When the renewal number is algorithmically right, and the outcome is still a loss.

By Adam Burdick, Goldshine Management Consulting

At a 240-unit garden community in the Washington, DC, metro area, I watched a resident decline her renewal, move out, and sign a new lease on a nearly identical unit in the same community for less than the amount she’d just turned down. The owner paid for a full turn to move his own resident down the hall at a discount. The vacated unit was then leased at a rate below the declined number anyway.

Run the audit afterward, and everything was algorithmically sound. The renewal followed the model. The new-lease pricing followed the model. Every screen was green. And at the end of it, rents didn’t grow, NOI went down, and the owner wrote a four-figure check for the privilege.

This Is Not an Argument Against Raising Rents

Let me be precise about the claim because it isn’t a plea to hold rents down. Increases are normal, and most residents accept them. The claim is narrower and more uncomfortable: somewhere in your renewal batch are residents who want to stay and will leave over a number no human would have sent them. The real question is whether anyone with judgment, authority, and the right incentives reads those renewals before they go out. At most companies, the honest answer is no. The batch is generated, the batch is compliant, and the batch is sent.

The Resident the System Can’t Say Yes To

Here’s the version I see most often. A guy moves for work and takes a seven-month lease. Short-term leases carry an algorithmically generated premium, so they pay $1,850 on a floor plan whose twelve-month rate is $1,600. The lease ends. He likes the place, his job worked out, and he wants to stay.

Keeping him means offering a number that looks like a rent cut. Converting him to a twelve-month lease at $1,650 would put him above your standard rate and lock in another year with zero vacancy. But the software reads $1,850 to $1,650 as an 11% decrease and won’t recommend it. The site team gets graded on trade-out, and nobody wants to defend a screen full of red. So, the renewal comes out to about $1,875, he finds the same floor plan across the street for $1,600, and he’s gone.

Count what happened. You lost a resident who asked to stay, with a clean payment file, and replaced him with an unknown. The system’s premium was meant to cover these costs during his lease term, but it’s not like those premiums are put into a separate account for turn costs. Those premiums were spent on the day his rent check cleared. You paid for a turn and ate three weeks of vacancy. The rate you refused to offer him is higher than the rate you accepted from the stranger. On the trade-out report, nobody did anything wrong.

The Ledger of One Unit
Rate he asked to stay at (12-month conversion)	$1,650
Renewal offers the system sent	$1,875
Outcome	Declined. Moved across the street.
Turn cost (make-ready, marketing, leasing)	$3,872
21 days vacant at $53/day	$1,113
Unit re-leased at	$1,625
Year-one cash vs. simply saying yes	about -$5,300

The rate he offered to keep paying is $25 a month more than the rate the stranger pays now. The $5,300 bought nothing.

A Rounding Error for Them. Real Money for You.

When this happens inside an 80,000-unit REIT, it registers as a blip on a dashboard. Averaged across a national portfolio, forced turns are statistical noise, and the pricing strategy that causes them still pencils out. That math is not available to you.

The national average turn runs $3,872 per unit, including make-ready, marketing, leasing time, and concessions, per Zego’s industry survey. [1] Spread that across a portfolio at the national turnover rate of 47.5%, and a 250-unit owner is carrying roughly $460,000 a year in turns. [2] Five avoidable ones in a quarter is not a blip on your screen. It’s a distribution.

Portfolio Size	Annual Turns	Total Yearly Cost
100 Units	48 Turns	$185,856
250 Units	119 Turns	$460,768
500 Units	238 Turns	$921,536

The Supply Wave Will Grade Your Renewal Process for You

Whatever slack existed for autopilot renewals is disappearing in supply-heavy markets. Look at Raleigh. The Triangle delivered more than 13,000 units in 2024 and roughly 12,000 more in 2025. [3] By the first quarter of 2026, two-thirds of properties in the market were offering concessions, among the highest concession rates in the country, and metro rents sat about 2% below the prior year. [4][5] Every one of those lease-up buildings is a cheaper option, with a leasing team focused directly on your residents.

National turnover sits at a two-decade low of 47.5%, and it’s tempting to read that as loyalty. [2] Residents stayed because there was nowhere cheaper to go. In Raleigh, there is now. Run your own arithmetic: if even one in five of your expiring leases faces a lease-up building down the road offering two months free, what happens to your renewal book? The historically low churn you’re underwriting is one supply cycle from gone.

The Fix Is People and Incentives, Not Software

None of this requires new technology. When we rebuild renewal processes for owners, the changes are unglamorous:

• A human reads the exceptions. Before the batch goes out, someone with authority reviews the renewals that look like the stories above: short-term premiums converting to standard terms, long-tenured clean-pay residents facing above-market increases, and any offer that exceeds what you’re currently asking strangers to pay for the same floor plan. That last check alone would have caught both residents in this article.
• Authority to say yes. The site manager can approve a conversion or a flat renewal without a corporate escalation that outlasts the resident’s patience.
• An avoidable-turn report. Definition: a resident declines a renewal and the unit re-leases at or below the declined offer within 60 days. That turn was forced, and it’s a different animal from the job transfer or the divorce you were never going to save. Count the forced ones, cost them at your real turn number, and put the total next to trade-out on the monthly report. A sample month at a mid-size portfolio: three forced turns at $3,872 is $11,616 spent chasing roughly $2,400 a year of premium.
• Comp that matches the goal. If every bonus in the chain keys on trade-out and none keys on retention cash, you have told your team exactly which residents to lose.

The whole program costs about an hour of someone’s month. One caught renewal pays for years of it.

What to Take Away

Before the next renewal batch goes out, ask one question: who reads it? If the answer is nobody, you aren’t pricing renewals. You’re generating turns and booking the paperwork as rent growth.

Sources

1. Multifamily Dive, “Turnover costs hold steady at nearly $4,000 per resident” (Zego survey) — https://www.multifamilydive.com/news/turnover-costs-4000-apartment-multifamily/696298/

2. CBRE multifamily turnover research, cited in “Apartment Turnover Rate: National Trends” — https://nspireexperts.com/apartment-turnover-rate/

3. Northmarq, “Vacancy levels off as new multifamily construction starts decline in Raleigh-Durham” — https://www.northmarq.com/insights/insights/vacancy-levels-new-multifamily-construction-starts-decline-raleigh-durham

4. Rental Beast, “Raleigh Rental Market Report Q1 2026” — https://blog.rentalbeast.com/post/raleigh-q1-2026

5. Axios Raleigh, “Raleigh rents were down year-over-year as new apartments come online” (March 2026) — https://www.axios.com/local/raleigh/2026/03/18/raleigh-rents-fell-year-over-year-apartment-construction

Adam Burdick advises multifamily owners on operations, pricing, and retention at Goldshine Management Consulting