Cohort Parameters

Addis Business Presets
250 users
1,500 ETB
32%
Typical friction from Telegram screenshot chasing, missed CBE slips, and manual SMS links.
6%
With automated pre-due SMS, direct USSD push, and webhook authorization.

Retention & Revenue Projection

Comparing Manual Friction vs Automated Recovery
Month 1 Revenue
375,000 ETB
Cohort signup baseline
Month 2 Retained (Manual)
170
80 subscribers lost to friction
Annual Churn Loss
1,440,000 ETB
Total revenue forfeited vs automated
12-MONTH COHORT REVENUE TRAJECTORY (ETB)
Automated Flow Manual Chasing
Optimal Addis Path Automated payment intent webhook + pre-due SMS reminder

Eliminate Day 30 churn by dispatching a localized Telebirr USSD deep-link 3 days prior, validating via Chapa recurring webhook, and triggering Telegram bot fallback notifications.

Timeline Manual Subs Manual Revenue Automated Subs Automated Revenue Preserved MRR

Addis Operational Playbook: Eliminating Month 2 Renewal Friction

1

Day -3: Pre-due SMS with Telebirr/CBE Quick Pay Deep Link

Send an automated SMS alert before the subscriber's billing cycle closes. Include an authenticated direct checkout URL rather than raw CBE account numbers.

2

Day 0: Chapa Webhook & Intent Verification

Listen for payment verification webhooks in real time. Automatically renew SaaS access and issue Ethiopian tax e-invoices without asking team members to check Telegram screenshots.

3

Day +2: Automated Telegram Bot Grace Period Prompt

If payment fails or slips, trigger a bot notification in their Telegram or WhatsApp chat offering a 48-hour grace window with one-tap payment retry.

Assumed renewal drops and rounded cohort accounting

Read the explanation

The saved calculator starts with a fixed subscriber cohort and two assumed renewal drop rates. At the default two hundred fifty subscribers and fifteen hundred Birr monthly fee, month one revenue is three hundred seventy five thousand Birr. Manual renewal uses a thirty two percent assumed drop, leaving one hundred seventy subscribers in month two. The automated scenario uses six percent, leaving two hundred thirty five. These bars use one point six pixels per subscriber: two hundred seventy two and three hundred seventy six pixels. Their sixty five subscriber difference times fifteen hundred gives ninety seven thousand five hundred Birr month two revenue difference. These are hypothetical scenario inputs, not measured Ethiopian churn or proof that automation produces the assumed improvement. From month three onward, the calculator multiplies the previous manual count by point nine six and automated count by point nine eight five, rounding to an integer at every month. At default month two values, month three manual count is round of one hundred sixty three point two, giving one hundred sixty three. Automated count is round of two hundred thirty one point four seven five, giving two hundred thirty one. Bars share one point six pixels per subscriber, widths two hundred sixty point eight and three hundred sixty nine point six. Revenue multiplies each rounded count by the fee. The twelve month totals sum those monthly revenues. This model has no new subscriber acquisition and no explicit fees, taxes, payment failures or changing renewal rates beyond the authored rules. The annual churn loss label is a separate shortcut, not the difference between the two twelve month cohort totals. It takes the original subscribers minus manual month two retained, multiplies by monthly fee, then by twelve. At default, eighty lost subscribers times fifteen hundred times twelve gives one million four hundred forty thousand Birr. The bars compare that shortcut using twelve months with a hypothetical eleven month multiplier: one million three hundred twenty thousand. At one display pixel per four thousand Birr, widths are three hundred sixty and three hundred thirty. The code uses twelve even though the drop happens in month two. Gateway and recovery text describe plans; this browser calculator does not execute payment webhooks or send messages. Export behavior and chart availability require independent checks; these outputs are assumptions rather than financial forecasts.

Model exported successfully.
Enjoy this tool? Build your own with Super