The number everyone quotes

Scroll through maker posts for a few minutes and you will see MRR everywhere. "Crossed $1k MRR." "Back to $4k MRR after a rough month." It has become the default scoreboard for indie software, and for good reason: for a subscription business, it is the clearest single view of where things stand.

But MRR is also easy to calculate loosely, and even easier to misread. This post covers what it measures, what it leaves out, and how to interpret it, both for your own product and when you see it in the catalog, where numbers are shown as reported by makers.

The basic definition

Monthly recurring revenue is the subscription revenue you can expect to receive every month, if nothing changes. It is a snapshot of current commitments, not a record of what arrived in your bank account.

The simplest version: add up what each active subscriber pays per month.

Say you have:

  • 30 customers on a $10 monthly plan
  • 8 customers on a $25 monthly plan
  • 6 customers on a $96 yearly plan

MRR is 30 × $10 + 8 × $25 + 6 × ($96 / 12) = $300 + $200 + $48 = $548.

Notice that the yearly customers contribute $8 each, not $96. That is the most common mistake: counting an annual payment as a month's revenue. It makes a good month look spectacular and the following eleven look like a collapse.

What does not count

MRR is meant to capture only the predictable, recurring part of revenue. Leave these out:

  • One-time purchases. Lifetime deals, setup fees, one-off templates or credits packs. They are real revenue, just not recurring.
  • Usage overages that vary wildly. If some months a customer pays $5 extra and some months $200, most people either exclude overages or use a conservative average.
  • Trials. A trial user is not paying yet. Count them when the first charge succeeds.
  • Failed payments. If a card has been declining for weeks, that customer is at risk; many makers remove them from MRR after a set number of failed retries.
  • Discounts at full price. If someone pays $5 on a 50% coupon, they contribute $5, not $10.

When you read another maker's MRR figure, you usually cannot tell which of these rules they followed. That is fine for rough comparison but worth keeping in mind.

MRR is a flow, not a pile

The headline number hides movement underneath it. A more useful view breaks each month's change into parts:

Component Meaning
New MRR Revenue from customers who started paying this month
Expansion MRR Increases from existing customers upgrading or adding seats
Contraction MRR Decreases from customers downgrading
Churned MRR Revenue lost from customers who cancelled
Net new MRR New + expansion − contraction − churned

Two products can both grow from $2,000 to $2,200 MRR in a month and be in very different shape. One added $250 of new revenue and lost $50. The other added $900 and lost $700. The second is running hard to stay roughly in place, and the underlying problem is retention, not acquisition.

Tracking these components, even in a spreadsheet, tells you where to spend your time. See Reducing Churn for Small SaaS for the retention side.

Revenue is not income

MRR is revenue before costs. From it you still subtract:

  • Payment and platform fees, which can take a noticeable percentage, more if you use a merchant of record.
  • Hosting and third-party services, which for AI-heavy products can scale uncomfortably with usage.
  • Tools and subscriptions you pay for to run the business.
  • Taxes, depending on where you live.

A product with $1,500 MRR and $1,200 of monthly model and infrastructure costs is a much thinner business than one with $1,500 MRR running on a single small server. When you browse the AI Tools category, keep that in mind: headline revenue in that space often carries heavier per-user costs than a simple CRUD app.

Small numbers are noisy

At low MRR, one customer is a big share of the total. If you have 20 customers and two cancel in the same week, your churn for the month looks alarming, but it might just be chance. If one agency upgrades to a big plan, your growth rate looks heroic.

A few habits help:

  • Look at trends over three months or more, not month to month.
  • Count customers as well as dollars. Growing MRR while customer count falls means you are becoming dependent on a few accounts.
  • Watch concentration. If your biggest customer is more than a quarter of MRR, losing them would hurt badly. That is worth knowing.

Reading other people's MRR

When you see a figure in a maker post, a few questions help put it in context:

  • How long did it take? $3k MRR after four years is a different story from $3k after four months, and both can be fine.
  • Is it revenue or profit? It is almost always revenue.
  • Is it one product or several? Some makers report a portfolio total.
  • Is it a peak or the current level? Posts tend to be written after good months.

None of this means makers are dishonest. Most are sharing openly and generously. It just means a single number, posted on a good day, tells you less than it seems to. The X posts in this catalog, which often come in threads over time, give a fuller picture than any one snapshot.

FAQ

Is ARR just MRR times twelve?

Yes. It is a run-rate projection, not what you earned in the past year. A product at $500 MRR has $6,000 ARR even if it launched last month.

Should I include one-time revenue anywhere?

Track it separately. It is real money and can matter a lot for a small product, especially lifetime deals. It just does not belong in the recurring number.

When does MRR start to feel meaningful?

That depends entirely on your costs and goals. For many solo makers, the first milestone that matters is covering the product's own costs, then covering a meaningful slice of personal expenses. Both are worth celebrating.

Related reading