Monthly Recurring Revenue

When a business sold something once and moved on, a single revenue figure for the month was enough. Sell $25,000 worth this month and that's the month's work; next month starts from zero again. As more companies moved to selling by subscription, that one figure stopped describing the situation.

The problem is that money of two very different kinds sits in the same row. When a company says it earned $75,000 this month, part of that is subscription fees that charge themselves automatically, and part is a one-time setup fee for standing up a new customer's environment. The first kind arrives again next month whether anyone does anything or not. The second ends there. With both written in one cell, next month is unpredictable.

This plays out badly in practice. A company reported $75,000 in revenue to its board, $30,000 of which was one-time setup work. The following month came in at $45,000, and everyone reading the number concluded the business had shrunk 40 percent. The recurring portion had been exactly $45,000 in both months.

Monthly Recurring Revenue (MRR) exists to split that cell in two. Count only the money that will still be there next month and the height the company is standing at becomes visible; add how much that height changed this month and you can see the speed as well.

Take every recurring contract, convert each to a one-month amount, and add them up. A hundred customers paying an average of $400 a month comes to $40,000.

There are four steps.

1. Keep only recurring items. Setup fees, training fees and one-time consulting don't come back next month, so they drop out.
2. Normalize every billing cycle to a month. A contract worth $4,800 a year enters as $400, not $4,800. A quarterly payment of $1,200 is also $400.
3. Use what's actually billed, not list price. A customer on a 20 percent discount counts at the discounted amount.
4. Add up every customer.

Step 2 is where numbers go wrong. Put the full $9,600 into the month an annual customer paid it and the chart spikes once and then flattens for eleven months. The principle is to spread the amount across the period being served rather than the moment cash arrived; the part not yet delivered is tracked separately as deferred revenue.

That covers the size. What teams look at more often is the change from last month, split into four pieces.

1. New MRR — added by customers paying for the first time this month.
2. Expansion MRR — added by existing customers buying more seats or moving to a higher plan.
3. Churned MRR — lost to cancellations and downgrades.
4. Net new MRR — the three combined. New of $6,000, expansion of $2,500 and churn of $4,000 leaves $4,500.

The absolute figure has no benchmark. Whether $40,000 is good depends on what last month was and what the growth cost, so what actually gets read is the rate of change and the composition.

The rate of change is the percentage added over the prior month. For an early subscription company, 10 percent a month is fast and 5 percent is steady. Holding the same percentage gets harder as the base grows: 10 percent of $8,000 is $800, while 10 percent of $400,000 is $40,000.

Composition is where the growth came from. Net new of $4,500 built from $8,500 in new business against $4,000 of churn is a different company from one that got there with $5,000 new and $500 churned, even though the headline is identical. The first is pouring water hard into a leaking tub, and the month marketing spend drops, revenue drops with it.

That's why churn rate travels alongside. It's recurring revenue lost divided by last month's recurring revenue — roughly 3 to 5 percent a month is the working benchmark when individuals and small teams are the customer base, and under 1 percent when the business sells to enterprises. Net revenue retention, which measures how much revenue existing customers hold on to once expansion is counted, tells you something further: above 100 percent, revenue grows even in a month with no new customers at all.

"MRR is $55,000 this month but net new is only $1,500."

The total looks healthy, but as much is leaving as arriving and growth has nearly stopped.

"That contract is setup work, so it doesn't hit MRR. Flag it separately."

A request to keep one-time revenue out of the recurring chart.

"For annual prepayments, please divide by twelve."

An instruction to calculate on the service period rather than on the cash received.

At Pagedeck, Grace pulls this number together in the first week of every month and sends it to investors. What goes out isn't a single total but a table broken into the four pieces: $52,000 last month, $3,600 new, $1,200 expansion, $1,800 churned, $55,000 now. Sending only the total guarantees a reply asking for the breakdown, so the breakdown goes first.

In a month when Sofia closes a $30,000 annual contract, Grace notes it under the table. Only $2,500 lands in recurring revenue while the whole amount hits the bank that month, and the balance and the chart would otherwise look like they disagree.

Reading the bank balance as recurring revenue is the most common mistake. The money arrived, so counting it in that month feels natural. But this metric follows when the service is delivered, not when the cash landed. Book an annual prepayment in full and the chart leaps once and stays flat for eleven months, and nobody can tell from it whether the company is growing.

Reading a rising total as good news is the other one. The number did go up, which is hard to argue with. But recurring revenue also jumps on a single large customer. In a month when one $6,000 account pushes the total up 8 percent, twenty small customers can be leaving at the same time, and next quarter gives the gain straight back. The total is the result; the condition is written in the pieces.