read the answers as reasons and patterns, not as a percentage of all your customers.
How did you hear about us is memory, not tracking
Shoppers remember the last thing they noticed, not the first thing that reached them, and many pick the easiest answer. Use the replies alongside your own order and ad data rather than instead of it.
A score over time only works if the question stays the same
Changing the wording of a rating or a 0 to 10 question, or its anchors, changes the answers. Pick the wording once and keep it, or the trend line measures your edits.
Order history only reaches signed-in customers
A placement aimed at customers with at least one order, or at first-time customers, is matched on the customer account. A shopper who checked out as a guest and is not signed in looks like any other visitor, so these surveys reach fewer of your buyers than your order count suggests. There is also no timer that waits a set number of days after delivery; asking on a later visit is the practical stand-in.
The preview shows a design, not your survey
Rating questions are shown on a ready-made survey template so you can read the question at phone size. Your own survey can use a different design and a different question type.
Shopper survey questions
What should I ask shoppers in a survey?
Start from the decision the answers will feed, then ask the fewest questions that can settle it. To learn which channel works, ask first-time customers how they first heard about you. To learn why carts stall, ask leaving shoppers what stopped them, as a single tap. The generator above builds a set for each goal.
How many questions should a popup survey have?
One or two for a popup, and never more than four. Put the one-tap question first, because it gets the most answers, and keep open text last and optional. A long survey in a popup is a form most shoppers close.
What are good post-purchase survey questions?
How happy are you with your last order, on a 1 to 5 scale; how easy it was to order; what almost stopped them ordering; and one open question about what to do better. Ask after the order has arrived, not before, so the answer is about the product and not the wait.
Why can only some questions be previewed here?
The preview uses a ready-made survey popup that asks a 1 to 5 rating question, so a rating question can be shown on it exactly as a shopper would see it. Multiple-choice, yes-or-no, open-text and 0 to 10 questions are copy-only on this page; the survey campaign in the Nudgesmart app builds those types.
Should I offer a discount for answering?
Only if you can live with the bias. A reward raises the number of answers and lowers their quality, because some shoppers answer for the code rather than to tell you something. For a one-tap question, the question itself is usually enough.
The right question at the wrong moment goes unanswered.
Feedback surveys are a campaign type in Nudgesmart, shown to the shoppers you choose, when you choose, as a popup on the store.
