How Do You Forecast MRR Growth?

Forecast each line of the MRR waterfall separately, then let the total fall out of the sum. Applying a blended growth percentage to current MRR produces a number that looks reasonable and explains nothing. A business can grow MRR 4% in a month while churn accelerates underneath, because expansion covered the loss. The aggregate hides the movement, and the movement is the part you can act on.

Forecast the waterfall, not the total

The monthly reconciliation runs from beginning MRR, plus new customer MRR and expansion MRR, less contraction MRR and churned MRR, to ending MRR. ORM builds the same reconciling waterfall on ARR by month, separating churned customer revenue, churned product revenue, and product downgrades on the loss side from new customer revenue, new product revenue, and product increases on the growth side. Every one of those lines has a different driver and needs a different method.

New MRR comes from pipeline, not from a trend line

The new-business line is a sales forecast wearing different units. It needs pipeline by expected close month, stage and segment conversion rates, a grounded win rate, and a realistic view of deal size.

Deal size is where teams overstate. Open pipeline routinely carries a higher average deal value than closed-won deals do, and the gap is often large. A pipeline carrying an $80,000 average deal size against $40,000 on closed-won deals is the shape of the problem. A new-MRR forecast built on pipeline value at face value runs high for exactly that reason. Build it on what deals actually close at.

Capacity caps the line from the other side. Ramping reps do not carry full productivity, so a plan that adds five reps in month one does not add five reps of new MRR until they ramp.

Expansion and churn need their own curves

Expansion and churn are properties of cohorts rather than of the whole base. A cohort's retention curve at month 12 predicts next year's churn far better than this month's blended churn rate does. Forecast churn against the renewal calendar, because a month with heavy renewal exposure carries different risk than a month with almost none.

Expansion is the line most often modeled as a flat percentage. Tie it instead to the accounts whose seats or usage are approaching their contracted ceiling, and to net revenue retention by cohort rather than one company-wide rate.

Build it at monthly granularity

Monthly MRR forecasts inherit the seasonality of the sales motion feeding them. ORM's view is that Q2 and Q4 run stronger than Q1 and Q3, with the third month of a quarter stronger than the first two. A model that spreads a quarterly new-business number evenly across three months lands short in month three and long in months one and two, every quarter. Building the sales forecast monthly from the start removes that error.