Subscription Business Model: How It Works and Its Risks

In 2013, Adobe told investors that moving its creative software to subscriptions would shrink its revenue and squeeze its margins for a while. That bet on the subscription business model shows both the cost and the payoff of trading one-time sales for a steady stream of payments.
This guide covers the subscription business model for recurring revenue: how it works, which numbers matter, what the data shows and where it breaks. Customers pay a repeating fee for ongoing access instead of buying once. The company swaps a large upfront sale for smaller payments it must keep earning.
How the Subscription Business Model Works
In a traditional sale, the customer pays once, and the company books the revenue right away. In a subscription, the customer pays monthly or yearly, and the company usually recognises that revenue ratably, spread across the service period. Adobe’s filings describe this effect, saying the model makes revenue more recurring and predictable because it is recognised ratably.
Subscriptions take a few common forms. Access models, such as software or streaming services, charge for continued use. Replenishment models ship products on a schedule. Consumption-based models bill by usage, and Zuora’s billing data shows them still producing promising revenue growth. Many companies now blend several of these.
Recurring Revenue Metrics Worth Tracking
In a subscription business model, retention does more work than acquisition, because every lost subscriber has to be replaced before the base can grow. Four numbers carry most of the weight. Monthly recurring revenue (MRR) is the predictable subscription revenue expected each month, and annual recurring revenue (ARR) is that figure annualised. Adobe’s definition of ARR multiplied current paid subscriptions by the average monthly price and by twelve, then added the annual value of enterprise term agreements. Churn is the share of customers or revenue lost in a period. Customer lifetime value, set against the cost of acquiring a customer, shows whether growth pays for itself.
Consider a hypothetical. 1,000 subscribers paying $20 a month produce $20,000 in MRR, or $240,000 in ARR. If 3 per cent cancel each month, the business loses 30 subscribers monthly. Compounded, that is about 31 per cent of the base over a year unless it keeps adding customers. That is why churn deserves more attention than almost any other line. Adobe expected renewal rates to become a key metric for its subscription offerings for the same reason.
What the Subscription Economy Data Shows
Zuora tracks more than 600 recurring revenue businesses through its billing platform in its Subscription Economy Index. In its April 2025 report, companies in the index grew revenue 11 per cent faster than the S&P 500 over the previous two years. The 2024 edition put 2023 growth at 10.4 per cent against 6 per cent for the S&P 500. Earlier editions showed a wider gap. From 2012 through 2018, index companies grew about five times faster, at 18.1 per cent a year against 3.6 per cent.
Two cautions apply. The index draws on Zuora’s own customers, so it reflects subscription companies that use that billing system, not the whole market. Revenue growth is also not profit. The more useful finding for operators concerns revenue models. Companies using four or more models grew average revenue per account 4.5 per cent faster than companies with one, and 2.3 per cent faster than those with two or three.
What Adobe’s Switch Reveals About Transition Costs
Adobe’s shift to a subscription business model began with the launch of Creative Cloud in May 2012. A year later, the company said new creative features would go only to Creative Cloud subscribers and that Creative Suite 6 would be the last major update. CNET reported at the time that the main plan cost $50 a month.
The financial strain was real. Adobe said perpetual license revenue and total net revenue would fall in fiscal 2013, and that expenses were not falling in step, which hurt net income and operating margin. Its argument was that lower entry pricing would bring in new users and raise long-term growth. A senior marketing director described it as layering: customers who join stay, and new customers add revenue on top. The practical lesson is to plan for a dip in reported revenue and margin during any switch.
Where Recurring Revenue Breaks Down
Churn is the main threat, and not all of it is a customer’s choice. Some cancellations come from failed card payments rather than lost interest. Adyen, citing research from Zuora’s Subscribed Institute, reports that businesses accepting five or more currencies grew about 8 per cent faster than those accepting one. That is a correlation, not proof that more payment options cause growth.
Pricing creates a second risk. Adobe’s pitch rested on a lower entry price, which widens the customer base but makes each customer worth less per month. Retention then has to carry the economics. A third risk is cost. If a subscriber’s usage grows while the fee stays flat, margins shrink. Consumption-based billing, which ties price to usage, is one answer, and Zuora’s index shows it is still growing.
Is a Subscription Business Model Right for Your Product?
Ask four questions before switching:
- Do customers need the product repeatedly?
- Does it keep delivering value after the first month?
- Can you serve a subscriber for less than the fee?
- Can people cancel without friction, so those who stay do so by choice?
A full switch is not the only option. Adobe kept selling CS6 on a perpetual basis while customers migrated. A hybrid, such as a one-time product plus an optional support or content plan, lets you test demand for recurring payments before you commit.
Conclusion
The subscription business model rewards companies that keep earning each payment, not those that simply collect them. Adobe’s filings show the trade-off plainly: lower reported revenue and margins first, steadier income later. Before changing your own pricing, pull twelve months of cancellation data, calculate churn by customer segment, and model the revenue dip a switch would cause. If those numbers hold up, test the model on one product line before rolling it out.