How to increase conversions and repeat purchases in an e-shop with customer feedback

Picture this. You're standing at a market with your stall, and people come up, browse the goods, and often willingly tell you in conversation what's missing, what they'd want differently, or what extras they'd like. This feedback is immediate, and you can react to it on the spot.

Online, it's a different story. A customer arrives and either buys or leaves. And often, you have no idea why. The direct dialogue is missing, and feedback has to be actively collected through reviews, ratings, or surveys. That's why it's all the more important to capture it at the right moment and connect it across the entire customer journey.

Continuous feedback collection can reveal new trends in customer behavior and flag changes in their expectations at the right time. For e-shops, it helps better understand where customers run into problems, what motivates them to buy, and what, on the other hand, leads them to abandon the cart or leave the site. Thanks to this, you can continuously optimize the customer journey, increase the conversion rate, reduce cart abandonment, and build greater customer loyalty — all of which translate into long-term growth for the entire e-shop.

Let's now look at three reasons why it makes sense to collect customer feedback.

Keeping a customer is cheaper than acquiring a new one. Acquisition costs in e-commerce keep rising long-term and margins are dropping. A customer who's already bought from you and was satisfied will come back at a fraction of the cost it would take to attract a new one.

Silent customer dissatisfaction costs money. A customer who has an issue usually won't tell you, and quite often won't even leave a review. They simply don't complete the order, don't come back, and don't recommend you. Without active feedback collection, the most important signals just slip away from you. Instead of clear answers, you're left watching a slow decline in performance in your data.

Collecting feedback is the cheapest and fastest research option. Rather than running A/B tests or paying an agency for a UX audit, you can simply ask the people who buy from you. Customer feedback often surprises you with how concrete and valuable it is. It's not just general impressions — you get specific input you can immediately turn into a better-functioning e-shop, higher conversions, and happier customers.

Feedback has the most value when it comes as close as possible to the moment the customer experienced what you're asking about. Asking a month after purchase how delivery went is too late. Asking about website usability a month after the visit is completely off. Here's an overview of the key moments — when to ask and what.

1. On cart abandonment (exit-intent)

When a customer leaves the page with a full cart, it's the right moment to ask why. A short pop-up with a single question:

"What stopped you from completing the order?"

This answer gives you a kind of insight you won't read out of analytics. Classic reasons tend to be shipping fees, a missing payment method, the requirement to register, or simply price comparison. Without asking, you're just guessing.

Example: Cart drop-off in e-commerce has long hovered around 65–75%. If your exit-intent survey reveals that 40% are leaving because of shipping costs, you have a concrete hypothesis to test (a free-shipping threshold, a different carrier). Even partially removing one reason for leaving means a measurable lift in conversion.

2. After completing the order (thank-you page)

Right after the order, the customer has a fresh impression of the buying process. It's the ideal moment for a CES question — that is, how easy was it to buy — or a short survey on what they'd improve. Careful: at this stage, asking about the product is premature, the customer hasn't seen it yet. At this point, ask primarily about the buying process.

Example: CES after purchase will show you the friction points that probably stopped people who didn't complete the order either. When 30% of customers rate the process 1–3 out of 5, you have a direct list of things that are slowing conversion for everyone else too.

3. After delivery

This is where it makes sense to ask about the product, packaging, and delivery. The ideal timing is 1 to a maximum of 3 days after delivery — that's the window when the customer has already tried the item but the impression is still fresh. Send a short email with an NPS question or CSAT on specific aspects, like quality, delivery time, and whether the item matches the description.

Example: This is the most important moment for retention. A satisfied customer at this point is your next repeat purchase. A dissatisfied one you don't reach out to simply won't return. A proactive response to a low NPS score (apology, resolution, compensation) measurably increases the chances the customer comes back.

4. On returns or complaints

A return is one of the most valuable moments for feedback — but few people ask. The reason for the return shows you systemic problems that are costing you money and that you wouldn't otherwise see.

Example: If 25% of returns point to "doesn't match the description/photo," you know the problem is on the product page — and the same problem, to a lesser degree, is also putting people off buying, because customers point it out in reviews. Better photos and descriptions raise conversion and reduce return costs in one move.

5. Continuously, with regular customers

For customers who buy repeatedly, it makes sense to run a broader survey twice a year. Ask, for example, what works for them, what bothers them on the contrary, and what new things they'd appreciate. These are your most valuable people and they're usually willing to answer because they already have a deeper relationship with your brand.

Example: Such surveys often bring suggestions for products you don't yet carry, or services (faster delivery, a loyalty program) that would increase purchase frequency. Regular customers tell you themselves how to make them buy more.

Key metrics: NPS®, CSAT, CES

Three standard metrics that work together, each measuring something different. You don't have to track all three at once, but you should know what each one says so you can pick the right one for the right moment.

NPS® – Net Promoter Score®

Question: How likely are you to recommend our e-shop to your family or friends?

NPS measures loyalty and willingness to recommend. Calculation: percentage of promoters (9–10) minus percentage of detractors (0–6). The result is a number between −100 and +100.

In e-commerce, NPS® is usually sent a few days after delivery. It works as a main summary indicator that you track over time. More important than the absolute value is the trend — a drop from 45 to 30 between quarters is a more serious signal than an absolute value of 30.

What to do with low NPS®: send detractors a follow-up question, "What would we have to change for you to recommend us?" These answers are among the most valuable data you can get.

CSAT – Customer Satisfaction Score

Question: How satisfied are you with [a specific thing]?

CSAT measures satisfaction with a specific transaction or aspect — that is, not the overall relationship, but one thing. It's well suited for questions like:

• "How satisfied are you with delivery speed?"

• "How satisfied are you with customer support communication?"

• "How satisfied are you with the product quality?"

Calculation: percentage of 4 and 5 ratings out of total responses. A value of 80% and above is a good standard, below 70% is a signal to act.

CSAT's strength is in its specificity. When you have CSAT on delivery at 65% and CSAT on products at 90%, you know exactly where to direct improvement. NPS would never tell you that.

CES – Customer Effort Score

Question: Was it easy for you to complete the purchase / resolve the complaint / find what you were looking for?

CES measures how much effort the customer had to spend. It's based on research showing that reducing customer effort has a bigger impact on loyalty than aiming for an "exceptional experience." People remember when something was easy, not when it was breathtaking.

In an e-shop, CES makes sense:

• after purchase — how easily did the checkout go

• after an interaction with customer service — how easily was the issue resolved

• on returns

When to use which metric

NPS for the overall feeling, CSAT for specific satisfaction with something, CES for effort in a process. In practice, an e-shop usually only needs a combination of two — NPS® after delivery as the main indicator + CSAT or CES at specific key moments (checkout, support, delivery).

How to turn feedback into real decisions

Collecting data is easy. Working with it is where most e-shops fail. A few principles that turn it into something useful:

Categorize responses. Split open-ended survey responses into categories (delivery, product, website, support, pricing). The moment you see that 30% of complaints are aimed at one thing, you have a clear set of priorities.

Combine qualitative with quantitative research. "People are complaining about shipping" is a hypothesis. Once you back it up with the fact that cart drop-off is 60% and the open-ended responses from the exit-intent survey confirm it, you have data you can make decisions on.

Close the loop with the customer. When someone gives a low rating and writes you something, get back to them. Not with an automated reply, but for real. These customers are often convertible back to satisfied — and either way, they'll see that feedback leads somewhere.

Report internally. A monthly report with NPS® development, CSAT at key points, and the top 3 complaints from open-ended responses. Without this, feedback won't go anywhere — it'll stay in the survey tool and no one will use it.

The most common mistakes e-shops make when collecting feedback

Surveys that are too long. A maximum of 5 questions for shorter contact (post-purchase), 10 for a periodic survey of regular customers. A 20-question survey gets either low response rates or shallow answers.

Bad timing. Sending a survey too early, before the customer has actually tried the product, or too late, when they no longer remember the purchase, leads to irrelevant answers.

Asking and then doing nothing. If a customer writes that they dislike long delivery times, and six months later writes the same thing again, they won't bother answering next time. They see that feedback leads nowhere.

Tracking only the NPS® numerical score. Without follow-up analysis of written comments, this leads to ignoring context and an inability to effectively address the root causes of customer dissatisfaction. Analyzing text comments is key for understanding specific problems, identifying hidden risks even among supporters, prioritizing investments, and gauging customer emotions.

Start measuring customer feedback with the SentiSnap app

Feedback isn't about having a long list of complaints. It's about having a system that consistently shows you where your e-shop is losing money and what specifically to improve first. In SentiSnap, you set up surveys for every point of the customer journey. Collect feedback at the right moment — when leaving the site, after purchase, after delivery, and on complaints. Then track everything clearly in one place in real time and have an immediate overview of what's working in your e-shop and what needs to be improved. Start for free and find out what your customers are really telling you.