Lunchtime foot traffic near offices: the midday peak and who it feeds
How big the noon rush is around office towers, how long it lasts and which towers have it, read from people-hours, for food, coffee and service businesses.
Part of the guide Location intelligence automation
Lunchtime foot traffic near offices is one of the sharpest peaks a street can have: thousands of workers leave their desks within the same hour or two, and the streets around a tower can carry more presence between noon and two than at any other time of the day. Footfall data shows this directly. For each tower you can see the busiest hour, the average presence in each hour of the day, and the share of the whole day that falls in the midday block from 12:00 to 16:00.
This piece is for salad bars, coffee carts, lunch subscriptions, caterers, gyms with lunchtime classes and pharmacies near business districts. It shows how to tell a true lunch street from a busy one, a made-up comparison of two towers, the traps, and how an AI employee measures every tower on your list each month.
What a lunch street looks like in the figures
The Mobility Footfall block splits each day into four parts, each with its hours: morning from 06:00 to 12:00, midday from 12:00 to 16:00, evening from 16:00 to 21:00 and late night from 21:00 to 06:00. It gives the share of the day in each part and the average presence an hour inside it, so a short four-hour midday can be compared fairly with a nine-hour late night. It also lists every hour of the day with its average, in footfall_by_hour.
Around a working office cluster, three signs usually appear together:
- A peak hour at 12:00 or 13:00, sometimes 14:00 in cities that eat later.
- A high average an hour at midday compared with the evening, once you divide by the hours each part covers.
- Weekdays far busier than weekends, which shows when you run the block once for weekdays and once for weekends.
A street that is busy at midday on Saturdays too is probably a shopping or leisure street, not an office one. Both can be good for lunch trade, but they need different menus and hours.
Measuring the right patch
An office tower is tall, not wide. With the distance set at 200 m or less, the block reads the tower's own cell of the map, a hexagon of about 0.1 km2 that holds the building, its plaza and the nearest crossing. Set it wider and it adds every neighbouring cell whose middle falls inside the circle, which can pull in a station or a shopping street and lift the figure for reasons unrelated to that building. For lunch decisions, the tower's own cell usually reads best. Use the same distance for every place you compare.
A made-up comparison of two towers in Nairobi
These figures are invented to illustrate the method; they are not measurements. A salad brand with two kiosks in Nairobi wants to add a third and has a lease offer at two towers in the Upper Hill business district.
| Measure, weekdays only | Tower one | Tower two |
|---|---|---|
| Footfall a day | 22,600 | 26,800 |
| Busiest hour | 13:00 | 10:00 |
| Midday share of the day | 34% | 22% |
| Average an hour at midday | 1,920 | 1,470 |
| Average an hour in the morning | 1,310 | 1,850 |
Invented figures for illustration only.
Tower two is busier over the day, but its peak is mid-morning and its midday hours are quieter than its mornings, a pattern that fits a building where many people arrive late or attend morning meetings and eat in. Tower one is smaller yet clearly comes alive at lunchtime. For a salad kiosk that does most of its trade between noon and two, tower one is the better lease. A coffee cart might choose the opposite.
Traps when reading office lunch figures
- Using all days. Weekends dilute an office pattern. Choose Weekdays in the block's Days setting.
- Comparing midday with late night by share alone. Late night covers nine hours and midday four. Compare the average an hour.
- Forgetting the delay. Each day arrives about ten days later, and the data starts on 17 October 2026, so a public holiday or a closed building in that window will show.
- Counting presence as customers. Many workers bring lunch or eat in a staff canteen. Footfall says how many are around, not how many buy.
- Reading one tower as the district. Two buildings on the same street can have very different working hours.
An AI employee that ranks towers by their lunch rush
In xMap Flows, an AI employee can find the office buildings in a district, measure the lunchtime rush around each, and send you the ranking every month. This kind of recurring check is what location intelligence automation is for.
- Schedule: the first Monday of each month.
- Area of interest: the business district you serve.
- Nearby places: Corporate office, plus Coworking space if you want them too.
- Mobility Footfall: Each incoming place, 200 m (the building's own cell), Weekdays, Footfall and Busy hours, the last 30 days.
- Sort and limit: by footfall_midday_pct, highest first, keep 20.
- Email with the list and a map of the 20.
What you get: the 20 buildings where the most of the day happens at lunch, each with its busiest hour and its hour-by-hour line, ready for a visit. Caterers can start from catering leads: offices and venues near my kitchen, and lunch services from office clusters for a lunch subscription.
Questions
What time is the lunch peak around offices?
Usually 12:00 or 13:00, sometimes later in cities that eat late. The block gives each place's own busiest hour and every hour's average, so you do not have to guess.
Can I see only weekdays?
Yes. Mobility Footfall has a Days setting: All days, Weekdays or Weekends. Weekends follow the place's own calendar.
Does footfall count how many people buy lunch?
No. It measures people-hours of presence near the building, not purchases.
How close should I measure around a tower?
Keep the distance at 200 m or less, which reads the tower's own cell of the map, about 0.1 km2. Wider circles pull in neighbouring buildings and stations.