Dwell Time Analytics: The WiFi Metric Most Venues Are Ignoring

What Dwell Time Actually Measures

Dwell time, in WiFi terms, is how long a guest's device stayed connected: from login to disconnect or timeout. It is a stand in for how long the guest stayed in the venue, and a rough one. Know where it goes wrong before you act on it:

• Portal timeouts cut sessions short. A captive portal authorises a guest for a set time. On Omada this is the Authentication Timeout: when the set time is reached, clients need to sign in again (TP-Link Omada, portal authentication). If your timeout is one hour, no session will look longer than an hour, however long the guest stayed. Our guide to choosing a guest WiFi session timeout covers how to set it.
• Phones drop off WiFi. A phone left face down on the table may drop the connection to save power and rejoin later. A guest who steps outside to take a call can still show as connected.
• One guest, several devices. A laptop and a phone are two sessions.
• Only connected guests count. Guests who never join your WiFi are invisible.

At the level of averages across many guests, these errors partly wash out. At the level of one guest on one day, they do not. Use dwell time to compare patterns (Tuesday against Friday, this month against last), not to judge single visits.

Where the data comes from. VoqadoWiFi records when each guest logged in and refreshes a "last seen" time when their device comes back through the portal. It does not pull minute by minute session data from your controller, so treat the session lengths in your VoqadoWiFi dashboard as rough. For how long each device has been connected right now, your controller is the better source: the Omada client list can be sorted by uptime (TP-Link Omada, monitoring clients), and the UniFi Network client list shows connected devices too.

Benchmarks by Venue Type

Dwell time benchmarks do not travel well between venues: the numbers depend heavily on your portal timeout and the way each platform measures a session. Two venues with the same guests can report very different averages.

What we can say is what usually drives dwell time in each type of venue, so you know what a change might mean:

Build your own baseline instead: four weeks of data, split by day of week and hour, with your portal timeout unchanged throughout. That is the number to compare against.

Dwell Time and Spend Correlation

In venues that sell by the item (another coffee, another round), a longer stay often means more orders. Not always: a co-working guest on one coffee for four hours stays long and spends little, and a table that lingers on a busy Friday may cost you a second sitting.

Worked example with assumed numbers, not measured VoqadoWiFi results: if guests in a coffee shop spend about £0.10 per minute they stay, a 20 minute visit is worth about £2 and a 90 minute visit about £9. VoqadoWiFi has no POS integration, so check the pattern against your own till data.

If spend per minute holds roughly steady in your venue, anything that keeps guests a little longer is worth testing: more comfortable seating, a food menu that justifies a second order, music at a sensible volume. If it does not hold steady, longer stays may simply mean fewer seats for the next guest.

Identifying Underperforming Dayparts

Split dwell time by hour of day. VoqadoWiFi's analytics already show logins by hour and by day of week in each venue's local time, which tells you when guests arrive. Pair that with session length to see how long they stay.

Illustrative example: a coffee shop sees long sessions from 10:00 to 12:00 and much shorter ones from 14:00 to 16:00. That afternoon gap could have several causes: the light changes and the room feels less inviting, the lunch menu has ended and nothing replaces it, or the after school rush makes the room noisy. The data shows you the gap. Standing in the room at 15:00 tells you why.

Things to try: an afternoon offer (pastry and coffee), a corner with sockets and good light, or a short afternoon menu. Illustration, using the assumed £0.10 a minute from above: if a change kept afternoon guests 15 minutes longer, that would be worth about £1.50 per guest. Check whether it actually did against your till before you keep it.

Dwell Time as a Satisfaction Proxy

A sudden change in average session length is worth a look. Illustrative example: average sessions in one week drop well below the previous four weeks. Before you blame the service, rule out the boring causes first: did someone change the portal timeout, did an access point go offline, did the WiFi get slower? Then look at the venue: a new staff member, a kitchen problem slowing service, building work next door, a menu change.

If you run post visit surveys, read the comments from the same week. Your reviews may say the same thing. Dwell time gives you the signal; the comments and your own eyes give you the diagnosis.

Layout Optimisation Using Multi-Zone Dwell Data

If your access points each cover a distinct area (dining room, bar, terrace, lounge), connection data per access point shows where guests settle and where they pass through. This data lives in your Omada or UniFi controller, which knows which access point each device is on. VoqadoWiFi reports by venue, not by zone.

Read it with care. Devices roam between access points, and coverage areas overlap, so a guest at a table near the bar may sit on the bar's access point. Treat it as a hint, then go and look.

A terrace with shorter stays than the dining room may have a comfort problem: wind, sun, traffic noise. A bar area with longer stays than the dining room may say the bar is the better experience, which is worth knowing when you plan staffing or seating. Change one thing at a time and compare a few weeks before and after.