The short version
- Retention is often held back by questions your data could answer but is not being asked.
- Compare customers who stayed with those who left, using cohorts by joining month.
- Test every finding with a control group, because correlation is not cause.
- Fix tracking and website crawling first so the data reflects real demand.
Customer retention is usually held back by a handful of questions your data could answer but is not being asked. This guide lists ten of those gaps, how to spot each in your own numbers, and what to do about it.
Retention matters because keeping a customer costs less than winning a new one. Bain & Company’s well-known estimate is that increasing retention by as little as 5 per cent can boost profits by as much as 95 per cent. Treat it as a rough indication rather than a promise, since the effect varies a lot by business.
None of the checks below needs a data scientist. A spreadsheet export from your CRM, analytics tool and support desk is enough to start.
What is a data gap in customer retention?
A data gap is a question about why customers stay or leave that you cannot answer from the data you currently look at. Examples include which early actions predict loyalty or when churn actually peaks. Closing a gap means finding the evidence, then changing something in onboarding, support or communication.
Ten data gaps worth checking
Gap 1: you do not know what a retained customer did in their first 30 days
Look at your best long-term customers and list what they did early on: finished setup, used the core feature, invited a colleague, booked a call. Then check how many new customers reach the same point. If most do not, you have found an activation gap, and onboarding is the place to fix it.
What to do: Build an onboarding checklist around those actions and send a short email sequence that points to each one.
Gap 2: you measure satisfaction but not how deeply people use the product
A satisfaction score tells you how someone felt last week. Usage depth shows whether they have built the product into their work. Compare churn between customers who use one or two features and those who use five or more. Where the gap is wide, adoption is a better early warning than a survey score.
What to do: Have your success team contact satisfied but low-usage customers and show them features that fit their job.
Gap 3: you never checked how first-response time affects churn
Most teams track time to resolution. Customers often judge you on time to first reply. Pull support tickets, group them by how long the first reply took, and compare churn over the next 60 days. You may find a threshold where waiting starts to cost you customers, even when the problem is later solved.
What to do: Set a first-response target separate from the resolution target, and write reply templates that acknowledge the issue quickly.
Gap 4: you assume customers leave at the first renewal
Many businesses watch the first renewal closely and then relax. Plot cancellations by customer age in months. If a second peak appears later, buyers are probably re-judging value for money once the novelty has worn off.
What to do: Schedule a value review before the second renewal, showing what the customer has achieved with your product or service.
Gap 5: you do not know where your best new customers come from
Paid referral schemes and affiliates get budget because they are easy to count. Ask new customers how they found you and compare retention by source. Often the customers who arrive through recommendations from happy users stay longer than those from paid channels.
What to do: Identify your most active customers, give them early access or direct contact with your team, and make sharing effortless.
Gap 6: you write off dormant free users and lapsed buyers
People who signed up, tried the product briefly and went quiet are rarely worth nothing. Segment them by the last thing they did. Someone who used one specific feature has a clear reason to return, and a message about that feature will land better than a generic newsletter.
What to do: Run a short win-back email sequence per segment and test an offer of a guided restart call.
Gap 7: you collect NPS scores but do not follow up on answers
Net Promoter Score asks how likely someone is to recommend you on a 0 to 10 scale. Promoters score 9 to 10, passives 7 to 8 and detractors 0 to 6, per the official Net Promoter System page. The number matters less than what you do with a detractor’s reply.
Look at the individual replies from detractors and passives, not only the average. Alert a named person when a low score arrives, contact the customer within a day, and track how many of those follow-ups end well.
Gap 8: your customer health score is based on guesses
Many health scores are built from whatever data is easy to export. Test them against history: did the customers who churned actually score low a month before? Quieter signals often predict better, such as fewer logins across a whole team, no support contact for months, or reports that nobody opens.
What to do: Rebuild the score from the signals that preceded real cancellations, then give each alert level a defined playbook.
Gap 9: you treat onboarding calls as optional
If only a small share of customers take up an onboarding call, compare their retention with those who did not. Be careful with the result: engaged customers are more likely to book a call, so the call may not be the cause. Run a small test where you proactively book calls for a random group.
What to do: Make the call standard for higher-value plans and ask the customer to state their goals for the first 90 days.
Gap 10: you send every product update to every customer
Check what happens to usage after a product announcement email. If less technical users go quiet or unsubscribe, the message may be making your product look more complicated than it is.
What to do: Segment announcements by role and plan, and write them around what the customer can now do rather than what you built.
From our work: When we review a client’s website and marketing, the most common gap is simple: leads and customers are counted, but nobody has looked at what the good ones did differently. Fixing tracking and reporting first usually shows quickly where follow-up emails, onboarding pages or enquiry forms lose people.
Not sure where you rank?
We’ll check your map position, your Google profile and your listings, and send you what to fix first. Free, no obligation.
How do you find retention patterns in your data?
Group customers into cohorts by the month they joined, then compare how many are still active after one, three, six and twelve months. A cohort is a group of customers who share a starting point. Differences between cohorts show where retention improves or drops.
Google Analytics 4 has a cohort exploration report that groups users by a shared attribute, such as acquisition date, and shows daily, weekly or monthly return behaviour. It tracks website and app behaviour. For paying customers, add an export from your CRM or billing system.
- Ask one specific question, such as why customers leave after 14 months.
- Pull the data from your CRM, analytics, support desk and invoices into one sheet.
- Compare stayers with leavers, not averages.
- Write a hypothesis, for example that customers who never use feature X cancel sooner.
- Test a small change and measure it against a group that did not get it.
Do technical website problems distort your data?
Yes. If search engines cannot crawl or index your site properly, you lose visitors before they enter your data at all, and tracking errors can skew what you do see. Fix crawling and tracking first so the data you analyse reflects real demand.
Our guide to five technical SEO errors that kill traffic is a good starting point. Our SEO services cover audits and fixes, and our email marketing work covers the onboarding and win-back sequences mentioned above.
Frequently asked questions
Customer churn is the share of customers who stop buying from you or cancel a subscription in a given period. You calculate it by dividing the customers lost in that period by the customers you had at the start. Tracking it by cohort shows when people leave, not just how many.
Find when and why customers leave, then change that moment. Compare customers who stayed with those who left, spot the early behaviours that separate them, and act on them in onboarding, support and communication. Test each change on a small group before rolling it out, and measure retention again afterwards.
There is no single good score, because it depends on your sector and customer type. NPS is the percentage of promoters (scores 9 to 10) minus the percentage of detractors (0 to 6). Compare your own score over time and against your direct competitors, and follow up individual replies.
Start with the date each customer joined, what they bought, when they cancelled or stopped buying, and basic usage or order history. Add support tickets and survey answers when you can. A clean spreadsheet export from your CRM and billing system is enough for a first cohort comparison.
Check headline retention monthly, so you see cohort changes before they become a pattern. Run a deeper review each quarter to test new hypotheses. Review again after any major change to pricing, onboarding or your product, since those are the moments retention is most likely to move.
Get help closing your data gaps
If you want a second pair of eyes on your tracking, reporting or the website your customers start from, contact LineUp. See more guides on the LineUp blog.
Search, ads and video for businesses that want more customers. Every guide is based on what we see in client accounts. See our services
