Two sentences that get said in the same meeting, by the same person, ten minutes apart:
"We've got loads of availability next week."
"There's no way we can fit another showing in."
Both are true. They are not a contradiction and they are not confusion. They are what happens when you count calendar slots and your business actually runs on something else.
The unit of capacity is the slot plus the drive
Open the calendar and you see half-hour blocks. Twelve of them on a Tuesday, four booked, eight free. Eight showings of spare capacity, apparently.
Now put the map next to it. Those eight free blocks are spread across a Tuesday during which somebody has to be at a house in one suburb at ten and a walk-up across the city at two. The real unit of capacity is not the half hour. It is the half hour plus getting there plus getting back, and once you count it that way a 30-minute showing at a scattered property is routinely a 90-minute commitment.
That single change of unit explains the whole problem. Your calendar says eight. Your week says three. And every conversation about "being more efficient with showings" is really a conversation about the map, not about the calendar or the effort of the person holding it.
Two things follow immediately. Leasing capacity at a scattered portfolio is a routing constraint, not an effort constraint — you cannot fix it by asking anyone to work harder, and a team that is already driving efficiently has no slack to find. And capacity is not one number for your firm. It is a different number for each property type, on each day, depending on what else is on the calendar that day.
A worked week, twice
Same problem, two cities, and the arithmetic runs differently in each. Every input below is an assumption, stated so you can substitute your own — none of these are published statistics, and your drive times are the only ones that matter.
Fourteen vacant units across Greater Phoenix
Assume: a broad, low-density metro; typical inter-property drive 25 minutes; showings 30 minutes; one person doing them; a 9-to-5 day with an hour for everything that is not driving or showing.
- One showing costs 30 + 50 = 80 minutes if you return to base, or 30 + 25 = 55 minutes if you chain straight to the next one.
- Chained back-to-back, a seven-hour day holds roughly seven showings. Unchained — an applicant here at ten, another across the metro at two — you get four or five, and the gaps are unusable.
- Fourteen vacant units needing, say, three showings each is 42 showings. At seven a day that is six full days of driving. At four a day it is ten.
The spread between those two numbers is not effort. It is sequencing, and it is worth approximately four days a week to whoever owns the calendar.
Fourteen vacant units across Greater Montreal
Assume: a denser core with a dispersed off-island periphery; drive 15 minutes between properties in the core and 45 minutes to the periphery; the same 30-minute showings.
- In the core, chaining gives you 30 + 15 = 45 minutes a showing — nine or ten in a day.
- One periphery property inserted into the middle of that day costs 90 minutes of driving and removes two core showings as well as its own.
- And there is a seasonal input a Phoenix operator does not have: winter driving inflates every one of those numbers for months at a time, and a January estimate built on a September drive time will be wrong in the direction that hurts.
The Montreal week is denser and therefore more efficient, right up until one outlying property lands in the middle of it. That is the general rule underneath both cities: the cost of a showing is set less by where it is than by what it sits between.
Clustering by geography or by property, and when each one loses you a lease
There are two ways to group, and the trade-off between them is real.
Cluster by geography — everything in the north end on Tuesday — and you minimise driving. The cost is that an applicant who wants Thursday for a north-end unit is offered Tuesday or nothing, and some of them are simply gone.
Cluster by property — all showings for the Elm Street building in one block — and you get the same efficiency at that address plus the ability to answer questions once. The cost is that the block is only worth running if enough people want that property in that window.
Neither is correct in general. What decides it is how scarce the applicant is. In a tight market, cluster hard: applicants have few alternatives and will move to fit you. In a soft market, where an applicant has a dozen options, insisting on your grouping is how the efficiency you gained on the road gets paid for in lost applications.
And this is where the vacancy cost from the per-day arithmetic comes back. If a day of vacancy at that property costs more than the driving you saved, you optimised the wrong quantity.
Keys and access are a routing cost, not an admin cost
The key that lives in an office you have to drive past is not a paperwork problem. It adds a leg to the route, and it adds it at the start of the day when the route is most expensive to change.
Whatever access model you run — and this article is deliberately agnostic about which — the question to ask about it is not "is this secure and tidy" but "how many legs does it add to a week." An access step that costs fifteen minutes each way, twice a day, four days a week, is two hours. That is a showing and a half you did not run, every week, forever.
Where batching helps, and where it quietly costs you
Batching is the obvious answer and it works — up to the point where it starts eating the thing you were protecting.
It helps when the properties are close together, when the applicants are flexible, when the market is tight, and when you can fill the block. It costs you when you are batching a slow property, because you have converted "available whenever you like" into "available Tuesday between two and four", and that offer is much weaker in a market where the applicant has options.
The honest test is one question: is this block full because demand is genuinely concentrated, or because I only offered two hours? If the second, the efficiency is imaginary — you did not compress the demand, you truncated it.
What changes when applicants pick the slot
Here is the part that actually moves the arithmetic, and it is not a smarter route.
When you choose the times, every slot is a negotiation, and the negotiation happens by email across a working day. When applicants choose from what you have made available, the negotiation disappears and your week fills itself in whatever shape you opened.
Releaser works this way: one shareable link per property, applicants complete your screening before they reach your calendar, and the ones who meet your criteria book their own slot.
But that only helps if the calendar you opened already encodes your constraints — and that is the honest catch in this whole approach. Self-booking does not know your map. It fills whatever you make available, so if you open a Tuesday that requires being in two suburbs at once, somebody will book exactly that, and the person driving will discover it on Monday night.
So the work moves earlier. Instead of negotiating each slot, you decide, in advance, which windows exist for which properties — north-end properties open these hours, the walk-up opens those, the periphery house opens a block on the day you are already out there. Then let people book inside it.
That is a configuration job rather than a scheduling job, and it is the difference between self-booking saving you a week and self-booking generating a route nobody could drive.
Turning geometry into an annual number
One last piece, because a routing problem you only feel on bad Tuesdays is easy to under-invest in. Turnover frequency tells you how often you will run this gauntlet.
In Canada, CMHC publishes tenant turnover rates, and the spread inside the Canadian data is large: broken out by rent quartile for 2025, CMHC reported Edmonton turnover running from 22.5% in the least expensive quartile to 33.6% in the most expensive, while Toronto ran from 7.9% to 12.1%. Applied to a scattered portfolio, that is the difference between turning a third of your units in a year and turning a tenth.
Multiply your own turnover by the showings each unit needs, then by the slot plus the drive rather than the slot, and you have your annual leasing-travel budget in hours. Most operators have never calculated it, and it is usually larger than the thing they are currently trying to optimise.
For US operators there is no equivalent published national turnover rate to look up — the Census Bureau's vacancy release does not carry one, and the NAA's benchmarking covers costs rather than frequency, with administrative and payroll expenses reported at $2,323 per unit for 2024, up 3.81% year over year. So the frequency number has to come from your own records. Twelve months of turns, counted by property type, is enough.
Releaser is available to property management firms in the United States and Canada, with custom configuration for different property types. Each property becomes one shareable link, applicants complete your screening before they reach your calendar, and the ones who meet your criteria book their own showing.
Sources. Canada: CMHC 2026 Mid-Year Rental Market Update and the 2025 Rental Market Report. United States: National Apartment Association on its 2024 Income/Expense IQ, and the US Census Bureau, Housing Vacancies and Homeownership. The two worked weeks are arithmetic from the assumptions stated in the text, not survey data.