Following the Movers: Using Migration Data to Time Market Entry

Every market-planning deck we see contains the same slide: a ranked table of metros with a population column and a growth-rate column. It is the most quoted number in the room and the least useful, because a population figure is a photograph of a market that has already happened. By the time a metro shows up on that list, the corridors worth owning inside it have been leased, and the rent you’ll pay reflects the growth you were hoping to capture.

Migration data is the same underlying phenomenon read as a moving picture instead of a still. It answers a different question: not how many people are here, but who is arriving, who is leaving, and where inside the metro they are landing. For a committee deciding which four markets get capital next year, that distinction is the difference between buying growth and paying for it.

A Market’s Size Is Not Its Direction

Stock is the count of people in a trade area today. Flow is the rate and composition of movement through it. Two corridors can report an identical 42,000 residents inside a ten-minute drive while one is quietly shedding households to the exurbs and the other is absorbing 3,000 net arrivals a year. On a demographic snapshot they are twins. As real estate decisions they are opposites, and the lease you sign will run seven to ten years — long enough that direction dominates level.

The practical fix is to stop treating population as a screening variable and start treating it as two variables. Level tells you whether a store can exist here at all: whether there is enough population, income, and daytime activity to clear your minimum. Flow tells you what the trade area will look like in year three, when your ramp curve is supposed to have matured. Underwriting only the first is how brands end up with a portfolio of stores that were correctly sized for a market that has since moved three miles north.

Read gross flows, not just net

Net migration is a subtraction, and subtractions hide things. A corridor with +1,200 net could be 1,400 in and 200 out — a stable, appreciating neighborhood — or 14,000 in and 12,800 out, which is a churning transitional market with completely different implications for brand loyalty, repeat frequency, and how fast your store reaches maturity. Ask for in-migration and out-migration separately. High-churn trade areas can still be excellent for convenience and daily-need categories and quietly hostile to anything that depends on a returning customer base.

The MSA Average Is the Enemy

Almost every migration statistic that reaches a boardroom is reported at the metro level, and almost every real estate decision is made at the corridor level. That mismatch is where most timing errors live. Growth does not distribute evenly across a metro; it collects along two or three arterials, usually where developable land, new school capacity, and a commutable path to employment intersect. Averaging that across a 3,000-square-mile MSA produces a number that describes no actual location.

The consequence is that a “fast-growing metro” is a claim about maybe fifteen percent of its geography. The rest is flat or declining, and it is where the available space usually is, because the growth corridors leased up first. Teams that enter on the metro number without resolving to corridor level reliably end up in the fifteen-percent market’s shadow: right city, wrong intersection, forecast built on someone else’s growth.

Working at corridor resolution means using drive-time geographies rather than county or MSA boundaries, which is the same argument we make in drive-time analysis vs. radius rings and in our guide to reading a new market remotely. If your migration inputs cannot be cut to a ten-minute drive polygon, they are context, not evidence.

A fast-growing metro is usually three fast-growing corridors and a lot of flat ground with better availability.

Rooftops, Then Retail: Where the Window Opens

Residential development leads retail demand, and the gap is measurable in the entitlement process itself. Permits are pulled, then construction runs, then units deliver, then households move in, then those households establish spending patterns. Retail rents, meanwhile, reprice on evidence — comparable leases, traffic counts, demonstrated sales — all of which are backward-looking. That asymmetry is the whole opportunity: for a period of roughly a year to two years, the rooftops are visibly coming but the rent still reflects the trade area as it was.

Enter too early and you carry a store through a ramp that hasn’t begun, burning cash while the households you underwrote are still in escrow elsewhere. Enter after repricing and you are paying a growth premium for growth you now have to share with the three competitors who arrived during the window. The discipline is not “be early” — it is knowing which stage of the pipeline you are looking at, and sizing your commitment to it. A corridor with framed buildings and a signed school bond is a different risk than one with a rezoning application.

What to watch, in order of lead time

• Rezonings and entitlements (24–48 months out). Directional only. Plenty die.
• Residential building permits and multifamily starts (12–24 months). The single most reliable early signal, and public in most jurisdictions.
• School enrollment by attendance zone (6–18 months). Confirms family households, not just units. A district adding portable classrooms is telling you something a census release won’t for two years.
• Utility connections and change-of-address filings (0–6 months). Near-real-time proof that occupancy is actually happening.
• Category search demand in the geography (continuous). Intent, updated monthly, and often the earliest evidence that arrivals are looking for what you sell.

Search Volume as a Migration Proxy

The frustrating thing about official migration data is its latency: the definitive numbers arrive well after the leasing decision had to be made. Search behavior does not have that problem. Because keyword volume can be read down to the city and region level — Semrush’s keyword research tools cover 26.7 billion keywords across 142 geographic databases, with search volume reported at city and region granularity— you can watch demand for your category form inside a specific corridor and track it quarter over quarter. It costs almost nothing and updates monthly.

Used well, this is a triangulation tool, not a forecast. If permits are up, enrollment is up, and searches for your category in that metro have climbed for four consecutive quarters while no competitor has opened, you have three independent signals agreeing. If searches are flat while construction booms, you may be watching a corridor filling with households who are not your customer — which is useful to learn before the LOI, not after.

The local-intent layer matters too. According to Semrush data, “near me” keyword variations total roughly 7.1 million US searches per month and grew 29% between Q1 2025 and Q1 2026, with immediacy-driven variants like “near me tonight” (+41%) and “near me open now” (+38%) rising fastest. In a corridor of recent arrivals, that behavior is amplified: new movers have no established habits, so they search for everything. We unpack the site-selection implications in near-me searches and site selection.

Who Is Moving Beats How Many

Thirty thousand arrivals is not a market. Thirty thousand arrivals whose median household income, age band, and household composition match your best-performing analog stores is a market. The two get conflated constantly, because headline migration numbers are counts and counts feel objective.

The composition questions that actually change a forecast: Are arrivals renters or buyers, and does your category index on one? Are they retirees relocating with assets but low daily trip frequency, or dual-income households with children and high visit frequency? Are they moving from higher-cost metros — arriving with price expectations that make your premium tier viable — or from lower-cost ones? Is the corridor gaining households while losing population per household, which reshapes basket size even as the count rises? Our work on demographic insights in site selection and on the evolution of the retail demographic goes deeper on translating those attributes into a sales model rather than a persona.

The Boomtown Everyone Already Found

The most expensive mistake in migration-led expansion is not missing a growth market. It is entering the one that has been on every conference slide for three years. Headline boomtowns attract capital the same way they attract residents, and by the time a metro is a consensus growth story, the growth is spoken for: the A corridors are leased, rents have repriced, and your category may already have four operators splitting a demand pool that looks large in aggregate and thin per store.

Migration tells you the size of the prize. It cannot tell you your share of it. That requires putting arrivals against competitive supply and remaining headroom — the capacity question we work through in how many stores a market can hold and in competitor analysis for expansion. A market adding 20,000 residents a year with no direct competitor frequently outperforms one adding 60,000 with eight, and only the ratio, never the numerator, tells you which.

This is also where an unglamorous secondary market earns a second look. Slower growth, but growth you own, a rent basis that isn’t pricing 2029, and a first-mover position that compounds. Those markets rarely win the ranked-table slide, which is precisely why they are still available.

Turning Flow Into a Sequenced Plan

Migration data does not produce a market list. It produces a sequence: which corridors are ready now, which get revisited in two quarters when permits convert to occupancy, and which are consensus plays to leave alone. That sequencing is the substance of a real retail market expansion strategy, and it only works if the migration read, the revenue forecast, and the leasing execution move together — a corridor identified eighteen months early is worth nothing if it takes fourteen months to get a deal done in it.

That’s the reason Locate keeps forecasting and brokerage under one roof: the analysis that finds the window and the execution that captures it are the same job. If you’re building next year’s market plan and want a corridor-level read on where the movers are actually landing, get in touch.

Common Questions

How do you use migration data for retail site selection?

Read it as flow rather than stock. Instead of asking how many people live in a market, ask how many arrived last year, how many left, where they came from, and what they look like on income, age, and household composition. Then drop below the MSA line: net migration for a metro is an average that hides which three or four corridors are absorbing almost all of the arrivals. Those corridors, not the metro, are the unit of decision.

What are the leading indicators of a growth market for retail?

Residential building permits and multifamily deliveries come first, typically 12 to 24 months ahead of occupancy. School enrollment by attendance zone confirms family households are actually arriving, not just units being built. Utility hookups and change-of-address filings give you near-real-time movement. And category-level search demand in that geography shows intent forming before census products catch up. No single one is sufficient; three pointing the same direction is a signal.

Is population growth enough to justify entering a market?

No. Growth tells you the market is getting bigger; it says nothing about how much of that growth is already claimed. A metro adding 60,000 residents a year with eight competitors racing in can be a worse bet than a market adding 20,000 with none. Migration data sets the numerator. Competitive supply and remaining capacity set the denominator, and only the ratio is a decision.

How can search volume data help evaluate a growth market?

Search volume is reported down to the city and region level, which makes it a cheap, fast proxy for demand forming inside a specific geography. If queries for your category in a corridor are climbing quarter over quarter while your competitors haven't opened there, you are seeing demand that census releases and traffic-count studies will not confirm for another year or more. It's directional, not a forecast, but it is one of the few signals that updates monthly.

How early should you enter a growth corridor?

Early enough that rents haven't repriced, late enough that the rooftops actually exist. In practice that usually means signing when residential delivery is underway and visible but retail rents still reflect the pre-growth trade area, roughly the window between permits pulled and certificates of occupancy issued. Entering before the rooftops means carrying a store through a ramp that hasn't started; entering after repricing means paying for growth you didn't underwrite.