Picture two versions of the same project.
In the first, the traffic engineer draws a wide net — twelve intersections, three corridors, two weeks of data collection. Half the report ends up documenting locations where the project's impact rounds to zero. The client pays for it. The schedule absorbs it. Nobody's happy.
In the second, the engineer scopes tight — six intersections, clean and fast. Then, four months later, during agency review, someone flags an unsignalized intersection two blocks outside the boundary that turns out to have a documented crash history and a school walking route running straight through it. Now there's a supplemental study, a delay, and an awkward conversation with the client about why it wasn't caught the first time.
Neither engineer did anything careless. They just answered the same question — how big should this study area be? — without a repeatable way to answer it. That's the gap this post is meant to close. (If you're not yet sure your project even needs a study, start with what a traffic impact analysis actually is.)
The Real Question Isn't "How Far?" It's "How Much?"
The instinct is to think of a study area as a shape on a map: a half-mile radius, a few blocks in every direction. That instinct is exactly what gets engineers in trouble.
A study area isn't a distance. It's a threshold. The question isn't "how far from the site should we look?" — it's "at what point does this project's traffic stop mattering enough to analyze?"
That reframe changes everything about how you scope.
Let the Trips Draw the Map
Here's the process that actually holds up under agency scrutiny:
Estimate trip generation
Use ITE's Trip Generation Manual (or your local agency's preferred source) to figure out how many new trips the project will add during peak hours.
Distribute those trips onto the network
Where do they actually go? A retail project sends trips in different directions than an apartment complex or a warehouse.
Watch where the impact fades
As you move outward from the site, the project's share of an intersection's total volume shrinks. Most agencies draw the line somewhere around 5–10% of existing peak-hour volume — beyond that point, the project's fingerprint on that intersection is arguably too faint to matter.
Stop there
Not at a mile marker. Not at "the next big intersection." At the point where the math says the impact becomes noise.
This is why a 50,000-square-foot retail project on a congested arterial might need a study area covering eight intersections, while the same-sized project on a lightly traveled rural road might need only two. The land use is identical. The context isn't — and context is what actually drives the boundary.
The Rules of Thumb Are a Floor, Not a Ceiling
Percentage thresholds are the backbone of scoping, but they're not the whole skeleton. A few things will pull an intersection into your study area even when the math says it's borderline:
It's a site access point
No exceptions, no math required: if traffic enters or exits here, it's in scope.
It has a documented crash history
A marginal volume increase at a location with a known safety problem isn't marginal anymore.
It sits near a school, transit stop, or high-pedestrian corridor
Vehicle percentages don't capture the kid crossing the street.
It's already programmed for a capacity upgrade
The "before" and "after" conditions might both need separate treatment.
The agency says so
Local guidelines can override the general rule entirely — always check before you assume ITE's playbook applies as written.
Think of the percentage threshold as your starting draft, not your final answer.
Scope It Once, Talk About It Early
The single most effective thing an engineer can do to avoid the "surprise intersection" problem isn't a better formula — it's a five-minute conversation.
Most agencies either require or strongly welcome a scoping meeting before the TIA gets underway. Bring your preliminary boundary, your trip generation numbers, and your reasoning. Let the agency flag anything they'd expect to see included. It's a much cheaper conversation to have in week one than in month four.
A Boundary With No Reasoning Is Just a Guess
When you submit the TIA, don't just show the study area — show your work. Reviewers respond far better to:
"We included the Main St / 5th Ave intersection despite it falling below the 5% threshold because of its documented pedestrian activity near the elementary school."
…than to a map with a dotted line and no explanation. The boundary itself will get questioned eventually. The reasoning behind it is what actually survives review.
The Takeaway
A good study area isn't the biggest one, and it isn't the smallest one — it's the one you can explain. Trip generation and distribution give you the math. Local knowledge, agency input, and a bit of engineering judgment give you the rest. Get this step right, and everything downstream — network design, data collection, scenario analysis — has a clear, defensible target instead of a guess.
Scope it once. Scope it well. Everything after this gets easier.
Want to go deeper?
Not sure a study is even required yet? Read whether a TIA is actually required for your project in Texas — it covers the thresholds and triggers that decide whether you need one in the first place.