How Much Beer Does a Bar Pour and Never Sell?
One venue, Aguila's, The Office. Six months of data, February through August 2026. Single site, so read this as one data point, not a benchmark.
What we measured
We installed in the keg room in February 2026 and synced to the venue's POS. Every pour recorded against every sale, 96 data points every three seconds: temperature, pressure, volume, sanitation.
The number we track is PourScore, the ratio of ounces sold to ounces poured. At install, The Office read 89.2%. Roughly one ounce in nine left the tap and never reached a ticket.
Where the ounces went
Foam. Overpours. Temperature and pressure running outside spec. Pours after close.
None of it appeared in the POS. A POS records what was sold. It has no record of what was poured, so the gap between the two is invisible to it by design.
That is also why manual counting misses this. The losses are small per pour and continuous. That pattern is hard for a person to catch across a busy shift and simple for a sensor to record.
What changed after calibration
Calibration is the process of correcting what the data exposes: lines pouring outside spec, temperature and pressure out of range, cleaning past due. The Office moved to on target and has held there.
Six months of results:
Recovered revenue: $41,438.17
Beer saved: 56.5 kegs, or 7,002 pints
Return on Floteq spend: 13.8x
Annualized run rate on recovery: $82,876.35
Volume did not change. No new taps, no menu change, no price increase, no additional guests. The same beer moved through the same lines. More of it was sold.
What it looks like on the P&L
Over the most recent eight week window, The Office ran $87,746.40 in draft revenue at a 72.7% margin.
Recovered ounces land almost entirely in that margin. The beer was already purchased, already tapped, already poured. A recovered ounce is not the same as a new sale. It costs nothing additional to produce.
What holds the number
Temperature stayed in range 97.82% of the time over the most recent window. Pressure stayed in range 99.51%.
Those two figures are why the score holds. Draft performance is not a setting you adjust once. It is the running output of equipment that moves, and the score tracks it continuously rather than at the end of a quarter.
Why this matters across locations
The Office is one keg room. A group running ten has ten independent versions of this number.
Without a shared measure, each site reports its own version of fine. The spread between the strongest and the weakest stays invisible until it reaches the annual, and by then it is a full year of pours.
One measure across every site changes that. You see which venue is losing ounces, on which lines, in the week it starts.
