UAE petrol station traffic counts: cars by hour on the road past a forecourt
How to read hourly car counts for the road outside a fuel station in the Emirates, from the main road to every road nearby, and what they can and cannot tell you.
Part of the guide Location intelligence automation
UAE petrol station traffic counts are available hour by hour, for both directions, on the roads within 500 m of any forecourt, going back to January 2021. For each station you get cars per day, the busiest hour, the busiest and quietest day of the week, the average speed, and a trend. That is enough to compare sites in a network, to staff the shop for the real peak, and to check whether the road a station depends on is growing or fading.
This is written for fuel retailers, forecourt shop and food operators, car wash and quick-lube businesses, and anyone valuing a roadside plot in the Emirates. It explains which road is being counted, how to read the direction split, a made-up example, and the weekly job an AI employee can take over.
Where the counts come from in a flow
The figures are a dataset inside the Regional data sources block: choose the United Arab Emirates, then Road traffic history. It is separate from Traffic Flow, which gives live speeds anywhere in the world but never counts cars. Hourly counts like these exist only for roads in the UAE, so the method in this article does not transfer to a station in another country.
You choose a period (the last 7, 30 or 90 days, the last 12 months, everything since January 2021, or your own dates), which days to include (all, weekdays, or weekends, meaning Saturday and Sunday), and what to get back: one row of totals per station, or a row per month, per day or per hour.
Which road is being counted
A forecourt is rarely beside just one road. There is the highway or arterial it serves, a service road that actually leads into it, and often a side street behind. The dataset looks for the nearest road of each of five road classes within 500 m and then does one of two things, depending on the Roads setting.
- The main road (the default). Each station gets the figures of one road: a named road in preference to an unnamed one, and the busiest of those. The row also says how many roads were found and lists them briefly with their class, distance and cars per day.
- Every road nearby. One row per road, so you can see the highway, the service road and the back street separately.
For fuel sites, start with every road nearby at least once. The main road is the right headline, but you want to know how far away it is. A count of sixty thousand cars a day on a highway 400 m off, reached only by a long slip road, is a different proposition from the same count on the road your entrance opens onto. The row gives the distance to the road in metres; read it before the count.
Direction is half the story
Counts are for both directions together, with a split by direction alongside. A station on a divided road serves one side. If the road carries a heavy morning flow towards the city on the far carriageway and a lighter one on yours, the headline flatters you. Ask for hour-by-hour rows over 30 days and look at the split at 07:00 and at 18:00. A station on the homeward side of a commuter road often does its best trade in the evening, which also changes when the shop should be fully staffed and when the car wash queue forms.
A made-up network comparison
These figures are invented to show how the reading goes. A retailer has three stations and wants to know which one justifies a larger shop.
| Station | Main road, cars a day | Distance to road | Peak hour | Weekend vs weekday | Trend |
|---|---|---|---|---|---|
| A, highway services | 82,000 | 60 m | 17:00 | weekend 15% lower | steady |
| B, arterial in a suburb | 34,000 | 25 m | 08:00 | about the same | rising |
| C, near a mall | 51,000 | 380 m | 20:00 | weekend 30% higher | steady |
Invented figures for illustration only.
Station A has the volume and sits on its road. Station C looks strong until you see the distance: its main road is the boulevard on the far side of the mall, and the road at its own entrance, found by switching to every road nearby, carries a fraction of that. Station B is the smallest today but is the only one whose road is growing, and its traffic is steady seven days a week. The retailer enlarges the shop at A, plans B for next year, and treats C as a mall-visitor site whose evenings and weekends matter more than its total.
What a count does not tell you
- How many cars turn in. Passing traffic is the pool, not the catch. Capture depends on access, visibility, price and what else is on the road. Your own till and pump data supplies that side.
- What kind of vehicle. The figure is vehicles counted; it does not separate trucks from cars, so it cannot split diesel demand from petrol.
- Today. The history ends a few days back. For the state of the road right now, use Traffic Flow for speed and Traffic incidents for closures.
- Anything across the border. A point outside the Emirates returns a note saying it is not covered.
Mistakes to avoid
- Quoting the busiest road within 500 m as if it were the road at the entrance.
- Comparing a month that includes Ramadan or a long public holiday at one station with an ordinary month at another. Use identical dates for all sites.
- Reading a low average speed at the peak as bad news. Slow traffic past a visible forecourt can help capture; standstill traffic that blocks the entrance does not. Look at the slowest hour beside the busiest.
- Judging a trend on the short one alone. The built-in trend compares the latest 30 days with the 30 before. For growth since 2021, ask for month-by-month rows.
The job an AI employee can do every month
Put your stations, and any plots you are considering, in a My data table with their coordinates. The flow is short:
- Schedule: the first working day of the month.
- My data: the stations and candidate plots.
- Regional data sources: Road traffic history, last 30 days, totals for each place, the main road.
- Calculated field: optional, for example cars per day divided by your own pumps per site if that is a column in the table.
- Sort and limit: by cars per day, or by the trend.
- Make a report or Email: the ranked table, with an xMap LLM summary of which roads rose and fell.
What lands in the inbox is a one-page ranking of the network with the peak hour for each site, and a line calling out any station whose road has dropped. To watch the competition on the same roads, a second flow with Nearby places and Only new reports each newly listed fuel station near yours. Together they are a small, practical piece of location intelligence automation for a forecourt business.
Questions
How far back do car counts for UAE roads go?
To January 2021, hour by hour, up to the last few days. You can ask for the whole period month by month to see how a road has grown.
Are the counts for one direction or both?
The main figure is both directions together, and each row also carries the split by direction, which matters for a station that serves only one carriageway.
Can I get the same counts for a petrol station outside the UAE?
No. Hourly car counts exist only for roads in the United Arab Emirates. Elsewhere, Traffic Flow gives live speed and congestion but not the number of cars.
What if the nearest road is an unnamed service lane?
The dataset prefers a named road within 500 m as the main road, and lists the others. Choose every road nearby to see the service lane's own figures.
Does a higher traffic count mean more fuel sold?
Not necessarily. It means more passing vehicles. How many stop depends on access, side of the road, visibility and competition, which you judge with your own sales data.