tl;dr
Customers only come back if they're satisfied with your product — understanding why they buy is the foundation of retention
Five sources for customer insight: support tickets, reviews, surveys, direct calls, social comments
Three things to understand: who they are, what they did, what you did
Use order/customer tags to push qualitative insights into Klar — then analyse them in the Cohort Comparison report to find what drives repeat rates and LTV
A customer will only come back if they're satisfied with your product. Set the product aside and there's very little else that matters. So understanding what a customer cares about — and how they use what they bought — is the most important input for your entire retention strategy.
Most brands don't spend enough time on this. The goal is to combine qualitative customer understanding with the data you already have in Klar, and use that to build a strategy that works for your specific customers.
Five ways to understand your customers
1. Customer service tickets — what do people ask about, and what do they complain about? This is often the most honest signal you have.
2. Reviews — what do customers write publicly, on your own review form or on marketplaces? Look for recurring themes in both positive and negative reviews.
3. Surveys — especially post-purchase surveys. They're the most scalable way to capture customer insight directly after a transaction, when motivation to respond is highest.
4. Direct calls — call customers shortly after they buy and ask them a few questions. It takes time but gives you depth that surveys can't match. A simple opener: "I saw you just bought — do you mind if I ask you a few questions?"
5. Social comments — what do people comment under your ads and organic posts? Unfiltered and unsolicited, it's often where the most honest reactions live.
If you're at scale, consider assigning one person to collect and organise all of this. It doesn't need to be a full-time role at most companies, but someone should own it and regularly brief the wider team.
Three things to understand about each customer
Once you're collecting input, organise it into three buckets:
Who are they? — who are they buying for (themselves, a family member, a friend)? What were their concerns before buying? What benefits were they hoping for? What were they using before to solve the same problem? How frequently do they actually use the product?
What did they do? — which product did they buy? Which channel did they come from? Did they subscribe to your newsletter? Did they complete an onboarding quiz? What was their purchase behaviour?
What did we do? — did we invite them to a loyalty programme? Did we add a handwritten card to the order? Did we upgrade them to express shipping? Did we follow up to check if they liked the product? Understanding the actions you took — and their impact — is what lets you optimise retention spend over time.
Combine customer understanding with data using tags
Tags are the best way to make qualitative insights analysable. When a customer responds to a post-purchase survey or contacts support, categorise their response and push it back as an order or customer tag in your shop system. Consistent naming conventions are key — for example, prefix all benefit-related responses with main_benefit_.
Once the tags are in Klar, go to the Cohort Comparison report and filter by tag. You'll be able to see how repeat rates and LTV differ by customer segment.
What this looks like in practice
Here's a real example: a brand asked customers what the main benefit they got from the product was. The responses became tags. In the Cohort Comparison report, the differences were significant:
"Looks good" → 30% repeat rate
"More active" → 35% repeat rate
"More confident" → 50% repeat rate, ~35€ more margin LTV after one year
That single insight has implications across acquisition (which benefit to lead with in ads), activation (how to frame early communications), and retention (reminding customers of the transformation they've experienced).
You only need one or two insights like this to meaningfully shift your retention numbers. It takes time to set up, but once it's running it's largely automated — and it becomes the foundation everything else is built on.
Part of the retention video series: