Two properties in the same portfolio both report 45 days.

At the first, units are rent-ready in five days and then sit for forty. At the second, the turn takes forty days and the unit leases within five of going live.

Same number. Opposite problems. One needs marketing and leasing attention, the other needs a maintenance and coordination fix, and nothing on the report tells you which is which.

Now put both into a portfolio average and report it to an owner.

Three metrics, one name

Before any benchmark is useful, you have to know which of these you are holding.

• List-to-lease measures from the day a unit is listed to the day it is leased. It starts when marketing starts. The make-ready period is not in it at all.

• Days vacant measures from move-out to move-in. It includes the turn, the listing period, the application and the gap before the new resident actually takes occupancy.

• Time to lease is used loosely and rarely defined. Sometimes it means one of the above, sometimes something in between.

A unit that spent 30 days in turn and leased 10 days after going live reports as 10 days on the first metric and around 40 on the second. Both figures are correct. Neither is wrong. They are answers to different questions.

This is the same problem we covered in what counts as a lead the number is fine, the definition underneath it is missing, and everyone quoting it assumes theirs is the shared one.

One source with a clearly stated methodology

Most days-on-market figures in circulation arrive without a definition. Apartment List's time on market index is worth using as your reference point precisely because it tells you what it counts.

Their methodology defines list-to-lease as the number of days between a unit's listing date and its lease date, built from listings active on their marketplace, with tracking beginning in 2019.

Their August 2026 report puts the figure at 32 days, up two days from the previous month. Three days longer than the same point a year earlier, and around two weeks longer than August 2021 when the market was at its tightest.

Note what that 32 days does not include. Not the turn. Not the days between a resident giving notice and the unit being ready to show. If your internal number is 45 and you are comparing it to 32, you may be comparing a figure that includes make-ready against one that does not, and concluding you have a leasing problem you do not have.

The seasonal swing is bigger than most performance gaps

Here is the part that quietly ruins casual benchmarking.

August 2026 came in at 32 days, which Apartment List note was the longest August in their series since tracking began. The overall record in that series was set in January 2026, at 41 days.

Nine days of variation inside a single year, driven by season rather than by anything any operator did.

So a property reporting 40 days in January is performing roughly in line with the national picture. The same property reporting 40 days in August is meaningfully behind it. Compare either one against an annual average and you get the wrong answer in both directions.

If you benchmark, benchmark against the same month last year. Comparing January to an annual figure is comparing two different market conditions and calling the difference performance.

About those asset-class benchmarks

Search for days-vacant benchmarks by asset class and you will find a consistent set of figures. Class A multifamily at 35 to 45 days with a target under 30. Workforce multifamily at 45 to 60 with a target under 40. Class C running past 60. Single-family per-turn vacancy at 30 to 45 days.

The shape is plausible and matches what most operators would guess. But follow those numbers back and they come from vendor-published content without a stated sample, a stated methodology, or a source. They are informed estimates presented as benchmarks.

That does not make them useless as orientation. It does mean you should not set a target against them, take them into an owner meeting, or treat a gap against them as a finding. We made the same argument about the funnel conversion figures everyone quotes, and about the 86% self-guided tour statistic. The industry has a habit of repeating a figure until repetition becomes provenance.

There is a further problem with asset-class benchmarks specifically. Class designations are not standardised. One operator's Class B is another's Class A minus. So even a well-sourced benchmark by class would be comparing categories that each company defines for itself.

The split that actually tells you something

Forget the benchmark for a moment. The useful number is not one figure, it is two.

• Make-ready days. From the unit becoming physically available to it being rent-ready.

• Marketing days. From rent-ready and listed to lease signed.

Take the two properties from the opening. Five and forty at one, forty and five at the other. Once the number is split, both diagnoses take about four seconds, and the interventions are completely different. One is a leasing and pricing conversation. The other is a scheduling and vendor conversation that leasing cannot fix no matter how hard it works.

Add a third number and it gets sharper still: notice-to-available days, the gap between a resident giving notice and the unit becoming physically available. That period can give operators a head start on planning the turn, and portfolios vary considerably in how effectively they use it.

Five dates, and most reports carry two

To produce that split, five dates have to exist against each unit:

Many operations can readily produce moved-out and leased dates. Everything between gets reconstructed afterwards from work orders, calendars and someone's memory of when the photos went up.

Which is why the blended number survives. Not because anyone prefers it, but because it is the only one the data supports without a week of reconciliation. Same structural issue we covered in why your response time number is probably wrong: without intermediate timestamps you know the total and you cannot say where it went.

Why the portfolio average is the weakest number you report

One figure across a mixed portfolio is close to meaningless as a diagnostic, and the primary data makes the case for you.

Apartment List's vacancy index uses stabilised properties rather than newly developed lease-ups, with properties entering the sample only after meeting its stabilisation criteria: active on the platform for at least six months and having reached 85% occupancy at least once. A property absorbing dozens of units at once behaves nothing like a stabilised asset turning a few units a month.

If the organisation publishing the benchmark separates those two states, your internal reporting probably should too.

The same applies within stabilised property. A Class A asset in a tight submarket and a Class C asset in a soft one have different demand curves, different renter behaviour, and different achievable numbers. Averaging them produces a portfolio figure that moves when your mix changes rather than when your performance does.

Report by property. Look at the spread, not the middle. The gap between your best and worst asset is usually a more actionable finding than either number alone, and it is the version that survives an owner asking why.

What to measure

• By property, not by portfolio. Always.

• Split into make-ready days and marketing days. The single highest-value change in this article and the one that changes what you do on Monday.

• Against the same month last year. Not against an annual average, and not against a figure from a different part of the season.

• With the definition stated on the report. One line saying which dates were used. When somebody asks whether 45 days is good, that line is your answer to "45 days of what."

• Separating lease-ups from stabilised assets. Different states, different expectations, different reports.

Common mistakes

• Comparing your days-vacant figure to a list-to-lease benchmark. Your number includes the turn and theirs does not. The gap looks like underperformance and is arithmetic.

• Benchmarking a winter month against an annual average. A nine-day seasonal swing sits inside a single year of the best available index.

• Treating vendor-published class benchmarks as targets. No methodology, and class definitions vary by company.

• Reporting one portfolio number to owners. It moves when your mix changes and it cannot explain itself. The owner-reporting version of this argument is in what your owner report doesn't say about leasing.

• Treating a long number as a leasing failure by default. Until the make-ready portion is separated out, you do not know whose problem it is, and leasing usually gets blamed because leasing is closest to the outcome.

Back to the two properties reporting 45 days. Both numbers are accurate. Both are useless, because the figure was produced by adding together two periods that belong to different teams and have different fixes.

Splitting them takes five dates recorded as they happen rather than reconstructed in a spreadsheet afterwards. That is the whole exercise, and it is the reason most portfolios still report one number.

That is one of the operational problems RIOO is designed to address. RIOO's leasing workflow holds unit availability and leasing status alongside applications and lease execution rather than in separate systems, and dashboards and reports read from that same data, so a question like how much of our vacancy is turn and how much is leasing becomes answerable by property.

Take your last ten turns and write down the rent-ready date for each. If that date is hard to find, you have located the reason the number has never been split.

Frequently asked questions

Q1. What is a good days-on-market figure for a rental property?
It depends entirely on which metric you mean. Apartment List's list-to-lease index, measured from listing date to lease date, sat at 32 days in August 2026 and peaked at 41 days in January 2026. A days-vacant figure that includes the make-ready period is not comparable to either.

Q2. What is the difference between days on market and days vacant?
Days on market, or list-to-lease, runs from the day a unit is listed to the day it is leased. Days vacant runs from move-out to move-in and includes the turn. The same unit can show 10 days on one and 40 on the other.

Q3. Are there reliable days-vacant benchmarks by asset class?
Figures circulate for Class A, workforce and Class C multifamily and for single-family rentals, but they are vendor-published without stated methodology, and class definitions are not standardised across companies. Treat them as orientation, not targets.

Q4. How much does days on market vary by season?
In Apartment List's index, the range across 2026 alone ran from 32 days in August to a record 41 days in January. That variation is larger than most performance gaps, which is why comparisons should be made against the same month in a prior year.

Q5. Which number should I actually report?
Make-ready days and marketing days, separately, by property, with the definition stated. The split tells you where the time went. The blended figure only tells you that it went.