FlowsBook a callSign inCreate free accountSign up
Roads and traffic

Traffic congestion monitoring around the places you care about

How to watch live speed, free-flow speed and the jam factor around chosen places on a timetable, when an alert is worth sending, and what this kind of data cannot tell you.

Part of the guide Location intelligence automation

By the xMap team · · 5 min read

Traffic congestion monitoring for a business means measuring, at fixed times, how fast traffic moves on the roads around specific places compared with how fast it moves when those roads are empty, and acting when the gap passes a line you set. You do not need a wall of screens. Three numbers per place, read on a timetable and kept over weeks, tell you which locations jam, when, and whether it is getting worse.

This is for logistics and service managers, retail and property teams, venue operators and city departments. It explains the three numbers, how to choose places and hours, when an alert earns its place, what the data does not give, and how to run it with an AI employee.

The three numbers that matter

Live road data in xMap Flows comes from the Traffic Flow block. For each place it reads the roads nearby at the moment the flow runs and adds these fields.

FieldWhat it meansHow to use it
Speed (km/h)How fast traffic moves now, averaged over the roads, weighted by the length of each stretchCompare it with free flow, not with the speed limit
Free-flow speed (km/h)How fast traffic moves on those same roads when they are clearYour baseline for each place
Jam factorA scale from 0 (free) to 10 (standstill or closed)The simplest number to alert on
Congestion (%)How much slower than free flow, never below 0Easy to explain to non-specialists
Traffic levelThe jam factor in words: Free flowing, Light, Moderate, Heavy or StandstillGood for a message headline

Fields Traffic Flow adds to each place. It also says how many road stretches it measured, how many are closed, the busiest road, a confidence from 0 to 1 and when the reading was taken.

A road with a free-flow speed of 60 km/h and a current speed of 30 km/h is 50% congested. The same 30 km/h on a road whose free flow is 35 km/h is barely congested at all. That is why the comparison with free flow matters more than the raw speed.

Picking places and hours

Start with the places where slow roads cost you something: depots and warehouses, stores with car parks, sites with timed deliveries, venues on event days, hospitals, schools. For each, choose how far to look. The roads next to the place (about 150 m) suit a shop front; every road within 500 m or 1 km suits a depot with several exits.

Then choose hours. Because the readings are live, a flow that runs at 08:00 tells you about 08:00 only. Pick the hours that matter to your operation, such as the first departure, the lunchtime rush, school pick-up and the evening return, and run at each one. Places within 300 m of each other share one reading, and one run measures up to 50 places, which is enough for most networks.

When to send an alert

Most readings should go into a record, not a message. Alerts work when they are rare and tied to a decision.

  • A threshold with a reason. Jam factor 6 or above at a depot during loading hours, because that is when departures start to slip.
  • A closed road. When roads_closed is more than zero at a site, say which road is busiest and suggest checking for a closure.
  • A change from normal. After a few weeks of readings you know what a typical Tuesday at 17:00 looks like. Alert when a reading is far outside it, not every time it is merely busy.
  • Low confidence held back. A reading with a low confidence figure is measured from few vehicles. Wait for the next one before raising the alarm.

What it does not give you

It is worth being clear, because people often expect more from live congestion data than it holds.

  • No car counts. Speed and the jam factor describe how traffic moves, not how much of it there is. A free-flowing road may carry many cars; a jammed one may carry few that are stuck.
  • No history on its own. Each reading is a snapshot. To see patterns, keep the readings yourself, in a sheet or a database, run after run.
  • No direction split or peak-hour figure. Run it at several hours to find the peak; look at the busiest road to see where the trouble is.
  • No reason. A slow road could be roadworks, a crash, rain or simply the rush hour. Pair it with Traffic incidents when you need the cause.

A made-up example: a furniture retailer in Toronto

Invented for illustration. A furniture retailer has six showrooms and a distribution centre around Toronto. Customers collect large items from the showrooms, and the retailer suspects two sites are hard to reach after work.

It runs Traffic Flow on weekdays at 08:00, 12:30 and 17:30, with every road within 500 m, and writes each reading to a sheet. After four weeks it averages the 17:30 readings. Four showrooms sit between Light and Moderate. Two sit at Heavy most evenings, with congestion above 50% and the same busiest road each time. The retailer moves its evening collection slots at those two sites to a later hour, and sends an alert to the distribution centre only when its loading-hour jam factor passes 6.

Running it with an AI employee

This is one of the simpler jobs to hand to an AI employee, and a good first step into location intelligence automation.

  1. Schedule: several times a day on weekdays, at the hours you picked.
  2. My data: your sites, from a table.
  3. Traffic Flow: the roads to measure around each.
  4. Branch: Jam factor at or above your line goes one way, the rest the other.
  5. On the Yes side, a Slack message or Email naming the site, its level and its busiest road. On both sides, a row in Google Sheets so the record keeps growing.

The template Road speeds outside my depots does this every hour; Road speeds around the stadiums applies it to venues. Each day the AI employee's Team inbox thread holds a note of how many readings it took and which places crossed the line.

Questions

Can I see how many cars use a road?

Not from live congestion data, which measures speed and how jammed the roads are. Hourly car counts since 2021 exist only for roads in the United Arab Emirates, in the Road traffic history dataset.

How often should I take readings?

As often as your decisions change. Many teams use three to five fixed hours a day; a depot planning departures might check every 30 minutes for a few morning hours.

What jam factor should trigger an alert?

There is no single right number. Start around 6 for places where delays cost money, then adjust after a few weeks of readings show what normal looks like at each site.

Does it keep past readings?

Each reading is live. Send every run to Google Sheets or your own database if you want a history to compare against.

Templates that do this

Read next

More on roads and traffic · Every article · Templates

Put it to work

An AI employee in xMap Flows does this on live data, on its own timetable. Start from a template or describe it in plain words.