NPS & SurveysCustomer Experience

NPS Case Study: Pre-Delivery vs Post-Delivery NPS

First published Feb 20, 2023Updated October 21, 20267 min read
Alexandra Panaitescu, Content Marketing Specialist
Alexandra Panaitescu
Content Marketing Specialist
Published: Feb 20, 2023Updated: Oct 21, 2026
NPS case study: a delivered order box with a survey envelope and a blue hourglass for survey timing
Quick Answer
This NPS case study follows two eCommerce stores that measured Net Promoter Score before and after delivery. The marketplace scored 71 before delivery and 82 after, so it fixed its website and support; four months later the scores were 78 and 84 [Omniconvert, marketplace NPS case study]. The fashion retailer scored 77 before delivery but only 51 after, a sign that its promise was higher than its product and service, and the post-delivery score fell to 44 when nothing changed.
Key Takeaways
  • Measure NPS twice: right after the order (website and promise) and a few days after delivery (product and service).
  • Read a low pre-delivery score as an online experience problem, and a low post-delivery score as a product, delivery or service problem.
  • Add an open follow-up question, because the reasons tell you what to fix.
  • Route every 0 to 6 score to customer support automatically, starting with your best RFM segments.
  • The marketplace raised pre-delivery NPS in its best segment from 72 to 89 in four months.
  • The fashion retailer that ignored a 26-point gap saw its post-delivery NPS fall from 51 to 44.
2,500+ brands use Omniconvert 70,000+ experiments run 13 years of CRO expertise Best-segment pre-delivery NPS: 72 to 89 in four months [Omniconvert, marketplace NPS case study]

This NPS case study shows what two eCommerce stores learned when they measured Net Promoter Score twice: once right after the order and once after delivery. The same survey gave opposite results. One store had a website problem; the other had a product and service problem. Each needed a different action plan, and only one of them acted on it.

Measuring NPS before and after delivery is one of the first steps we take with clients in our Customer Value Optimization work. For the basics of the metric, read the Net Promoter Score guide.

Why measure NPS before and after delivery?

Pre-delivery NPS is defined as the score customers give right after they place an order, and it rates the website and the promise. Post-delivery NPS is defined as the score they give a few days after they receive the order, and it rates the product and service.

One order holds two experiences: buying on the site and living with what arrives. A single survey mixes them. Two surveys show the gap, and the gap tells you where to act:

  • Low pre-delivery NPS: fix the online experience, such as search, checkout, payment and speed.
  • Low post-delivery NPS: fix the products, delivery and service, or stop promising more than you deliver.

Clients often ask how their score compares with industry benchmarks. That question matters less than whether you meet your own customers' expectations, because product type and audience change what a normal score is. Our guide to what a good NPS score is covers the context. For the background of the Net Promoter Score as a strategy tool, see the BYU strategy notes; for the math, see the NPS formula.

The two clients in this NPS case study

The case study covers a large multi-category marketplace and a mid-size fashion retailer. Neither measured customer satisfaction proactively before the project; both reacted only to complaints.
Source: Omniconvert (client profiles)
Client Profile Catalog Scale
1. Marketplace 11 years on the market; books, electronics, cosmetics, clothing, food, home and garden 110K+ SKUs 1.3M orders per year, over 1 million customers
2. Fashion retailer 5 years on the market; urban streetwear for men, women and children 2K+ SKUs sold 89K delivered orders, 59K+ active customers (2020)

Both had the same pain point: no qualitative feedback. So they did not know whether they had a product problem or a marketing problem, which processes to fix first, what stopped customers from becoming repeat customers, or what would keep them loyal.

How the NPS system worked

The setup combined two NPS surveys with an open follow-up question, a link to customer data and an automatic support ticket for every score from 0 to 6, rolled out to the most valuable segments first.
  1. Survey before delivery
    An NPS survey on the thank you page rated the buying experience while it was fresh.
  2. Survey after delivery
    A personalized email a few days after the customer received the order rated the product and service.
  3. Ask why
    A follow-up question after the 0 to 10 rating asked what went wrong and what the store could do better.
  4. Alert support on detractors
    Each score from 0 to 6 created a ticket in the support tool and an email to the support team, who then contacted the customer.
  5. Roll out by customer value
    The alerts started with the top three RFM segments, then expanded to first-time and high-potential customers, so the support team had time to adapt.

Omniconvert Commander (formerly Nexus) held the NPS answers next to each customer's RFM segment and connected to the clients' ticketing systems, so support could prioritize by customer value. We encourage clients to adapt our eCommerce Blueprint to monitoring and using Net Promoter Score. It is a detailed framework, but clients who adopt it can act before an unhappy customer leaves.

Results for client 1: the marketplace

The marketplace's pre-delivery NPS was 11 points below its post-delivery NPS, so the problem was on the website. After four months of fixes, pre-delivery NPS rose from 71 to 78 and post-delivery NPS from 82 to 84.

One month after launch, the marketplace had thousands of answers. Pre-delivery NPS was 71 and post-delivery NPS was 82 [Omniconvert, marketplace NPS case study]. Customers were happy with what they received but struggled to buy it.

The open answers explained why. The follow-up question matters because a score alone does not say what to fix, which is one of the known pitfalls of NPS. Customers reported problems with search, filters and online payment, plus technical errors and low site speed. This happened while the marketing team celebrated a new mobile design, and 80% of traffic was on mobile. Customers also did not understand the fees on their order, how to use loyalty points, or the delivery estimates for carts with products from several sellers.

The real-time feedback gave the development team clear priorities and made the issues visible to the whole company, so most of them were fixed quickly. The answers also became input for the conversion rate optimization roadmap and for clearer communication about fees and delivery. Pro tip: when sales drop, ask a follow-up NPS question such as "What could almost stop you from shopping with us today?" The answers are the best user experience audit you can get.

The NPS survey was also connected to the support CRM, where agents saw the score, the feedback and the customer's RFM segment together. The marketplace now treats loyal, profitable customers as VIPs, with dedicated support representatives and benefits such as discounts or shipping options.

Source: Omniconvert (marketplace NPS case study, month 1 vs month 4)
Group Pre-delivery NPS Post-delivery NPS
All customers 71 → 78 82 → 84
Soulmates (best segment) 72 → 89 77 → 88
Lovers (second-best segment) 72 → 83 About 90, stable

The largest gains came in the Soulmates segment, the customers who buy most often and spend most. Matching NPS answers with RFM segments showed who the best and worst customers were, what they bought and when, and what friction they met. A post-delivery score of 82 was already high for a store with 1.3 million orders a year, so the gain there was small; moving it further needs an over-delivery approach and personalized treatment.

Results for client 2: the fashion retailer

The fashion retailer's pre-delivery NPS was 77 but its post-delivery NPS was 51, a 26-point gap. It kept running flash sales, and four months later the post-delivery score had fallen to 44.

One month after launch, the retailer had the opposite result: pre-delivery NPS of 77 and post-delivery NPS of 51 [Omniconvert, fashion retailer NPS case study]. The website and campaigns raised expectations that the products and service did not meet.

The cause was clear in the open answers. Low product quality appeared in 48% of responses and low service in 29%. The brand acquired customers almost only with discount campaigns and flash sales, which attract buyers who rarely become loyal and cost margin. Making the case for qualitative research was hard, because discounts always hit the short-term sales target.

The brand kept sending flash sales, even to its most valuable customers. Pre-delivery NPS stayed at 77, but post-delivery NPS fell from 51 to 44. A top customer who orders a special outfit for a birthday and gets it late, or with missing buttons, is not a small problem.

The first step was to find the "toxic" products: those with low NPS. The children's clothing category and a newly launched type of men's jackets hurt the experience most. The store then began to promote products that existing customers rated well. The ongoing plan:

  • Treat NPS feedback in real time.
  • Increase the response rate of the post-delivery survey.
  • Remove toxic products.
  • Add more products that customers buy again, and find high-potential products.

The first result was a change of mind: the decision makers accepted that NPS is central to their customer value work. Repairing negative experiences takes far more time and effort than preventing them.

When to send an NPS survey in eCommerce

Send one NPS survey right after the order, on the thank you page or by email, and a second one a few days after delivery. Each survey has a different owner and a different fix.
Source: Omniconvert
Moment What it rates Who acts on it
Thank you page, right after the order Website, checkout, payment, marketing promise eCommerce, CRO and development teams
Email, a few days after delivery Product quality, delivery, packaging, service Product, logistics and customer support teams

A common mistake in practice: the post-delivery email is scheduled a fixed number of days after shipping, not after delivery. Customers whose parcel is late then answer before it arrives, rate the wait instead of the product, and pull the post-delivery score toward the pre-delivery one. The gap that this case study depends on disappears. Trigger the second survey on the delivery confirmation instead. The NPS survey best practices cover channels and sample size, and a post-purchase email flow with NPS shows how to automate the second survey.

What this NPS case study shows

The same metric, measured at two moments, gave each store a different diagnosis. The store that acted on it improved; the store that ignored it lost ground.
  • The gap is the insight: compare the two scores before you compare yourself with an industry average.
  • The open answer is the fix: the reasons named search, payment, fees, product quality and service.
  • Segment the score: the best gains came in the most valuable RFM segments.
  • Discounts hide problems: sales can look healthy while post-delivery NPS falls.

Once NPS is linked to customer data, you can read it by delivery stage, RFM segment, location, brand, category and channel, and find the exact part of the journey to fix. For the link between these scores and repeat purchases, read our guide to the Net Promoter Score for eCommerce, and for more examples, see these CRO case studies.

NPS Case Study FAQs

1What is the difference between pre-delivery and post-delivery NPS?

Pre-delivery NPS is asked right after the order, so it rates the website, checkout and marketing promise. Post-delivery NPS is asked a few days after the customer receives the order, so it rates the product, delivery and service. A gap between the two shows which part of the journey to fix first.

2When should you send an NPS survey in eCommerce?

Send two surveys: one on the thank you page or right after checkout, and one a few days after the order is delivered. Trigger the second survey on the delivery date, not the shipping date, so customers rate the product they actually received.

3What did the marketplace case study show?

The marketplace started with a pre-delivery NPS of 71 and a post-delivery NPS of 82, so the problem was the buying experience on the site. After it fixed search, filters, payment and speed issues and gave its best customers priority support, pre-delivery NPS rose to 78 and post-delivery NPS to 84 in four months. In its best segment, pre-delivery NPS rose from 72 to 89.

4What did the fashion retailer case study show?

The fashion retailer had a pre-delivery NPS of 77 but a post-delivery NPS of 51, so its promise was higher than what customers received. Low product quality appeared in 48% of responses and low service in 29%. It kept running flash sales, and the post-delivery score fell to 44.

5How do you act on low NPS scores?

Create a support ticket automatically when a customer scores 0 to 6, start with your most valuable customer segments, read the open follow-up answers to find the cause, and fix the products, pages or processes that come up most often. Then measure the score again after the change.

6Should you compare your NPS with industry benchmarks?

Benchmarks give context, but they matter less than your own trend and the gap between your pre-delivery and post-delivery scores. Product type and audience change what a normal score is, so first check whether you meet your own customers' expectations.

Measure Both Moments, Then Act on the Gap

Set up two NPS surveys this week: one on the thank you page and one triggered a few days after delivery, each with an open follow-up question. After a month, compare the two scores. If the pre-delivery score is lower, audit and test the buying experience; if the post-delivery score is lower, look at products, delivery and service before you spend more on campaigns.

Route every 0 to 6 answer from your best customers to support the same day. Omniconvert Explore runs the on-site NPS survey and the A/B tests that check each fix.

Alexandra Panaitescu, Content Marketing Specialist
Content Marketing Specialist
Alexandra Panaitescu is a B2B content marketing specialist with over 8 years of experience building data-driven content strategies and inbound campaigns that help businesses grow, from generating qualified leads to establishing brand authority and revenue.

Ask customers at the right moment, read why they scored you low, then test the fix. With Omniconvert Explore, you can run NPS surveys and A/B tests on the same platform.

See Explore →

Find out where your customer experience breaks

Omniconvert Explore runs NPS and on-site surveys, A/B tests and personalization in one tool, so you can ask customers at the right moment and test the fix they ask for.