.png)
Frameworks, core principles and top case studies for SaaS pricing, learnt and refined over 28+ years of SaaS-monetization experience.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
What Is Annual Recurring Revenue ARR a Complete Guide for SaaS Executives
ARR has become the shorthand for SaaS health. Boards ask for it before they ask about revenue. Sales leaders use it to size pipelines. Investors use it to judge growth quality. Yet many companies still treat ARR as a simple annualized revenue number, then make pricing, discounting, and product decisions that inflate the metric while weakening the business beneath it.
The central question is not whether ARR matters. It does. The question is what ARR should govern. Monetizely's position is clear: ARR should be the executive scorecard for recurring-contract growth, retention, and expansion. It should not be the pricing metric that determines what customers pay for. A company that confuses those two jobs will often create predictable-looking ARR and fragile economics.
Annual Recurring Revenue is the annualized value of recurring customer contracts that are active at a specific date. It measures the subscription run rate embedded in the installed base.
For a contract with a fixed recurring value, the basic calculation is:
[ \text{ARR} = \sum \text{Annualized recurring contract value for active customers} ]
Where contract terms vary, a more precise calculation annualizes each active recurring contract:
[ \text{ARR} = \sum{i=1}^{n}\left(\frac{\text{Recurring contract value}i}{\text{Contract days}_i}\times365\right) ]
Varonis uses that daily annualization method for active SaaS, term-license, and maintenance contracts. SailPoint uses the same approach for active subscription agreements. Both companies also state plainly that ARR is an operating metric, not GAAP revenue and not a forecast of future revenue.
The distinction matters most at the point of sale. A customer that signs a $120,000 annual subscription on December 15 contributes $120,000 of ARR immediately if the contract is active. The company will recognize only a small portion of that amount as revenue in December under a ratable subscription arrangement.
The following exhibit shows a disciplined ARR calculation for a SaaS company at quarter end.
| Contract item | Recurring contract value | Term | ARR treatment | ARR contribution |
|---|---|---|---|---|
| Enterprise platform subscription | $120,000 | 12 months | Include full annual recurring fee | $120,000 |
| Two-year workflow subscription | $240,000 | 24 months | Annualize contract value | $120,000 |
| Annual committed event bundle | $30,000 | 12 months | Include contracted recurring minimum | $30,000 |
| One-time implementation project | $25,000 | One time | Exclude | $0 |
| Usage overages above commitment | $18,000 | Variable | Exclude until contractually committed | $0 |
| Total ARR | $270,000 |
The table shows the governing principle: ARR counts recurring contractual value, not every dollar a customer may spend. That boundary gives the metric integrity.
A company can choose to include recurring support, committed usage minimums, or term subscriptions if those items are contractually recurring and renewed as part of the offer. It should exclude implementation, training, custom work, one-time data migration, and uncommitted overages. The policy must be explicit and stable.
ARR becomes misleading when leaders use it as a substitute for every other financial measure. It is not. Each measure gives management a different view of the business.
Varonis reported $745.4 million in ARR as of December 31, 2025, while also reporting $1.10 billion in remaining performance obligations. Those figures describe different things: active recurring contract run rate versus contracted revenue that has not yet been recognized.
Confusing ARR with revenue produces a common planning failure. The CFO forecasts a healthy ARR ramp. The operating plan assumes that the same growth will appear in revenue and cash. Then multi-year deal timing, implementation delays, or a shift from upfront licenses to ratable SaaS revenue changes the income statement without changing the underlying commercial momentum.
SailPoint provides a useful example. Its ARR reached $1.125 billion as of January 31, 2026, up from $876.7 million a year earlier. The company notes that ARR can grow faster than revenue when more incremental business comes from SaaS contracts with ratable revenue recognition.
A written ARR policy protects the number from commercial pressure
ARR is not a universal accounting standard. It is a management definition. That fact makes governance more important, not less important.
An ARR policy should settle five questions before the sales organization is measured on the number:
Amplitude's December 31, 2025 disclosure demonstrates why the policy matters. It defined ARR as the annual recurring revenue of subscription agreements in force at a point in time, including certain recurring premium services, while stating that ARR does not represent annualized GAAP revenue or a revenue forecast.
The strongest rule is simple: every ARR inclusion should survive a skeptical question from finance, customer success, and the customer who signed the contract. If the answer depends on a sales exception, the item does not belong in ARR.
Public-company ranges are useful reference points, not universal targets
Executives often ask for the “right” ARR growth rate or net retention benchmark. Public disclosures can establish a credible reference range, but no management team should use a generic range as a pricing target.
The relevant comparison set is companies with a similar customer profile, sales motion, product maturity, contract structure, and cost base. A $20 million self-serve analytics vendor should not manage against the same expansion target as a $1 billion identity-security company with a large enterprise installed base.
The table below summarizes selected public B2B SaaS disclosures. The range is useful because it anchors the conversation in reported results rather than vendor surveys.
| Metric | Reported company examples | Observed range | Management read |
|---|---|---|---|
| Year-over-year ARR growth | Varonis: 16.0% as of December 31, 2025; Amplitude: 17.0% as of December 31, 2025; SailPoint: 28.3% as of January 31, 2026 | 16.0% to 28.3% | Growth above this set usually requires strong new-logo execution, expansion, or both. |
| Dollar-based net retention | Amplitude: 104% as of December 31, 2025; monday.com: 110% as of December 31, 2025; SailPoint: 113% as of January 31, 2026 | 104% to 113% | Retention above 100% means expansion and price realization offset contraction and churn within the existing base. |
| Large-account ARR concentration | Amplitude: customers above $100,000 ARR accounted for 78% of ARR as of December 31, 2025; monday.com: customers above $50,000 ARR represented 41% of ARR as of December 31, 2025 | Meaningfully different by model | Account concentration changes the risk profile, sales capacity need, and renewal exposure of the ARR base. |
Amplitude reported $366 million of ARR and 104% trailing-twelve-month dollar-based net retention at December 31, 2025. Varonis reported $745.4 million of ARR at the same date. SailPoint reported $1.125 billion of ARR and 113% dollar-based net retention as of January 31, 2026.
monday.com offers a further reminder that ARR quality sits inside the customer mix. As of December 31, 2025, its customers above $50,000 in ARR represented 41% of total ARR, and the company reported 110% net dollar retention.
The implication is not that every company should target 110% retention or 25% ARR growth. The implication is that ARR growth needs an explanation by source: new customers, product expansion, price increases, contract term changes, or acquired revenue.
An ARR bridge reveals whether growth is durable
A single ending ARR number hides the forces that created it. The executive team needs an ARR bridge every month and every quarter.
Consider a company that begins the year with $10 million of ARR.
The company posts 30% ARR growth. Its gross retention is 87%, while net retention is 106% once expansion and contracted price increases are included. The business is growing, but the bridge reveals that new-logo acquisition and account expansion are carrying a meaningful churn burden.
That distinction should change management action. A company with 30% ARR growth and 87% gross retention does not have the same operating problem as a company with 30% growth and 96% gross retention. One needs a retention intervention. The other may need more efficient demand generation or capacity to serve expansion.
The pricing metric determines ARR quality before the forecast does
The temptation to make ARR the pricing objective is understandable. Annual subscriptions are predictable. Seats are easy to quote. A large upfront annual commitment makes the sales dashboard look better.
Yet the easiest model to annualize is not always the best model to sell.
Monetizely's 5-Step Pricing Framework starts with Goals and Segmentation, then moves to Packaging - Designing Offers That Fit, Choosing the Right Pricing Metric, Finding the Right Price Points, and Operationalizing Agentic AI Pricing. The order is essential. Goals determine what growth the company needs; segments determine which buyers have distinct needs; packages create offers for those buyers; the pricing metric decides what the customer is charged for; price points set the rate; and operations make the model work in billing, quoting, product telemetry, and renewals. The full sequence is developed more fully in Monetizing Agentic AI - see Footnote 1.
ARR belongs downstream of the pricing-metric decision. It should measure the recurring value produced by a sound model. It should not decide whether the company charges by seat, location, workflow, active record, transaction, committed usage, or completed outcome.
A strong pricing metric passes four tests:
- Buyers can connect it to the value they receive.
- The company can measure and invoice it accurately.
- The metric protects gross margin as usage scales.
- Sales teams can explain it without creating procurement friction.
A seat model may pass all four tests for collaboration software, where each employee receives ongoing access and the buyer budgets by headcount. A transaction model may fit payments, document processing, or communications platforms where value rises with completed activity. A fixed platform fee may fit infrastructure that must be available even when usage is uneven.
The wrong response is to choose seats because they create clean ARR. That decision turns ARR into the tail that wags the business.
AI and agentic products sharpen the problem because cost and customer value can move at different speeds. A seat-based AI assistant may have heavy users whose inference costs exceed the revenue from their licenses. An autonomous agent may produce value measured in resolved cases, qualified meetings, or completed tasks, while a seat price ignores that output.
The Agentic Monetization Spectrum, or AMS, clarifies the choice. It scores an agent on three dimensions: zero-human ability, or how much human work remains; operational domain, or whether the agent handles a task, a workflow, or work across functions; and output/cost ratio, or how sharply customer value rises relative to compute cost. As autonomy, domain breadth, and output value increase, the strongest pricing metric moves away from a human seat and toward a measurable output or outcome.
The following scorecard shows why ARR can remain stable while the underlying billing metric changes.
| Product archetype | Zero-human ability | Operational domain | Output/cost ratio | AMS score | Primary pricing metric |
|---|---|---|---|---|---|
| AI coding assistant that drafts code while an engineer reviews and ships it | Medium - 2 | Medium - 2 | Inflecting - 2 | 6 / 9 | Developer seat |
| AI customer-service agent that resolves requests across channels and systems | Large - 3 | Large - 3 | Exponential - 3 | 9 / 9 | Resolved customer interaction |
A coding assistant can create ARR through annual developer-seat commitments because the developer remains the buyer's practical unit of work. Cursor is a useful example of packaging that separates individual, team, and enterprise needs primarily through administrative and governance features while keeping the core coding value consistent.
An autonomous customer-service agent requires a different architecture. The primary meter should be completed resolutions, with an annual minimum commitment that creates budget certainty and ARR. The annual commitment is the contract structure. The resolved interaction is the pricing metric. Treating those as the same thing is the source of many AI pricing mistakes.
ARR becomes useful when it changes decisions. The best operating reviews do not stop at ending ARR, new ARR, and churn. They connect the bridge to product use, packaging, discounting, gross margin, and renewal risk.
Each monthly review should ask whether expansion came from more users, more workflows, more usage, higher prices, or a broader package. A rise in ARR from discounts that pulled forward multi-year signatures deserves a different response than a rise driven by customers adding a second product.
Sales compensation also needs care. Paying only on new ARR can reward deals with excessive discounting, weak implementation readiness, or commitments that customers cannot sustain. Paying only on net retention can discourage the sales organization from opening new markets. A balanced plan needs both growth and durability.
The executive question is therefore not, “How do we maximize ARR?” It is, “What kind of ARR are we building, and does the pricing model make that growth repeatable?”
Actions that put ARR in its proper place
Create a board-level ARR bridge by product, segment, and cohort. Require every material movement to be tagged as new logo, expansion, price increase, contraction, churn, acquisition, or currency movement.
Set separate targets for ARR, revenue, cash collections, gross retention, net retention, and gross margin. One target cannot carry the operating plan for a subscription business.
Make pricing-metric approval a product and finance decision, not a sales-forecast decision. The proposed meter should show customer value, cost behavior, billing feasibility, and expected renewal behavior before launch.
Track ARR quality alongside ARR quantity. Measure renewal concentration, discount levels, implementation completion, product adoption, and margin by cohort.
Build annual commitments around the customer’s budgeting cycle, then select the primary meter around delivered value. For AI agents, a committed annual minimum can support ARR while the billed unit remains a completed outcome.
Footnotes
- Monetizing Agentic AI. https://www.amazon.com/Monetizing-Agentic-AI-Handbook-Transformation/dp/B0H7Z13VKJ/
- Amplitude, Form 10-K for the fiscal year ended December 31, 2025. (sec.gov)
- Varonis Systems, Form 10-K for the fiscal year ended December 31, 2025. (sec.gov)
- SailPoint, Form 10-K for the fiscal year ended January 31, 2026. (sec.gov)
- monday.com, fourth-quarter and fiscal-year 2025 results, filed February 9, 2026. (sec.gov)