Self-storage ECRI calculator: what one rate increase is worth

Whether to send this batch, at this percentage, this month. It is the arithmetic behind a single decision, not a rate program.

The gap between a rent and your street rate measures opportunity. It says nothing about whether the tenant stays, and that is the only thing the return depends on.

Six lines, all of them visible on the page:

With the defaults, 100 tenants at $118 and a 12 percent increase over twelve months: each tenant who stays pays $14.16 more a month, or $169.92 over the year. 92 stay, for $15,632.64. Eight move out at $177 each, for $1,416. Net: $14,216.64.

• Gain from the tenants who stay: tenants notified, times one minus the added move-out rate, times average in-place rent, times the increase percent, times the months the increase is collected.
• Cost of one move-out: one month of rent for each month the unit sits empty, plus the move-in concession given to the replacement tenant, plus any other cost entered.
• Expected cost: tenants notified, times the added move-out rate, times the cost of one move-out.
• Net impact: the gain, less the expected cost.
• Break-even added move-out rate: the gain per tenant who stays, divided by that gain plus the cost of one move-out.
• Net as a share of the best case: the net impact divided by the gain if every tenant stayed.

Read the net as a well-bounded estimate, not a forecast. Two assumptions make it optimistic and two make it conservative, and they do not cancel.

• The replacement is assumed to pay the same rent once the concession ends. Where in-place sits below street a move-out is cheaper than modelled, and where it sits above street it is dearer.
• Turn labor, a lock, and any marketing or commission cost on the replacement count only if entered as other cost. Left at zero, the cost of a move-out is a floor.
• Ordinary churn is not modelled: every tenant who stays is credited the full months collected. Shorten the months for a batch likely to turn over anyway; at eight months the default break-even falls from 49 to 39 percent.
• The tenants who leave are assumed to pay none of the increase, which understates the gain.
• The added move-out rate is your estimate of churn the increase causes beyond normal churn. Fit it to what your own past batches did.
• There is no street-rate ceiling here. Check the new rent against your own asking rate before sending, and keep delinquent tenants out of the batch entirely.
• Notice periods, any statutory limit on increases, and the contents of the notice are set by the lease and by state law. Confirm both with counsel.

The break-even rate is the number to argue about. At the defaults it is 49 percent: this increase pays unless it pushes out roughly half the batch, which it will not. That is why gap-based programs usually do net positive, and it is not a reason to stop checking.

What the batch total hides is the distribution. A program can net positive while a specific slice of it, long-tenured tenants in soft unit types, loses money on every raise. Run those tenants as their own batch and the net tells you so.

Raise the percentage and the gain moves in a straight line while churn accelerates, so run the same batch at several percentages and compare the net rather than the gain.

In Verillum AI, that same expected-value step is the last gate on every per-unit recommendation, and what each past batch actually did is read off the next rent roll rather than guessed. A demo shows it running against a full roll.