How to Forecast How Crowded a Venue Will Be at a Given Hour

Quick answer

Short answer: use a venue busyness forecast. It learns a venue's usual pattern for every hour of the week from anonymized, aggregated phone location data, combined with other real-world signals such as weather, events and holidays. With the BestTime API you send a venue's name and address once, get a 0-100 forecast for all 168 hours of the week, then read the day and hour you care about.

A value of 100 is the venue's own busiest hour of a normal week. It is not a share of capacity and not a head count.

Forecast a venue for a given hour, step by step

1. Identify the venue. A name and address are enough, for example "Colosseum, Piazza del Colosseo, Rome".
2. Create the forecast. You get busyness for every hour of every day of the week, plus a venue ID to use in later requests.
3. Pick the day and hour. Days run from 6 AM to 5 AM the next morning, so 1 AM on a Saturday night counts as Friday night.
4. Read the value. 0-100, relative to the venue's busiest hour, with a label such as below average, average or above average.
5. Plan around it. Pick the quietest or busiest hour that fits, or compare the same hour across days.

What goes into a busyness forecast

The output is busyness per venue and hour. The API does not expose personal identifiers, device IDs, raw location traces or visitor counts.

Why the scale is 0-100, relative to the venue's own peak

Each venue is measured against itself. 100 is the busiest hour of that venue's normal week, and 50 means about half as busy as that peak. 0 means the venue is closed or nearly empty.

• Not capacity. 100 does not mean the venue is full. It means this is as busy as the venue usually gets.
• Not a head count. 80 at a small cafe and 80 at a stadium are very different numbers of people. Both mean "close to its usual peak".
• Easy to compare in time. You can see at a glance when a venue is busy or quiet, and rank hours or days for the same venue.

A relative scale works for any venue type and size without knowing its capacity, and it answers the question most people ask: when is it busy, and when is it quiet? Each hour also gets a label, from below average to above average, compared with the venue's weekly average.

How to use the forecast

API example: forecast a venue, then read one hour

First create the forecast with your private API key. The first forecast for a venue costs 2 credits.

curl -X POST "https://besttime.app/api/v1/forecasts?\
api_key_private=YOUR_PRIVATE_KEY&\
venue_name=Colosseum&\
venue_address=Piazza+del+Colosseo+Rome+Italy"

The response contains the forecast for every day of the week and a venue_id in venue_info. Then read one day and hour with your public key. day_int runs from 0 for Monday to 6 for Sunday, and hour from 0 to 23.

curl "https://besttime.app/api/v1/forecasts/hour/raw?\
api_key_public=YOUR_PUBLIC_KEY&\
venue_id=VENUE_ID&\
day_int=5&\
hour=14"

Response, trimmed, with example values:

{
  "status": "OK",
  "analysis": {
    "hour_raw": 85,
    "hour_analysis": {
      "hour": 14,
      "intensity_txt": "Above average",
      "intensity_nr": 1
    }
  }
}

hour_raw is the 0-100 busyness for Saturday at 2 PM. To find which venues will be busy or quiet at an hour, rather than checking one venue, use the Venue Filter API. Full request and response fields are in the BestTime documentation.

Limits to keep in mind

• Relative, not counts. If you need how many people came, use a door counter at your own venue or a visit-count provider such as Placer.ai or Advan Research.
• Coverage depends on signal. Busy public venues such as bars, restaurants, stores, museums and attractions are covered best. Small or rarely visited places may have no forecast.
• Forecasts show the usual pattern. A last-minute closure or an unplanned crowd is not in a forecast made in advance.
• Public venues only. Forecasts are not meant for private workplaces, homes or staffing plans.

Other ways to estimate how busy a place will be

See the foot traffic data providers compared, or where to get foot traffic data for free.