Three numbers claim to tell you how many residents stayed, and they will not agree. Renewal rate counts lease decisions. Retention rate counts households. Turnover rate counts units. One building, one year, three denominators, three answers, and all of them can be correct at once.
Most reporting arguments start here. Someone reports a strong renewal rate on Monday and a rough turnover rate on Thursday, and nobody can say which one is lying. Neither is. Here is what each number counts, and the small set of cases that decide where it lands.
Turnover rate counts units, not people
Turnover rate is the share of units that changed occupant during a period. It measures the property, not the residents.
That is the definition sitting underneath the Canadian market data you are most likely to benchmark against. In its Rental Market Survey, CMHC counts a unit as turned over if it was occupied by a new tenant who moved in during the past 12 months, and states that a unit can be counted as being turned over more than once in a 12-month period. Two things follow. A unit that turned twice contributes twice, so the figure is not bounded the way a share of your unit count feels like it should be. And the survey covers row and apartment structures with three or more units, collected in the first two weeks of October, so a scattered-site portfolio of houses, duplexes and individual condo units sits outside that universe entirely.
Before you compare your turnover to a market figure, match the period, the property type and the counting rule. Without that, the comparison is decoration.
Renewal rate counts decisions, and most tenancies never make one
Renewal rate is renewals divided by the tenancies that reached a renewal decision during the period. The numerator is easy to pull. The denominator is where the number breaks.
Across most of Canada, a fixed term running out is not a decision. Ontario's Landlord and Tenant Board guide is direct about it: the end of a lease does not mean a tenant has to move out, and a tenant who paid rent by the month keeps the right to stay as a monthly tenant, with the rules of the former lease still applying. Nobody signs anything. So if your denominator is renewal offers sent and your numerator is renewal agreements signed, every resident who rolled over month to month drops out of both sides of the fraction. The number you publish then describes the smaller group who signed paper, not the larger group who stayed.
Two definitions survive contact with that reality. Pick one on purpose:
- Term-end retention. Of the tenancies whose fixed term ended in the period, how many were still in place 30 days later, whether they signed a new term or continued month to month. This is the number that answers whether people stayed.
- Signed renewal rate. Of the tenancies you offered a new fixed term, how many signed one. This is a measure of your renewal campaign, not of resident behaviour. Keep it if fixed terms are part of how you operate, and never present it as retention.
Whichever you choose, write the denominator next to the number every time you report it. If you want the legal side of why the fixed term is not the event people assume, we covered that in the lease renewal myth.
Retention rate follows the household
Retention rate is the share of households in place at the start of a period that are still in place at the end. It is the only one of the three that tracks people rather than paper or property.
That is what makes it the right home for internal transfers. A resident who moves from a one bedroom to a two bedroom down the hall is a turnover on the first unit and a retained household at the same time. Both statements are true. Turnover has to count that move and retention has to not count it. If your two numbers rise and fall in lockstep, one of them is being computed wrong.
Four cases decide where your number lands
Run these through your definition before you publish anything. They are where portfolios of every size lose comparability, including with their own prior year.
- Mid-term move-outs. A broken lease, an assignment or an eviction in month seven never reaches a renewal decision at all. It belongs in turnover and in retention. Leave it out of the renewal denominator and your renewal rate can sit in the nineties while a third of your units turn.
- Month-to-month continuations. A resident who stays without signing is a retention. Decide once whether they also count as a renewal, then apply that everywhere, including last year's figures when you restate them.
- Internal transfers. Turnover yes, retention no. Tag the move-out and the move-in as a matched pair in your system, or you will book the household as lost and the new tenancy as a win, which flatters two numbers at the same time.
- Units that turn twice. A unit that housed three tenancies in a year is one unit and two turnovers. CMHC counts it twice. If your own report caps each unit at one, say so on the report, because you are measuring something else.
Report all three, and label them
The fix is not to pick a favourite. It is to publish all three with their denominators attached, over one period everyone has agreed to.
- Fix the period first and use the same one for all three numbers. A calendar year and a lease year give different answers from identical data.
- Freeze the denominator as a rule, not as a query. Tenancies whose fixed term ended between January 1 and December 31 is a rule. Whatever the renewals report returned this morning is not, because it shifts as records get edited.
- Count move-outs from move-out dates, not notice dates. Notice gets withdrawn and dates get pushed.
- Put the definition in the report, one line under each figure. The next person to read it will not have this conversation to refer back to.
This matters more the smaller you are. In a 40 unit portfolio, one household is 2.5 points, so a single unresolved question about whether an internal transfer counts can move your headline number further than any real change in how long people are staying.