ARR and MRR (Annual Recurring Revenue and Monthly Recurring Revenue)

Definition: MRR (Monthly Recurring Revenue) is the predictable subscription revenue a business collects every month, normalized to a monthly figure regardless of how each customer actually pays. ARR (Annual Recurring Revenue) is that same figure annualized — MRR multiplied by 12. Together they are the two headline metrics subscription and SaaS companies use instead of raw revenue, because they isolate the recurring, predictable part of the business from one-time, lumpy, or non-repeating income.

How It Works

What Counts as Recurring Revenue

Only revenue from active subscriptions counts toward MRR — a customer paying $99 a month for a live subscription contributes $99, full stop
Annual contracts are normalized to a monthly figure: a customer paying $1,200 upfront for a one-year plan contributes $100 to MRR, not $1,200 in the month it was collected
One-time revenue — setup fees, professional services, custom onboarding work, a single hardware purchase bundled with a subscription — is excluded from MRR entirely, even though it shows up on an income statement
Usage-based or metered revenue is trickier: many companies include a trailing average of recent usage charges, but because it isn’t guaranteed to repeat at the same level, some companies exclude it or report it separately from “committed” MRR
Discounts and promotional pricing reduce MRR to the amount actually being collected, not the list price a customer would otherwise pay
Multi-year contracts follow the same normalization logic as annual ones: a two-year contract worth $24,000 in total still only contributes $1,000 to MRR

Committed vs. Self-Serve MRR

Committed MRR comes from customers locked into a contract for a minimum term, even if they’re billed monthly, and is treated as the highest-quality, most predictable slice of MRR
Self-serve, month-to-month MRR comes from customers who can cancel at any time with no contractual commitment — valuable revenue, but inherently less certain to still be there next month
A company blending both types into a single MRR figure without distinguishing them can look just as “recurring” as a peer with a much higher share of committed, contracted revenue, even though the two businesses carry very different risk
Investors evaluating ARR during due diligence often ask what share of it is contractually committed versus freely cancellable, since that split materially affects how confidently the figure can be projected forward

Multi-Currency and Multi-Region MRR

Companies selling in multiple currencies typically convert everything to a single reporting currency at a consistent exchange rate, so the MRR figure stays comparable from one month to the next
Using a live, floating exchange rate for MRR reporting can make the metric swing with currency markets rather than actual business performance, which is why many companies fix a rate and only revisit it periodically
This matters more as a company expands internationally, since currency-driven swings can otherwise be mistaken for — or used to obscure — real changes in the underlying MRR waterfall

The MRR Waterfall

Because MRR moves in more than one direction every month, most companies track it as a “waterfall” that breaks total MRR movement into its components:

New MRR — recurring revenue from brand-new customers signed in the period
Expansion MRR — additional recurring revenue from existing customers upgrading, adding seats, or buying add-ons
Contraction MRR — recurring revenue lost from existing customers downgrading or reducing usage, without fully canceling
Churned MRR — recurring revenue lost from customers who cancel outright, directly driven by Churn Rate
Reactivation MRR — recurring revenue recovered from customers who previously churned and later came back
Some companies further split New MRR into self-serve versus sales-assisted signups, since the two acquisition motions often carry very different CAC and LTV (Customer Acquisition Cost and Lifetime Value) profiles even though both count identically toward the waterfall

Net new MRR for the period is:

\text{Net New MRR} = \text{New MRR} + \text{Expansion MRR} - \text{Contraction MRR} - \text{Churned MRR} + \text{Reactivation MRR}

Looking at the waterfall instead of the single net number reveals why MRR moved, not just that it moved — a flat month could mean nothing happened, or it could mean strong new sales exactly offset a serious churn problem
Expansion and contraction are what make net revenue retention diverge from gross retention, since expansion is the piece that can push net retention above 100% even with some churn present
Reactivation is usually the smallest waterfall component, but it’s worth tracking separately, since it signals whether “lost” customers are recoverable with the right win-back effort
Most finance teams maintain the waterfall as a running monthly ledger rather than recomputing it from scratch each period, since each month’s ending MRR is simply the prior month’s ending MRR plus that month’s net new MRR

From MRR to ARR

ARR = MRR \times 12

ARR is simply MRR scaled to a yearly view — it doesn’t imply the company has annual contracts, or that the number is somehow more “locked in” than MRR
ARR is used whenever a figure needs to feel like an annual-scale business outcome: fundraising conversations, board decks, and public statements about company size, since “$6M ARR” communicates scale more intuitively than “$500K MRR”
MRR is used for the month-to-month operating rhythm: sales targets, marketing spend pacing, and how finance and leadership actually run the business week to week, since a full year is too slow a feedback loop to catch problems early
Because ARR is just MRR × 12, a change in MRR this month instantly changes the ARR figure too — ARR is a snapshot multiplied out, not a separately measured number
Some companies wait to report ARR until MRR has stabilized into a real recurring pattern, since annualizing a single unusually strong (or weak) month can produce a misleading ARR figure
A related figure, ARR run rate, is sometimes calculated from a shorter recent window (like the last full quarter, annualized) rather than the current single month, smoothing out one-off spikes while still reflecting recent momentum better than a trailing twelve-month actual figure would

MRR vs. ARR at a Glance

	MRR	ARR
Time scale	Monthly	Annual (MRR × 12)
Primary audience	Internal operators, finance, sales leadership	Investors, board, public-facing scale statements
Best for	Spotting problems quickly, month-to-month pacing	Fundraising, valuation multiples, “size of business” framing
Volatility	More sensitive to single-month swings	Smooths short-term noise into a yearly-scale number
Typical review cadence	Weekly or monthly	Quarterly or at fundraising/board milestones

MRR Growth Rate

The health of a subscription business is judged less by the absolute MRR number and more by how fast it’s growing:

\text{MRR Growth Rate} = \frac{MRR_{\text{this month}} - MRR_{\text{last month}}}{MRR_{\text{last month}}}

Early-stage startups are often expected to sustain double-digit month-over-month MRR growth to be considered on a strong trajectory, though this naturally decelerates as the revenue base gets larger
A slowing MRR growth rate, even with MRR still rising in absolute terms, is one of the first things experienced investors flag, since it signals the growth engine is losing efficiency
Net revenue retention — expansion and contraction and churn all netted against starting MRR, covered in depth under Churn Rate — determines how much MRR growth a company gets “for free” from its existing base before a single new customer is signed
Comparing gross new MRR added to net new MRR added shows how much of the growth engine’s output is being absorbed by churn and contraction before it ever reaches the bottom line
A widely referenced heuristic called the “Rule of 40” suggests a healthy SaaS company’s annualized MRR growth rate plus its profit margin should add up to roughly 40% or more — capturing the idea that it’s fine to grow fast while unprofitable, or grow more slowly while solidly profitable, but being weak on both fronts at once is a warning sign

The MRR Quick Ratio

A related metric puts the waterfall’s growth and loss components into a single efficiency score:

\text{MRR Quick Ratio} = \frac{\text{New MRR} + \text{Expansion MRR}}{\text{Contraction MRR} + \text{Churned MRR}}

A Quick Ratio of 4 means a company is adding four dollars of new and expansion MRR for every dollar lost to contraction and churn — a commonly cited healthy benchmark for a growing SaaS business
A Quick Ratio near or below 1 means the business is roughly treading water or shrinking, regardless of how strong new sales look in isolation, since losses are eating up an equivalent or greater share of gains
Unlike the net new MRR figure, the Quick Ratio is unaffected by the size of the revenue base, which makes it easier to compare growth efficiency across companies of very different sizes
Tracking the Quick Ratio over several months reveals whether growth is becoming more or less efficient, independent of whether absolute MRR is still increasing
Some teams calculate two versions side by side — one including new-customer MRR only, and one including expansion only — to see separately whether new business or existing-account growth is carrying more of the load

ARR and MRR Milestones by Stage

Rough, commonly cited stage benchmarks give founders and investors a shared sense of whether ARR is tracking normally for a company’s age and funding stage:

Stage	Illustrative ARR Range	Typical Focus
Pre-seed	Often pre-revenue to a few hundred thousand	Finding early signal, not yet optimizing MRR growth rate
Seed	Roughly $0–$1M	Proving early Product-Market Fit and a repeatable MRR waterfall
Series A	Roughly $1M–$3M, growing quickly	Demonstrating a healthy MRR growth rate and improving net revenue retention
Series B and beyond	Roughly $3M–$10M or more	Scaling go-to-market while defending margins and retention

These ranges are rough industry heuristics, not fixed rules — they vary enormously by vertical, price point, and sales motion, and a company well outside them isn’t automatically off track. See Seed Round vs Series A for how these milestones interact with fundraising expectations.

Why It Matters

MRR and ARR isolate the predictable, recurring core of a business from one-time revenue, giving a much cleaner signal of actual momentum than total revenue alone
Investors size and value subscription businesses largely off ARR, often applying a revenue multiple directly to it, which is why the composition and growth rate behind ARR matters as much as its headline size
MRR is granular enough to catch problems within weeks, while waiting for annual or even quarterly revenue figures would let a growth or retention problem compound for months before anyone notices
The MRR waterfall exposes whether growth is coming from healthy new business and expansion, or is being propped up while churn quietly erodes the base underneath it
MRR and ARR directly shape how long a company’s cash lasts relative to its spending, since predictable recurring revenue can be weighed against Runway and Burn Rate with far more confidence than lumpy, one-time revenue
A rising ARR figure alongside deteriorating net revenue retention is a warning sign that gets missed if a team only ever looks at the top-line number
Because ARR scales cleanly from MRR, it gives founders, employees, and investors a simple, shared shorthand for company size and progress without needing to walk through a full income statement
Clean, consistent MRR and ARR definitions build trust with investors during due diligence, since inflated or inconsistently defined recurring revenue is one of the fastest ways to damage credibility in a fundraise

Common Pitfalls

Counting annual contracts at full value in the month collected: a $12,000 annual deal is $1,000 of MRR, not a one-month spike followed by eleven months of nothing — this single mistake is the most common way founders accidentally inflate MRR
Including one-time fees in MRR: setup fees, onboarding charges, and professional services revenue aren’t recurring, and folding them in inflates the metric’s meaning even if the revenue itself is real
Ignoring the MRR waterfall and only reporting the net number: a flat net-new MRR figure can hide a serious churn problem being masked by strong new sales, or vice versa
Multiplying a single strong (or weak) month by 12 to get ARR: annualizing one unusually good or bad month produces a distorted ARR figure that doesn’t reflect the actual run rate
Blending usage-based revenue into MRR as if it were fully committed: metered revenue that could drop next month isn’t the same quality of “recurring” as a fixed subscription fee, and treating them identically overstates predictability
Not adjusting MRR for discounts and promotions: counting list price instead of what customers actually pay overstates both MRR and the ARR built on top of it
Comparing ARR across companies with different definitions: one company’s ARR might exclude usage revenue and discounts while another’s doesn’t, making a head-to-head comparison misleading without checking the underlying methodology first

ARR and MRR in Fundraising and Board Reporting

Board decks typically lead with ARR because it communicates company scale in a single, annual-feeling number that’s easy to compare quarter over quarter
Term sheets and valuation conversations frequently reference an ARR multiple (for example, “8x ARR”), tying the accuracy and composition of that ARR figure directly to company valuation
Fundraising due diligence routinely includes a full MRR waterfall review going back a year or more, since investors want to see the components of growth, not just the net trend line
Internally, operating reviews lean on MRR and its components because monthly granularity surfaces problems — a bad sales month, a spike in churn — while there’s still time to react within the same quarter
Investors pay close attention to whether ARR growth is coming from durable expansion and new-logo growth versus one-off enterprise deals that may not repeat, since the latter overstates the health implied by the ARR figure alone
Some founders present ARR alongside a bookings figure (the total contract value signed, regardless of when it’s recognized as recurring revenue) — conflating the two is a common source of confusion, since bookings can substantially outpace the ARR actually “live” and generating monthly revenue so far

Example

A project management SaaS startup ends July with 1,000 customers on a single $50-a-month plan and no other pricing tiers, giving it a clean MRR of $50,000 and an ARR of $600,000. During August, the waterfall looks like this: 80 new customers sign up on the same $50-a-month plan, contributing $4,000 of New MRR; 30 existing customers upgrade to a $90-a-month power-user tier, adding $1,200 of Expansion MRR; 10 customers downgrade to a cheaper $30-a-month plan, subtracting $200 of Contraction MRR; and 25 customers cancel outright, removing $1,250 of Churned MRR. No previously churned customers reactivate this month. Net new MRR for August comes to $4,000 + $1,200 - $200 - $1,250 = $3,750, bringing MRR to $53,750 and ARR to $645,000 — an MRR growth rate of $(53{,}750 - 50{,}000) / 50{,}000 \approx 7.5\%$ for the month. The same month’s Quick Ratio comes to $(4{,}000 + 1{,}200) / (200 + 1{,}250) \approx 3.6$ — solid, though a shade under the commonly cited benchmark of 4.

Looking only at the ending MRR number, August looks like a solid month. But the waterfall tells a sharper story: Churned MRR of $1,250 nearly equaled New MRR, and without the $1,200 of Expansion MRR from existing power users upgrading, growth would have been far weaker. The founders realize their new-customer funnel, not their retention, is actually the weakest link — expansion and a still-healthy pace of new signups are doing most of the work of covering for a churn rate that needs attention.

Six months later, preparing a fundraising deck, the company reports “$780,000 ARR, growing 6% month-over-month” to prospective investors. During due diligence, an investor asks for the underlying MRR waterfall rather than accepting the headline number, and finds that a chunk of the ARR growth over that period came from three large annual contracts signed in a single month — deals that inflated that month’s apparent New MRR before growth settled back into a more modest, sustainable pace. The company’s ARR is real, but the investor’s request for the waterfall, rather than just the topline ARR figure, is exactly the kind of scrutiny that separates a durable growth story from one that merely looks impressive at a glance.

Referenced by

Dig deeper

Spot an error in this note?