Customer Retention Rate (CRR) Calculator
From starting customers, ending customers and new acquisitions, compute the true retention of existing customers (CRR).
輸入資料
計算結果
重點速覽：CRR = (ending customers - new customers) / starting customers x 100%. Subtracting new customers is essential — otherwise new wins are mistaken for retained old customers and CRR is overstated (even absurdly >100%). Example: start 1,000, new 200, end 1,100 → retained 900, CRR 90% (100 churned). CRR + churn = 100%. WARNING: Compare with peers/history; varies by industry and period. Education, not advice.
計算公式
留存舊客 = 期末客戶 − 期間新增客戶。
客戶留存率 = 留存舊客 ÷ 期初客戶 × 100%。
$$\text{Retained} = \text{End Customers} - \text{Acquired Customers}$$$$\text{CRR} = \dfrac{\text{Retained}}{\text{Start Customers}} \times 100\%$$使用說明
- Enter starting customer count.
- Enter ending count and new acquisitions.
- View retained customers and CRR.
客戶留存率 (CRR) 計算範例
| 情境 | 期初客戶 | 期末客戶 | 期間新增 | 留存舊客 | 留存率 |
|---|---|---|---|---|---|
| 基準 | 1,000 位 | 1,100 位 | 200 位 | 900 位 | 90.00% |
| 流失偏高 | 1,000 位 | 1,050 位 | 250 位 | 800 位 | 80.00% |
| 零流失 (全留) | 1,000 位 | 1,200 位 | 200 位 | 1,000 位 | 100.00% |
留存舊客 = 期末客戶 − 期間新增；留存率 = 留存舊客 ÷ 期初客戶 × 100%。務必扣除新增客戶，否則會把新客誤算成留下的舊客。
理財情境案例
案例一：正確計算客戶留存率
某訂閱服務期初有 1,000 位客戶，期間新招來 200 位，期末帳上共 1,100 位。
第一步：留存舊客 = 期末客戶 − 期間新增 = 1,100 − 200 = 900 位。第二步：留存率 = 留存舊客 ÷ 期初客戶 × 100% = 900 ÷ 1,000 × 100% = 90%。
關鍵在於『扣除新增』:期末雖有 1,100 位看似成長，但其中 200 位是新客;若直接用 1,100 ÷ 1,000 = 110% 就荒謬地超過 100%，因為把新客誤算成留下的舊客。真正的留存率是 90%,代表期初的 1,000 位中流失了 100 位。
案例二：留存率、流失率與獲客成本的關係
承上，留存率 90% 意味著同期流失率為 100% − 90% = 10% (留存率 + 流失率 = 100%)。若期初 1,000 位中流失擴大到 200 位 (期末 1,050、新增 250、留存 800)，留存率降至 80%,流失率升到 20%,是生意健康度惡化的訊號。
為何留存如此關鍵？因為留住舊客的成本通常遠低於獲客成本 (CAC)。若拉一個新客要花 200 元 CAC，而留住一個舊客只需少量關懷成本，那麼與其花大錢補回流失的客戶，不如先堵住流失。
實務判讀：(1) 留存率沒有放諸四海的標準，訂閱制/SaaS 追求很高的月/年留存，零售餐飲波動較大，應與自身歷史及同業比較看趨勢；(2) 高留存搭配高客戶終身價值 (LTV) 才是健康商業模式的基礎。本站另有『流失率計算器』與『獲客成本計算器』可搭配分析。
常見問題
Why subtract new customers?
Because the ending count mixes two groups — old customers kept and new ones won. Retention should measure only how many of the original stayed, so the numerator must be ending minus new; otherwise you'd count new customers as 'retained' and inflate the rate.
How is retention related to churn?
They sum to 100%. Retention tells 'how many kept', churn tells 'how many left' — opposite but complementary; analyse both, and use the churn-rate calculator for the同期 figure.
What is a good retention rate?
No universal standard — depends on industry and period. Subscription/SaaS pursue high monthly/annual retention; retail/F&B fluctuate more. Best practice: compare with your own history and peers, watch the trend — rising retention is a positive signal.
Why must we subtract new customers (again)?
It is the most error-prone step. The ending ledger mixes 'old customers still there' with 'newly acquired'. CRR wants pure retention of the original base. Directly dividing ending by starting can exceed 100% (e.g. 1,100/1,000 = 110%, logically impossible). Correct: retained = 1,100 - 200 = 900; CRR = 900/1,000 = 90%. This honestly reflects 'of 1,000 originals, 900 stayed, 100 left'. Subtracting new customers strips their contribution so retention isn't masked by acquisition — a company can look like it's growing while quietly leaking old customers; correct CRR exposes that.
How do retention, churn and CAC interplay, and why does retention matter?
They depict the health of your customer economy. Retention + churn = 100% always. Retaining an existing customer usually costs far less than winning a new one (CAC) — keeping customers extends the value of each acquisition (higher LTV). Healthy models satisfy 'LTV >> CAC' (often LTV/CAC >= 3), and high retention is the engine behind LTV. So rather than pouring all resources into acquisition, plugging churn first is often higher-ROI. Pair with the churn-rate, CAC and LTV calculators.
相關工具
參考資料
內容審核：香港計算器財經團隊。計算邏輯與公式參考香港金融管理局（HKMA）及投資者及理財教育委員會（IFEC）之個人理財計算指引，結果僅供參考，實際以相關機構公佈為準。