A virtual waiting room hides overselling behind averages, and the three numbers that expose it are inflow per minute, outflow per minute, and walkaway rate by time slice. Queue-it's Monitor shows inflow, visitors in the room, outflow per minute, maximum outflow, and wait time. The walkaway rate is the key lost-revenue proxy, and an overall 4 percent walkaway can hide 22 percent at the Saturday 2pm peak. Watch the numbers by slice, not by day.
Inflow, outflow, and the gap between them
Inflow is the rate visitors arrive. Outflow is the rate the queue releases them to the store. The gap between them is the queue itself, and the gap is where overselling hides. When inflow runs ten times outflow for the first five minutes, the queue is growing faster than the store can serve it, and the wait time climbs with every minute. The dashboard shows all three in real time, and the numbers that matter are the per-minute figures, not the totals.
Maximum outflow is the number that decides the drop. It is the fastest rate the store can release visitors without falling over, and it should be set from the store's actual capacity, not from the size of the audience. A store that can handle 500 checkouts a minute should release 500 a minute, even if 50,000 people are waiting. Releasing faster does not sell more. It crashes the checkout and loses the sale.
Related reading on this site: Home and Pre-Queue Registration: Verified Entry Before the Drop Opens cover the neighbouring problems in more depth.
Walkaway rate is the lost-revenue proxy
Walkaway rate is the percentage of visitors who leave the queue before reaching the store, and it is the closest thing to a lost-revenue number the queue produces. An overall walkaway of 4 percent sounds fine. A walkaway of 22 percent at the Saturday 2pm peak is a different story, and it is the story that matters, because the peak is when the inventory is still there and the demand is highest. Averages hide the peak. Slices reveal it.
The fix is to read the numbers by slice, not by day. Break walkaway, wait time, and outflow into fifteen-minute windows and look for the slice where the queue broke. The slice that broke is the one where outflow was too low, wait time was too long, or the queue page gave visitors no reason to stay. Fix the slice and the average fixes itself.
The abuse view is a different dashboard
The same monitoring shows the abuse signals. Queue-it's Monitor includes an abuse view that tracks CAPTCHAs solved, proof-of-work challenges rejected, and fast re-enters, visitors who leave the queue and rejoin immediately to reset their position. Those numbers do not affect the walkaway rate, but they affect who gets to buy. A drop with a clean walkaway rate and a high fast re-enter count sold its inventory to the visitors who gamed the queue, not the visitors who waited.
Read the two dashboards together. The performance numbers say whether the queue served the drop. The abuse numbers say whether the drop served the right visitors. A drop that sold out in four minutes with a 4 percent walkaway and a clean abuse view went well. A drop that sold out in four minutes with a 4 percent walkaway and a fast re-enter spike sold out to the wrong people.
Frequently asked questions
What three numbers tell me a drop is oversold?
Inflow per minute, outflow per minute, and walkaway rate by time slice. When inflow runs far above outflow, the queue grows faster than the store can serve it, and the walkaway rate by slice shows where visitors gave up.
What is walkaway rate?
The percentage of visitors who leave the queue before reaching the store. It is the closest thing to a lost-revenue number the queue produces, and an overall 4 percent can hide 22 percent at the Saturday 2pm peak.
What is maximum outflow?
The fastest rate the store can release visitors without falling over, set from the store's actual checkout capacity. Releasing faster does not sell more. It crashes the checkout and loses the sale.
How do I spot queue abuse in the analytics?
Look at the abuse view: CAPTCHAs solved, proof-of-work challenges rejected, and fast re-enters. A high fast re-enter count means visitors are leaving and rejoining to reset their position, and the drop sold to the visitors who gamed the queue.