Retention is the metric that determines whether your business is sustainable. A business with high retention grows compoundly — every new customer adds to a base that keeps paying. A business with low retention is running on a treadmill — you have to acquire customers constantly just to stay flat.
Most "retention strategy" content is generic: "focus on customer success," "communicate proactively," "measure NPS." These aren't wrong, but they aren't specific enough to act on.
Here are the 10 tactics I've seen actually move retention numbers, ranked by impact-to-effort. The first three, done well, produce more retention gain than the other seven combined.
First: what retention actually IS
Two distinct concepts get called "retention":
Customer retention rate: the percentage of customers who are still with you at a given time. If you have 100 customers on Jan 1 and 88 on Jul 1 (with no new customers added), your 6-month retention is 88%.
Net revenue retention (NRR): the percentage of revenue you keep from your existing customer base after considering upgrades, downgrades, and churn. If your existing base was paying $100k/month and now pays $110k/month (upgrades exceeded churn), your NRR is 110%.
NRR is the metric that matters most. A SaaS with 105%+ NRR grows even without new customers. Below 90% means you're bleeding.
Benchmarks:
- Below 70% NRR: business is failing
- 70-85%: struggling
- 85-95%: healthy for consumer, weak for B2B
- 95-105%: good for both
- 105-120%: excellent
- Above 120%: rare and world-class
The 10 tactics below are what actually gets you from "struggling" to "excellent."
Tactic 1: Nail activation for new customers (highest impact)
The problem: most customers who churn do so in the first 30-60 days. They signed up, tried to use the product, didn't get value quickly, disappeared.
The fix: obsess over the first-week experience. Specifically, get every new customer to their "aha moment" — the specific action that predicts they'll stick around.
Concrete moves:
- Identify your activation event (see the SaaS onboarding fixes post for the specific method)
- Measure the percentage of new signups who hit the activation event within 7 days
- For every new signup who doesn't activate in 24 hours, send a personal email asking what's blocking them
- Iterate the onboarding flow every 2 weeks based on data
Expected impact: improving 7-day activation by 20 percentage points typically improves 90-day retention by 15-25 percentage points. This is by far the highest-leverage retention work.
Tactic 2: Talk to customers who churned (second highest impact)
The problem: most founders don't know WHY customers churn. They assume (usually wrong).
The fix: every week, contact 5 customers who churned in the last 30 days. Personal email, from you. One question: "You canceled recently — was there something specific that didn't work?"
Response rate: 25-40% typically reply. What they tell you is worth more than any dashboard.
Common patterns you'll discover:
- A specific feature is missing or broken (fixable)
- The value proposition wasn't clear (fixable via marketing/onboarding)
- Wrong customer fit — they shouldn't have signed up (fixable via better targeting)
- Their needs changed (rare — this is legitimate churn)
What to do with the info:
- If 3+ people say the same thing: high priority to fix
- If pattern is unclear: keep asking more people
- Track systematically in a shared doc
Expected impact: consistently doing this for 90 days typically produces 3-5 fixable insights that, when addressed, move retention 5-10 percentage points.
Tactic 3: Build a "resurrection" flow for dormant users
The problem: many customers don't formally cancel — they just stop using the product. If you don't reach out, they'll eventually cancel or fail to renew.
The fix: an automated (but human-feeling) flow for customers whose usage has dropped.
Concrete moves:
- Define "dormant" (e.g., no login in 14 days for a daily-use product)
- Trigger a personal email when a customer goes dormant: "Hey, noticed you haven't logged in — anything blocking you?"
- Follow up if no response: a specific tactical offer ("here's a quick way to [get value]")
- If still no response after 21 days: one final email offering help before considering cancellation
Expected impact: typically saves 15-25% of at-risk customers who would have churned silently.
Tactic 4: Make cancellation valuable (not painful, but valuable)
The problem: most cancellation flows are punitive or manipulative — "are you SURE you want to cancel? Read our 50 reasons to stay!" This damages trust even for the customers who do cancel (they'll remember and never come back).
The fix: make cancellation genuinely useful — extract value from the exit conversation.
Concrete moves:
- The cancellation flow includes ONE screen asking why (specific options + open text)
- Immediately after they cancel: a personal email from a real person offering to help transition their data OR asking if there's anything you could have done differently
- Never hide the cancel button; never make it require calling
- For high-value customers: offer a personal call as an option
Expected impact: the direct impact is small (5-15% recovery), but the indirect impact is huge — customers who cancel gracefully often come back later or refer others.
Tactic 5: Segment by customer type and treat differently
The problem: treating all customers the same means both under-serving your best ones and over-investing in ones who won't be there long anyway.
The fix: segment customers by expected lifetime value and treat differently.
Concrete moves:
- Score customers by combinations of usage, plan tier, and demographic signals
- Your top 10-20% get white-glove treatment (proactive check-ins, direct founder contact, priority support)
- Middle 60-70% get standard service
- Bottom 10-20% you might actually LET churn (they're not profitable to retain)
The counterintuitive part: trying to save every customer costs more than letting the wrong ones go. Focus retention effort on customers actually worth retaining.
Expected impact: typically 10-20% improvement in NRR because retention effort is now proportional to customer value.
Tactic 6: Add "expansion" motions, not just retention motions
The problem: you can only retain 100% of your revenue by not losing anyone. To grow NRR above 100%, existing customers need to spend MORE over time.
The fix: systematic expansion motions.
Concrete moves:
- Usage-based pricing: customers pay more as they use more (natural expansion)
- Additional seats: teams grow, they add more users
- Upsells to higher tiers: when they hit limits, offer the next tier
- Additional modules: add-on features priced separately
- Cross-sells: related products that solve adjacent problems
Warning: don't design expansion motions that feel exploitative. "You just hit your limit! Upgrade now!" screens the moment they hit any limit is a bad customer experience.
Expected impact: the difference between 95% NRR (retention only) and 115% NRR (retention + expansion) is enormous over time — it's the difference between flat and compounding growth.
Tactic 7: Invest in the "second week" experience
The problem: most founders focus on onboarding (week 1) and long-term customer success (month 6+). The "second week" — days 8-21 — is often ignored, and it's where a lot of silent churn happens.
The fix: deliberate touchpoints in the second week.
Concrete moves:
- Day 7: check-in from a real person, asking what they've achieved and what's next
- Day 14: relevant tip based on what they've done (or haven't)
- Day 21: soft push toward the "next level" of usage (a feature they haven't tried, an integration to set up)
Expected impact: typically 5-10% improvement in 30-day retention, because you're catching people BEFORE they drift away.
Tactic 8: Build "switching costs" into the experience
The problem: if your product feels easy to switch away from, some customers will switch away.
The fix: legitimately valuable switching costs (not manipulative lock-in — real value that gets lost by switching).
Concrete moves:
- Historical data they've accumulated (past reports, custom setups, saved templates)
- Custom configurations (rules, workflows, integrations they've built)
- Network effects (team members, shared resources, community)
- Learning curves paid for (training, expertise developed)
The distinction: BAD lock-in = "you can't export your data." GOOD switching costs = "you've spent 6 months configuring the product to your exact workflow, and switching means starting over."
Expected impact: cumulative — hard to attribute directly, but products with real switching costs consistently have 20-40% better retention than substitutes.
Tactic 9: Communicate value proactively (not just react to problems)
The problem: most communication with customers is transactional — bug fixes, feature announcements, billing. Very little is about VALUE received.
The fix: systematic value communication.
Concrete moves:
- Monthly "your usage report" showing what they accomplished with the product
- Quarterly business review calls for higher-value customers
- Milestone celebrations ("you just crossed 100 [thing]!")
- Case study opportunities for customers who've had big wins
The specific tactic that works well: at renewal time, send them a summary of their usage over the past year — quantified. "You used [Product] for 240 hours, tracked 1,847 items, and saved an estimated 96 hours vs the manual process." This reminds them of value at exactly the right moment.
Expected impact: 5-15% improvement in renewal rate for annual-billed customers.
Tactic 10: Fix the top 5 support issues (the boring one)
The problem: customers who repeatedly hit friction leave. Repeated small annoyances add up.
The fix: analyze support tickets. Fix the top 5 root causes systematically.
Concrete moves:
- Weekly review of support tickets by category
- Every quarter: identify the top 5 categories by volume
- Prioritize fixes to root causes (not just symptomatic support)
- Track ticket volume by category over time — you should see it declining as fixes ship
Expected impact: hard to measure directly, but every removed friction extends customer lifetime by weeks or months.
The 3 things that DESTROY retention
Before the tactics, know what NOT to do:
1. Silent pricing changes. Raising prices without warning or grandfathering is the fastest path to mass churn. See how to raise prices without losing customers.
2. Broken key features. If the specific feature that drives most of the value breaks — even for a week — you'll see churn 60-90 days later as those customers evaluate alternatives.
3. Disappearing support. Customers who had a support ticket that took 3+ days to resolve are 3-4x more likely to churn than customers whose tickets were resolved same-day. Support quality is a retention driver.
Measuring retention properly
Track these metrics weekly, at minimum:
- Gross retention: % of customers still paying (excludes upgrades)
- Net revenue retention: % of revenue kept from existing customers
- Cohort retention curves: what percentage of customers from Cohort X are still active after N months
- Churn reasons: categorized reasons why customers cancel
The specific insight cohorts reveal: are your NEW cohorts retaining better than old ones? If yes, your onboarding is improving. If no, you have a systemic problem.
A specific example
I worked with a founder whose SaaS had 68% net revenue retention — bleeding customers.
Diagnosis (took 2 weeks):
- Talked to 20 recent churners (Tactic 2): main pattern was "we never really got it working for our workflow"
- Analyzed activation (Tactic 1): only 24% of new signups completed the activation event within 7 days
- Analyzed support tickets (Tactic 10): 40% of tickets were about one specific onboarding step that was confusing
Fixes (over 3 months):
- Rewrote the confusing onboarding step (moved friction from month-1 to no friction)
- Added a personal email to any user who didn't activate within 48 hours (Tactic 3)
- Started a monthly "your usage report" email (Tactic 9)
- Segmented top customers for white-glove attention (Tactic 5)
Results:
- 7-day activation went from 24% to 51%
- 90-day retention went from 62% to 79%
- Net revenue retention went from 68% to 91%
- All this without changing pricing or acquiring differently
The lesson: retention improvements are often achievable through operational changes, not product changes. Founders assume they need to build more features to keep customers. Usually they need to fix the flow around the features they already have.
What to do this week
If you haven't measured retention:
- Calculate your 30, 60, and 90-day retention rates today
- Calculate your net revenue retention
- This is baseline. Everything else is measured against it.
If retention is below benchmark:
- Start with Tactic 2 (talk to 5 churned customers this week)
- Then Tactic 1 (fix activation for new users)
- These two, done for 90 days, typically move the needle more than everything else combined
If retention is healthy but not great:
- Add Tactics 3 (resurrection flow), 5 (segmentation), and 9 (value communication)
- These are the leverage-multipliers on top of a solid foundation
If retention is world-class:
- Focus on Tactic 6 (expansion) — you're already keeping people, now grow revenue per customer
- Consider Tactic 8 (switching costs) — protect the moat
If you're not sure where to start, related reads:
- The 5 metrics your startup needs — retention is #1 among them
- The SaaS onboarding fixes that move activation — the foundation
- How to price your SaaS — pricing affects who joins, which affects retention
Retention is the compounding metric of business. Improve it 15% and you don't just get 15% more revenue — you get 15% more sustainable-growth over years. It's the highest-leverage work most founders under-invest in.
---
If your retention is weaker than you'd like and you're not sure why, reach out via the contact page with your numbers and a description of your product. I'll help you identify the specific gap.