Does Recurring Revenue Make a Business Predictable?

A subscription business begins the month with something a shop cannot take for granted: customers who have already agreed to keep paying.

That is a real advantage. The business does not have to win every sale from scratch. It has a starting base, renewal dates and some visibility into what customers will spend.

But the reassuring word is recurring. It can make a forecast sound more certain than the customer relationships behind it. A payment can recur until a customer cancels, cuts the number of seats, changes plans or decides a price increase is no longer worth accepting.

Recurring revenue helps make a business predictable. The degree of predictability depends on what keeps the revenue recurring.

Three numbers that run on different clocks

Adobe provides a useful current example. Its 10 September 2026 release covers fiscal Q3, ended 28 August. It reports:

Subscription revenue represented about 97.4% of total revenue, calculated from those figures. This is clearly a subscription-led business. Yet ARR is neither a cash balance nor a promise that the next year’s reported revenue will equal that amount.

Adobe’s September 10-Q measures ARR as annual subscription contract value across two customer groups, using fixed fiscal-year exchange rates and revalued comparatives. It includes Semrush’s contribution.

Those details matter when comparing growth. A changing currency basis or an acquired customer base is different from existing customers choosing to spend more. Dividing ARR by four and comparing it mechanically with quarterly revenue mixes a closing snapshot with activity across the quarter.

First, follow the customers who were already there

Imagine a service with 1,000 customers, each paying ₹10,000 a year. Its annual recurring run-rate starts at ₹100 lakh.

At renewal, 100 customers leave. The remaining 900 accept a 10% price increase, taking their annual price to ₹11,000. Those original customers now represent ₹99 lakh of run-rate: 90% customer retention, but 99% revenue retention.

Add 100 new customers at the new price and the total reaches ₹110 lakh. The business reports 10% growth, while its original cohort—the customers it started with—has become slightly smaller in revenue terms.

That is not necessarily a bad result. It is a different result from growth built on almost everyone staying and spending more. The difference affects how much new selling the business must do to keep growing.

FOLLOW THE SAME CUSTOMERS

Growth in the total. What about the cohort?

A fictional service starts with 1,000 customers paying ₹10,000 a year: ₹100 lakh of annual recurring run-rate. At one renewal date, some leave, prices change and new customers join.

Customers renewing 90%

50%100%

Annual price change +10.0%

−10%+30%

New customers 100

0400

Starting cohortGrowth hides churnPrice offsets departuresSteady renewals

Existing cohort’s annual run-rate

₹99.00lakh · 99.0% net revenue retention in this example

Total ending annual run-rate

₹110.00lakh · +10.0% from ₹100 lakh

900 original customers renew; 100 leave; 100 new customers join. Ending customer count: 1,000.

Starting run-rate₹100.00 lakh

Lost customers, at old price−₹10.00 lakh

Price change on retained customers+₹9.00 lakh

New customers, at new price+₹11.00 lakh

Ending run-rate₹110.00 lakh

Follow the cohort calculation

Retained cohort: 900 customers × ₹11,000 a year = ₹99.00 lakh. Divide by the starting ₹100 lakh to get 99.0% net revenue retention. New customers contribute ₹11.00 lakh separately. Everyone starts at the same price and retained customers keep the same seat count and plan.

Choose “Growth hides churn.” The ending run-rate still grows 10%, but only 80% of the original customers renew. Their revenue retention falls to 88%; new customers fill the gap.

Now choose “Price offsets departures.” A 25% price increase offsets 20% of customers leaving. The original cohort’s revenue retention is 100%, even though one in five customers has gone. A stable rupee figure does not establish a stable customer relationship.

Every number in this example is fictional. It is a renewal-date snapshot, not a calculation of Adobe’s retention or the revenue recognized during a year.

Retention has more than one meaning

Customer retention counts how many customers stay. It does not capture the size of their contracts: losing one large customer can matter more financially than losing several small ones.

Gross revenue retention follows the original customers’ revenue after losses and downgrades, excluding expansion. Net revenue retention also includes expansion within that same cohort, such as additional seats, upgrades or accepted price increases. New customers belong outside both calculations.

Definitions vary, so check the company’s calculation and measurement period. Net retention above 100% means the measured cohort’s expansion outweighs its losses. It does not mean nobody left.

Neither Adobe’s Q3 release nor its 10-Q discloses a customer-retention or net-revenue-retention percentage. Total growth cannot tell us how a fixed customer cohort behaved.

Smoother accounting can conceal a change in demand

Suppose a customer pays ₹12,000 upfront for a one-year service recognized evenly over that year. Cash arrives at the start; revenue is recognized at ₹1,000 a month as the service is provided. A monthly payer on otherwise similar terms produces a different cash schedule, even if revenue recognition is the same.

Revenue, billings and cash collection should not be treated as interchangeable. Adobe’s remaining performance obligations include deferred revenue and unbilled contracts awaiting recognition; they are a different measure from ARR.

Adobe primarily recognizes subscriptions over their term, with certain offerings measured by usage. Its filing warns that earlier agreements can delay the reported impact of weaker new subscriptions or renewals.

That is a useful distinction: smoother reported revenue is not automatically steadier underlying demand. Existing contracts can cushion a quarter while the next set of renewals becomes less certain.

What makes the recurrence credible?

The strongest explanation usually starts with the customer’s work. A product used every day, embedded in a team’s processes and worth more than it costs has a better reason to be renewed than a service someone bought once and barely uses.

Switching costs can help: migration, retraining or integrations take effort. But inconvenience is not unlimited pricing power. If the product stops solving the problem well, customers have a reason to bear that effort.

Contract length changes the timing of risk too. An annual agreement may improve near-term visibility while leaving a renewal decision ahead. A usage-based subscription can recur with variable spending. A customer might remain active while using fewer transactions or seats.

Price and retention should therefore be read together. A higher price may improve revenue immediately, but the next renewal is where acceptance becomes clearer. The interactive example holds those inputs independent to explain the arithmetic; real customers respond to prices.

Predictable revenue still needs attractive economics

Keeping customers requires product development, reliable service and support. Winning replacements requires sales and marketing. A growing subscriber base can bring rising delivery costs as well as revenue.

An upfront annual payment can help fund operations, but it comes with a service obligation. It is not all freely distributable profit on the day it arrives. To judge the business, we still need cash flow, margins, acquisition costs and the investment needed to keep the product relevant.

Before relying on a recurring-revenue forecast, I would ask:

1. What does the metric include? Read its scope, currency basis and treatment of acquisitions.
2. What happens to the same customers? Separate renewal losses, reduced spending, expansion and new sales where disclosed.
3. When can behaviour change? Check contract duration, renewal timing and usage exposure.
4. What does it cost to preserve the base? Follow servicing, product investment and customer acquisition alongside revenue.

Recurring revenue is valuable because the business starts with relationships rather than a blank sales book. Its predictability is earned through those relationships continuing to be useful to customers.

The subscription tells us how payment repeats. Understanding the business means asking why the customer will keep choosing it.