Customer retention rate is the percentage of existing customers a business keeps over a set period, calculated by subtracting new customers acquired during the period from the ending customer count, dividing by the starting customer count, and multiplying by 100. A company that begins a quarter with 200 customers, ends with 215, and picked up 30 new ones along the way retained 92.5% of its original base. The number strips out marketing wins and shows only whether the customers you already had chose to stay.
The Formula
Three inputs drive the calculation:
- S: customers at the start of the period
- E: customers at the end of the period
- N: new customers acquired during the period
The math is ((E − N) ÷ S) × 100. Subtracting new customers from the ending count is the step that makes this a retention measure rather than a growth measure. Without that subtraction, you are measuring overall customer change, which mixes acquisition performance into what should be a loyalty signal.
A Worked Example
A SaaS company begins the year with 100 customers. Over twelve months it signs 20 new accounts and loses 10 existing ones. The ending count sits at 110. Plug the numbers in: (110 − 20) ÷ 100 = 0.90, or a 90% retention rate. Ninety of the original hundred customers stayed.
The next year, starting from 110 customers, the company acquires 40 new accounts and loses only 5, ending at 145. Retention climbs to (145 − 40) ÷ 110 = 95.5%. Those five points translate directly into fewer accounts to replace and lower acquisition spending needed to hold the customer base steady.
Retention Rate and Churn Rate
Churn rate is the mirror image. If retention is 90%, churn is 10%. The two describe the same reality, but the framing shifts urgency. “Reduce churn from 10% to 7%” and “raise retention from 90% to 93%” set the same goal, and most subscription businesses track both, reporting retention externally while using churn internally to flag cancellation spikes.
Choosing a Measurement Period
The right interval depends on how often customers interact with the product. Subscription software companies usually measure monthly, because a bad month of churn compounds fast in a recurring revenue model. Retailers and businesses with longer purchase cycles often measure quarterly or annually, since shorter windows pick up seasonal buying patterns rather than genuine loyalty shifts.
Consistency matters more than the specific choice. Switching between monthly and quarterly measurement makes trend comparisons meaningless. Many public companies align retention tracking with their quarterly financial reporting cycle so the data feeds directly into earnings discussions.
What Counts as a Good Retention Rate
Benchmarks vary widely by industry because switching costs, contract lengths, and product stickiness are not the same everywhere. Annual averages typically land around:
- Media and professional services: 84%
- Insurance and automotive: 83%
- IT services: 81%
- Financial services and telecommunications: 78%
- Consumer services: 67%
- Retail: 63%
- Hospitality: 55%
A rate well below your industry’s average usually points to an internal issue rather than a market one. Context still matters. A startup in its second year will churn more than an established competitor sitting on entrenched enterprise contracts, and comparing across those situations misleads more than it informs.
When Headcount Isn’t Enough: Revenue Retention
The basic formula counts every customer equally, whether they pay $50 a month or $50,000. That blind spot is why many businesses also track revenue-based versions of the metric.
Gross revenue retention measures the percentage of recurring revenue kept from existing customers after cancellations and downgrades, ignoring any expansion from upsells. It answers how much baseline revenue is at risk. Net revenue retention (NRR) adds expansion revenue back in. An NRR above 100% means existing customers are spending more over time than the business is losing to churn. Top-performing SaaS companies maintain NRR of 111% or higher, growing revenue from the installed base without acquiring anyone new.
Tracking both headcount and revenue retention catches distortions that either would hide on its own. A company can show 95% customer retention while revenue retention sits at 80%, because the customers leaving happened to be large accounts. The reverse also happens: strong revenue retention can mask the loss of many small customers when a few enterprise expansions absorb the difference. Losing a large volume of small accounts often signals product-market fit problems in a specific segment, even when the top line looks fine.
Why the Number Matters Financially
Acquiring a new customer typically costs six to seven times more than keeping an existing one, so every percentage point of improved retention drops close to directly onto the bottom line. Research from Wharton has found that a 5% increase in retention can improve profitability by 25% or more.
The clearest expression of this is customer lifetime value. LTV is the total revenue a customer generates over the whole relationship, and higher retention lengthens that relationship without any additional acquisition spending. A widely cited benchmark for a healthy business is an LTV-to-customer-acquisition-cost ratio of at least 3:1, meaning each customer generates three dollars in lifetime value for every dollar spent to win them. Below 2:1 usually signals overspending on acquisition. Far above the benchmark, at 8:1 or 10:1, can paradoxically indicate underinvestment in growth. Retention rate is the lever that most directly moves where a business sits on that scale.
Cohort Analysis for Sharper Answers
A single retention number for the whole customer base is useful but blunt. Cohort analysis groups customers by when they started and tracks each group separately over time. The process runs in five steps:
- Define the cohort by a shared starting point, such as everyone who signed up in March or made a first purchase in Q2.
- Choose time intervals (daily, weekly, or monthly) based on how often customers use the product.
- For each interval, count how many members of the cohort are still active, counting each customer once per period.
- Divide active customers by the original cohort size and multiply by 100.
- Display cohorts as rows and time periods as columns in a table or heatmap, with retention percentages in each cell.
Patterns invisible in an aggregate number show up quickly this way. Customers from one marketing channel may churn at twice the rate of organic sign-ups, or a cohort onboarded after a product redesign may retain dramatically better than earlier ones. Those are targeted findings that lead to targeted fixes.
Data Pitfalls That Distort the Number
The formula is simple, but bad inputs produce bad outputs. The three data points need to come from one consistent source, whether a CRM, subscription billing platform, or point-of-sale system. Mixing sources creates double-counting problems, especially when customers hold multiple accounts or interact through different channels.
A few hygiene issues trip companies up repeatedly. Duplicate accounts inflate both starting and ending counts, artificially boosting retention. Free-trial users who never convert sometimes linger in the customer database and distort the numbers in the other direction. Reactivated customers who left and came back are a judgment call: retained or new? Most businesses treat them as new acquisitions to keep the metric honest. Whichever rule you adopt, apply it the same way every period, because the value of retention rate lives in the trend, not in any single reading.