Is ZIP Code Data Enough to Choose a Rental Property? Same Street, Same ZIP, 17 Points Apart

Whenever I'm on a call with a new client, they usually already have an idea of the state and/or city they're interested in investing in. Sometimes they've even narrowed it down to a ZIP code. These are always the first filters everybody reaches for, and they're useful to some degree, but in most affordable housing markets they're the weakest one you can use because the devil truly is in the detail, and it's detail almost nobody gets to.

So when you're choosing a rental property, which tells you more: the ZIP code or the neighborhood?

Here's what that looks like with two real houses we analyzed recently for our own inventory. A contractor I buy from offered me both of these properties in the same week. Same street, same ZIP, same school zone, same market page on Redfin. I ran my full neighborhood analysis model on each so you can see the difference for yourself.

Before I put these two side by side, I should briefly explain how I grade neighborhoods. While lots of investors grade A, B, C and D, I prefer to look at phases.

Phase 1: This is a neighborhood showing very poor data and no tangible signs of a positive trajectory. You'll find cheap houses here, and they might look great on paper, but you don't want to own property here unless you're looking for a problematic tenant base, high operational costs, and a lack of appreciation.

Phase 2: This is my sweet spot. It's showing positive improvement on most data points, but there's still room for further improvement and value growth. The area is on a positive trajectory, but there's still some risk here. We need some level of risk in order to find the reward.

Phase 3: This is where all or most of the improvement has already happened. The neighborhood and its residents are stable, but there's little room for improvement or value growth other than slow, long-term appreciation. This can be a great place to park money long-term, but it isn't where you'll find the kind of cash flow we need as rental property investors.

With all that in mind, let's get into it.

ZIP code vs neighborhood: what does a ZIP-level search show?

The same thing for both. That's the problem.

I'll call them House A, on the 5100 block of the street, and House B, on the 6100 block. Both are 1925 frame houses, renovated, in a C-class part of Kansas City's east side. I'm not naming the house numbers, because one is a contractor's house and one is in our inventory, and neither owner needs their address in a blog post.

Here is what an out-of-state or foreign buyer would find in twenty minutes online.

The jobs figures differ by a point because the two circles are drawn from different spots. Both are led by health care. On every line, the two houses look like the same purchase.

The rent limit deserves one note. HUD's figure is a ceiling on what a voucher can pay, and it is not your rent. It's the same for every house in the ZIP, which is exactly why it can't tell these two apart.

What does the tract and quarter-mile data show?

When you stop looking at the ZIP and focus in on the much smaller and more relevant census tract data, a very different picture starts to emerge.

House A sits in census tract 77. House B sits in tract 79. A census tract is a small area of about 1,200 to 8,000 people that the Census Bureau uses to publish local data. For crime I use live police department reports and go smaller still, to a quarter-mile circle around the house, which is the 5 to 10 minute walk a tenant actually lives with.

There's one point of context it's important to remember when you read this. Every tract figure is a Census five-year estimate, 2015 to 2019 against 2020 to 2024, and the margins of error at this scale are wide.

And one more thing to be plain about before going further. Two houses isn't a dataset. My point here is that a single ZIP can hide a split this big. While that is a very hard lesson I learned early on in my investing journey, this example doesn't tell you how often that happens. My experience? In these cheaper secondary markets where we tend to invest, it happens a lot. The housing markets here should be measured street by street.

Why is crime going two different ways a mile apart?

I don't know why, and I'd be guessing if I told you. What I can show you is that it is.

Both houses sit in the same police patrol area. Here are all reports within a quarter mile of each, from Kansas City Police Department open data, each report counted once.

Like most of these data, I grade crime on direction, not level. Falling crime rates drive demand for rentals and home buyers, and that fuels future value and rent growth. The level around both houses is still above the city average, and I say that to every client. But over a ten-year hold, which way it's moving matters more than where it is today.

So, I compare 2022 with this year's pace. For House A, reported crimes were down 16% against 2022. This year on its own it's down 30%, against a 10% fall across the city. Three times the city's improvement, within a short walk of the front door. That's a neighborhood very clearly getting safer.

For House B, the same sum tells us reported crime is up 11% against 2022. This year it's down 8%, which is roughly the city's own fall. In other words, the area around House B is going sideways while the city improves.

I'm comfortable with the direction at House A. I'm not comfortable with it at House B. That said, none of this replaces my property manager's opinion on the street, and a drive-by of the block. The data and the legwork work together.

Why does the better area have the worse eviction number?

This is the finding I'd want you to take away, because it's the one that catches careful buyers.

I look at evictions at the census tract level. House B's tract files evictions at 0.87 times the Jackson County rate this year. House A's tract files at 1.85 times. On that number alone, you'd assume House B looks a safer bet in terms of eviction rates. But again, I weight the trajectory higher than the raw snapshot of today.

As you can see, evictions in House A's tract peaked in 2024. Since that peak, filings are down 41%, over the same years that poverty in the tract fell from 26% to 18%. The area was in trouble, and it's coming out of it. That is a clear sign the tenant base is improving. Problematic tenants are being replaced by more of those who pay and stay, which is what we want as a landlord.

House B's tract tracks the county eviction rate almost exactly, year after year. Its low rate rests on long tenancies and a large Section 8 voucher base, and a voucher tenant is a safety net, not a rent upside. Section 8 tenants tend to stay for longer periods of time, and leave of their own volition rather than get evicted, which usually means tenants with fewer housing choices. That's a different kind of area from A, not a safer one, and the Census data in the next section says which kind.

Two things to keep straight when you read eviction numbers. A filing isn't an eviction. Nationally about 4 in 10 filings end with the tenant leaving, and Kansas City tenants have had a right to a lawyer since 2022, which slows every case down.

And a high filing rate is a screening instruction, not a verdict. Your best tool for reducing eviction risk is and always has been careful tenant screening. In a tract filing at nearly twice the county, I'd verify income at three times the rent with pay stubs and bank statements, the same way I'd screen any applicant, and I'd want a manager who starts the process the day rent is late.

What do incomes, owners and empty homes say about direction?

They say House A's tract is recovering and improving, and House B's isn't.

House A's tract is improving on every measure. Incomes are rising after adjusting for inflation. In today's dollars the median went from $30,437 to $41,859, up $11,422 or 37.5%.

Poverty fell by a third in five years, and from 47.5% a decade ago.

The number of empty homes has halved.

The number of homeowners went up, and most of them have been there for decades. People tend to stay.

House B's tract reads the other way on every line that matters.

Owners down from 40% to a third. Income fell to $25,491, which is 63% below the city's median. The share of residents living in poverty almost doubled to 42.2%. And crucially, the number of folk who own their own home fell considerably. That tells me homeowners are moving out and renters are taking their place, and the tract's voucher base is larger than House A's. That's not a good sign for future growth.

One more thing I'd rather show than skip. Census median rent in House A's tract rose 14% between the two periods, from $736 to $838. Inflation over the same stretch was about 23%, so in real terms the tract's median rent fell. I still grade A's rents as improving, because that median includes small apartments and the houses tell a different story.

Three-bed houses within 1.5 miles ask $1,350, four-beds $1,550, and the rent paid on voucher homes in the tract rose 53% between 2021 and 2025, about 11% a year. This is a great example of why it's important to dig down below the readily available headlines and get some context.

Why does the number and trajectory of home ownership matter so much to me? Because turnover is the biggest running cost on a rental, and a block where homeowners stay is a block where tenants stay too (at least in my experience).

Where does the 96-point gap come from?

Mostly from crime and the people data. Here is my full scorecard.

I grade eleven parts of an area. Each gets a level from 1 to 5 for where it stands, and a trend of up, steady or down. The two combine and get weighted, and the total out of 552 becomes a score out of 100. Phase 2 (where I see the best balance of risk and reward) runs from 50 to 69.

The gap is 96 points. Crime is 45 of it. Demographics and income are 30. Owner-occupancy and rents are 12 each, and House B picks up 3 on investment because a 240-home development is 0.4 miles from it. So 75 of the 96 points come from two segments, and both of them are about the people on the street and what happens to them.

Notice where the two houses tie. Sales, employment, schools, housing stock and vouchers all score the same. That's the ZIP-level data again, and it's five of eleven segments. If I'd stopped there as most remote investors do when doing their own research, the two houses would look like the same investment. That's the core message I'd like to deliver with this article. These markets really are street by street. There are diamonds, but there is also a lot of rough.

Why do C-class areas change block by block?

Because the recovery of an area like this one doesn't arrive evenly. It arrives in pockets, and the pockets have edges.

Picking the right city gets you to a metro where the numbers can work. It doesn't pick the street, and the street decides the outcome. I've written before about the A to D neighborhood framework I use.

B-class areas are around the city median, few empty homes, a strong base of owners.

C-class areas have more empty homes and more upside, and they are street by street. There are good streets and bad streets inside the same name on the map.

You'll find better cash flow in a C area, but with more risk. A B area will give you stability, but it might not deliver the cash flow you need to sustain and maintain the asset and debt service.

What I look for in a C-class area is a mix. Some renovated houses selling at $175,000 to $220,000. Some average houses at $120,000 to $150,000. Maybe one or two vacant distressed properties.

That mix tells me investment is happening but isn't finished, which means there's upside left. A street with nothing but $40,000 houses and no buyers is a different thing, and renovating one to a high standard doesn't change what it is.

I learned that the expensive way. By 2022 I'd scaled to 124 properties by buying the cheapest houses I could find and renovating them. Too many of them were on the wrong streets. My portfolio collapsed in 2023, and the thirty I own today are the ones that were on the right ones. I rebuilt my whole buy box around that lesson, and this scoring method is what came out of it.

The two Walrond houses in this article show the edge of a pocket. Public money is going into the corridor near both of them: a 12-acre health and housing campus about a mile from A, a 240-home development 0.4 miles from B, and a bus rapid transit line to downtown half a mile west. B is closer to all of it and still scores 17 points lower. That's why investment carries only 3 of 92 points in my method. A development breaking ground tells you what might happen. Owner-occupancy, incomes and the crime trend tell you what is happening, and what you're likely to experience as a landlord.

There's a second reason this matters for a buyer from overseas. If you're buying renovated turnkey properties, and from 4,000 miles away you probably are, the house's condition is off the table. You're not buying a project. So what you're really buying is the street, and the only question that matters is where the street is heading.

What would the listing have told you?

Nothing that separates them. I mean that literally.

A listing carries the ZIP, the schools, the price history and a map. For these two houses every one of those is the same. The photos would both show a renovated 1925 house. The checks you can run on a seller's numbers would come back similar, because the numbers they test are about the house and the ZIP.

What my report sees that a listing doesn't:

Everything in that table is public. You could pull all of it yourself with a laptop and an afternoon, and I'd encourage it. What takes longer is knowing which geography to use for each one, and that's the part this article is about.

There is one document a buyer does get that helps. The neighborhood section of the appraisal records the owner-to-tenant split and the price trend as the appraiser saw them, and it's the only piece of third-party local data most buyers ever hold. It's a start. It isn't a trend.

And public data runs a year or two behind the street. That's why every report I run ends with questions for a property manager who works in that area, and why how I vet a manager includes asking which streets they'd avoid. On House A, that field report is still open, and it's a condition of the purchase.

This is the report that sits behind every house we offer. It's the first thing I run, before the price, before the lease, before the loan, and sourcing the right house in the right area starts with it.

What is the single-metric trap?

Picking one number and letting it decide. Each of the three a buyer reaches for first would have misled you here.

The school rating. Both houses feed a high school rated 1 out of 10. If that's your filter, you'd have passed on both, and you'd have passed on the better part of Kansas City's east side with them. Schools score the same for A and B. They don't separate the two.

The ZIP price trend. Up 7.7% in a year, with price per square foot up 19%. That's a good number, and it covers both houses equally. It would have told you to buy either.

The eviction rate. This is the one that bites, because it looks like exactly the sort of careful, local check I'm recommending. On the latest number, House B's tract is the safer one. On the trend, House A's tract is the one coming right. One figure, no history, wrong house.

The fix isn't a better single number. It's a level and a direction for every segment, and a score that forces you to look at all of them before you look at any one.

What did I do with each house?

I added House A to our inventory, with conditions, and I passed on House B. I should say plainly that I'm paid an advisory fee by the contractor if a client buys House A, and nothing at all on House B. Weigh that against everything above.

I rated House A as a "buy with conditions", and the conditions are about the block rather than the tract. The city has a condemned house at the demolition stage about a fifth of a mile away, and in August it referred a property across the street for clean-up. My checklist still has a drive-by, the property manager's field report and the status of those two cases on it. The tract is clearly recovering. The street is a little behind it, and I want to see it before a client commits.

House B is in a Phase 1 neighborhood, scored at 41.1, with no clear sign of lasting improvement, and I don't buy Phase 1. That's not a comment on the house, which is renovated and tenanted, or on the contractor, who offered me both in good faith. It's a comment on a quarter mile of data. There are deals I turn down, and this is what one looks like.

If you take one practice from this, make it the geography. Grade every part of an area on the smallest patch with reliable data: a quarter mile for crime, the tract for incomes, owners and evictions, listings within a mile or so for rents. Use the ZIP for context and never for the verdict. The foreign investor starter kit has the checklist I use, and it's free.

The bottom line

The ZIP code told me these were the same house. The street told me they weren't.

I'd rather own the house in the area that's recovering and heading somewhere positive than the one in the area that's standing still, even when the recovering one has the scarier eviction number today. Direction is what you own over ten years. A snapshot is what you own for a month.

Two houses on one street don't prove this happens everywhere. I think it happens more than most buyers would guess, and I know it happened here. I've been wrong about streets before, which is the whole reason I grade them this way now.

At the end of the day, investing is a game of probabilities. Looking at a tract instead of a ZIP doesn't make an area safe. You can still have bad tenants, and other problems. In fact, if you own rental properties long enough, those things will eventually come around and you'll have to deal with them when they do. What this analysis does is improve your probability of a smoother ride and a positive long term outcome, and make it much harder to buy the wrong house for a right-looking reason.