Method · accessibility intelligence · CONTRACT C
How the drive-time figure is produced, and what it does not claim
Drive time is the most-used and most-abused figure in site analysis, usually because two very different things are both called a "trade area". This page documents what LocateRight actually computes, how sampled origins are handled, and — deliberately prominently — what the result cannot establish. The full computation set is on the methodology page; the underlying data blocks are on data sources.
How the measurement is produced
- Define the study area and the target. The study area is a centre, a
latitude and longitude, and a radius. The destination is the business target, preferably
identified by Google
place_idso the routing target is an unambiguous place rather than a bare coordinate. - Sample origins across the study area. Origins are sampled across the study area rather than routing every address in it. The default probe uses 36 sample points at a 1500 m radius, and the number of points actually used is reported on the result.
- Route each origin to the target. Each sampled origin is routed to the
business target in driving mode, producing a travel time in minutes and a distance in
kilometres per point. Traffic awareness is available as an option and is reported as
traffic_awareon the result, so it is always visible whether the numbers you are reading considered it. - Exclude unroutable points, and count them. Points the routing service cannot resolve are excluded from the statistics and counted separately. This matters: a point that fails to route is not a point that takes zero minutes.
- Report the bands with the coverage statement attached. The result carries average, fastest and longest travel times, the share of routable points within 10, 15 and 20 minutes, a histogram of the distribution, the actual number of routing requests made, and the note that these are sampled points rather than the full population.
The percentage bands are explicitly shares of the routable points. If a fifth of the sample cannot be routed, the bands describe the four fifths that could — and the coverage note is there so that reading is the correct one.
Radius and drive time answer different questions
Both define a "trade area". Neither is wrong; using them interchangeably is.
| Straight-line radius | Drive time | |
|---|---|---|
| Measures | Geometric distance | Road-network travel time |
| Obstacles respected | None | Rivers, rail, one-ways, motorway junctions |
| Cost | Low — geometry only | Higher — routing requests per probe |
| Failure mode | Includes unreachable areas, excludes near-but-blocked ones | Small sample; sensitive to traffic time of day |
| Best for | Fast screening, dense urban grids | Realistic catchment where network shape matters |
| Typical error | Systematic overstatement of catchment | Sampling error, and traffic variation |
In most real settings the two disagree substantially, and the disagreement is not noise — it is the shape of the road network. Where a river, a rail line or a motorway junction sits between the site and the population, drive time is the more honest measure and radius is not. See trade area analysis for how the two are combined in a full analysis.
What this result does not establish
| Question | Answerable? | Why |
|---|---|---|
| How much of the study area is within 15 minutes by car? | Yes, for the sampled points | Directly measured |
| How many people live inside that catchment? | Not from this | Requires the population block — see WorldPop |
| How many of them are my customers? | No | Nothing here encodes intent or category affinity |
| What are travel times at 8am on a Tuesday? | Not reliably | Traffic-aware figures are a snapshot, not a time profile |
| Can a customer legally and comfortably walk or transit there? | No | Driving mode only; walking, cycling and transit are different networks |
| Is the catchment exactly what these bands show? | No | It is sampled points, and the result says so on its face |
Drive-time accessibility: FAQ
How does LocateRight measure drive-time reachability?
It samples a set of origins across the study area, routes each one to the business target in driving mode, and reports travel-time statistics over the points that routed successfully. The default probe uses 36 sample points across a 1500 metre radius, and the number of points actually used is reported with the result.
What happens to locations that cannot be routed?
Unroutable points are excluded from the statistics and counted separately, rather than being treated as zero minutes or silently dropped. The percentage bands are shares of the routable points only, so a result is never inflated by points that failed to resolve.
Does this tell me how many customers I will get?
No. It tells you how much of the sampled study area falls within given driving times of your target location. Population inside that area, and whether those people are your customers, are separate questions answered by other data blocks. The result is reported as sampled points, not the full population.
Is a straight-line radius different from drive time?
Yes, and the difference is large in most real settings. A radius is a geometric circle that ignores rivers, rail lines, one-way systems and motorway junctions, so it routinely includes places that are unreachable in the stated time and excludes places just beyond the line that take longer. Both methods are legitimate; they answer different questions, and the trade-off between them is covered on the trade area analysis page.
Is traffic considered?
Traffic awareness is an option on the probe and is reported as traffic_aware on the result, so it is always visible whether the figures you are reading were produced with traffic considered or not. Peak-hour and off-peak travel times can differ substantially.
Measure a catchment and read its coverage
Run an analysis and get the drive-time bands with points used, unroutable count and coverage statement attached — never a score.