
Most rental property portfolios are benchmarked against a metro-level rent or price index. That's fine for a headline. It falls apart once you need to make a decision about a specific group of homes.
Two conversations with clients this year made that clear to us.
The problem: one metro, many markets
One conversation was with an asset manager at an institutional investor that owns a scattered-site SFR portfolio. Atlanta makes up more than a quarter of it. Sitting on the LP side, the asset manager wanted an independent way to check rent growth assumptions going into next year's business plan.
The asset manager's point: homes in the same MSA can behave very differently depending on where they are, their schools, household incomes and the age of the home. A single Atlanta number doesn't tell you what to expect for your homes. Her team wants to dig as deep as needed to get comfortable with an assumption, without rebuilding a data shop in-house.
We hear the same request from operators. They already buy metro-level indexes from the major research firms. What they want is something closer to apples-to-apples, along the lines of: how did 1980s three-bedroom homes on the west side of Tampa do?
Why it matters: indexes are slow, and the gap costs real money
The other conversation was with an asset management lead at a large institutional SFR operator, who described what that gap costs. Like many operators, his firm marks homes by taking a BPO at stabilization and then adjusting that value each quarter with a metro index. That can run for years before the home gets a fresh BPO.
The trouble is that indexes are slow to turn. In softening markets like Florida and Texas, book values ended up ahead of what homes would actually sell for. When it was time to sell, some investors were surprised by the haircut. His example was a cohort in San Antonio that needed a 25% discount.
The takeaway: a more granular, more accurate read on value, at a competitive cost, would set expectations up front and make disposition conversations much easier, especially for private equity investors with separate accounts at large operators.
What we did: a like-for-like test in Atlanta
To show what a segmented view looks like, we took about 7,000 closed institutional SFR leases in metro Atlanta. of which 3,534 fall in the two comparison windows. We grouped them into 10 clusters by:
Area: East, Northeast, South and West/Northwest, defined by school district
Size: 2–3 bedrooms vs. 4+
Year built: before 2000 vs. 2000 and later, where there was enough data to split
For each cluster we compared leases signed in the last six months (May–Nov 2026, 1,002 leases) with leases signed 12–18 months earlier (May–Nov 2025, 2,532 leases).
Comparing raw medians is misleading, because the homes that happen to lease in each period differ. In one cluster, recent leases were 170 sq ft smaller, which made rents look 4% lower when like-for-like they were flat. So within each cluster we controlled for square footage, year built, bedrooms, bathrooms and zip code. That isolates the true change in rent for comparable homes. We report each result with a 90% confidence range.
What we found
Atlanta institutional SFR rents were essentially flat, year=over-year, like-for-like. But "flat" hides real differences:
Three clusters moved clearly:
Northeast 4+ bed homes (Gwinnett, Barrow, Forsyth) rose 4.4% y-o-y.
Older 2–3 bed homes in West/Northwest (Paulding, Cobb, Douglas) rose 1.5%.
Older 2–3 bed homes in the South (Clayton, Henry, South Fulton) fell 3.0%.
That is a 7-point spread inside one metro, in a market a metro index would call flat. For an LP checking a business plan, or an operator marking a book, those are very different assumptions.

What to watch
Thin recent data. Recent lease volume runs about 150–200 a month, versus about 450 a year earlier, so smaller clusters have wider ranges.
Operator mix. Who is leasing matters. In the South pre-2000 cluster, one large operator's share of leases rose from 16% to 46%, so part of that −3.0% may reflect one landlord's pricing.
The bottom line
Metro indexes are a starting point, not an answer. Segmenting by area, size and vintage, and comparing like with like, shows where rents are actually moving, with an honest range around each number. That helps LPs pressure-test assumptions, operators set renewal and disposition expectations, and investors avoid surprises at sale.
Want to see this for your markets or your portfolio? Get in touch with Locate Alpha.
