What Is a Good NPS for Ecommerce? How to Compare Scores

A useful ecommerce NPS benchmark comes from comparable customers answering the same kind of question at a comparable point in their experience. There is no single score that establishes whether every store is doing well. Start with a clearly defined baseline for your own store, then use external benchmarks only when their methodology fits.

A positive NPS means you received more promoter answers than detractor answers. It does not establish that most customers are promoters, that a particular customer will buy again or that revenue increased because of a change you made.

What the score actually tells you

NPS subtracts the percentage of detractors from the percentage of promoters. Scores of 9–10 are promoters, 7–8 are passives and 0–6 are detractors. Passives count in the denominator, even though neither group in the subtraction includes them. Bain's scoring explanation gives the definition; our calculation guide works through the arithmetic.

Keep all three group counts beside the score. Different distributions can produce the same result and suggest different follow-up questions.

Match the kind of NPS before comparing numbers

Bain distinguishes competitive benchmark, relationship and experience NPS. A score collected after a purchase and a broader relationship survey answer different questions. A competitive research program also uses a different collection process from a brand surveying its own customers.

For a Shopify store, the practical distinction is timing. A first-time buyer on the Thank you page has just placed an order. They can discuss choosing a product and checking out, but usually cannot yet judge delivery or using that product. Repeat buyers may bring earlier experience to the same question.

Record what customers could reasonably assess when they answered. Avoid comparing a checkout survey directly with an after-delivery survey and labeling the difference a change in loyalty.

How to evaluate an ecommerce benchmark

Before using a category average, look for its source and methodology. An unexplained “beauty brands typically score X” table is not enough to set a target for your store.

If that information is missing, label the comparison as uncertain. Do not turn an unsupported range into a promise about retention, growth or customer satisfaction.

Build a baseline you can repeat

Choose the decision before the target. For example: “What makes repeat buyers hesitate to recommend our store?” is a more actionable starting point than reaching an arbitrary score.

1. Define the audience. Decide whether the question is for all orders, first orders or repeat orders. Keep a record of any additional conditions.
2. Choose the collection point. Use the same page and question wording for the period you intend to compare.
3. Record the counts. Save promoters, passives, detractors and total valid NPS answers alongside the score.
4. Read the reasons. Use an optional follow-up to identify issues worth investigating. Keep the follow-up relevant to the experience the buyer has had.
5. Record changes. Note changes to the survey, audience, store, support or delivery process before reviewing the next period.

There is no universal minimum number of responses that makes a result trustworthy. Small groups can move sharply when one answer changes. Larger samples can still omit customers who chose not to respond. Report those limits and avoid treating a small movement as statistically established without analyzing the actual data.

An example of a fairer comparison

The following numbers are illustrative, not OrderSurvey customer results:

The observed score increased by ten points. That is a description of these answers. Before attributing it to your work, check whether the audience, question, timing and response collection changed. Look for supporting comments and operational evidence. These counts alone do not prove that a store change caused the increase or that revenue improved.

Set up a consistent survey in OrderSurvey

Use an NPS question on the Thank you and/or Order status page, with a conditional follow-up such as “What could we have done better during your purchase?” Test the low, middle and high score paths before publishing. The NPS follow-up guide covers that workflow.

Use first/repeat-order audiences and supported order conditions when they fit the research question. Existing customer-tag targeting is unavailable on these online blocks. See the segmentation guide for supported conditions and the role of the Thank you block in first-order classification.

The Order status block does not wait for delivery. It runs on eligible page visits; the app has no delivered-status or days-since-delivery display condition. If you need feedback after product use, establish that context explicitly instead of assuming the page identifies it.

Shopify POS is a separate placement with different capabilities. The conditional online workflow above is for Thank you and Order status pages; do not assume the same branching and targeting apply at the retail counter.

The dashboard's NPS card summarizes valid NPS-question answers from the last 90 days across surveys. It is not automatically a separate score for each audience or calendar month. With Unlimited, export CSV and isolate the relevant survey, question and dates in your spreadsheet; count NPS answer rows rather than every question in the export.

Turn the baseline into a useful next step

Choose one repeated concern to investigate, assign an owner and record what you change. Track whether that concern appears again, alongside the score and response count. If a low NPS answer needs attention, Unlimited can attempt a Slack, Teams or custom webhook alert after the response is saved. Your team still owns customer contact and resolution.

Free includes 100 responses across the store per rolling 30 days and conditional follow-ups. Unlimited is $24 USD/month for unlimited responses, CSV export and webhook alerts, among its features. Compare plans.

Install OrderSurvey to start a consistent feedback program, or book a personal setup walkthrough to check both survey blocks and your first response together.

Frequently asked questions