On this page
When a bar decides to take consistency seriously, the business case is almost always built on liquor cost. Tighten the pours, reduce the variance, recover the margin. That case is real and it is usually enough to justify the work on its own.
It is also the smaller half of the number.
The cost you can see
Over pouring, comps that were not tracked, waste, and stock that leaves without a corresponding sale. This is the part every inventory tool in the category is built to find, and they find it reasonably well. It is countable, it appears in a variance figure, and an owner can look at it and decide whether it is worth acting on.
The limitation is not the counting. It is that a variance figure for a four week period tells you that something happened somewhere in twenty eight days. Acting on that requires knowing which nights and which products, which is a different question from how much.
The cost you cannot see
A guest comes in for the first time and has a good night. They come back three weeks later. This time the drink is different, or the wait is longer, or the thing that made the first visit good is not there. They do not complain. They do not leave a review. They simply do not come back, and no system you own records that anything happened.
This is the expensive one, and it is invisible by construction. A bar with a slow leak of second visits looks fine on a weekly revenue line for a long time, because the top of the funnel is still working. The damage shows up much later as a business that needs constant new custom to stand still.
What the review study can and cannot tell you
We classified 13,063 guest reviews across 41 Texas venues, twice, using separate models, and checked the result against state alcohol tax filings. That work is on the research page and it is the only dataset we cite on this site.
What it is good for is showing what guests actually write about when they write about a venue, and how that maps to how the venue was trading. What it cannot do is tell you how many people had a poor second visit and silently stopped coming, because those people by definition did not write anything. Nobody has that number. Anyone who quotes you one has made it up.
So the honest form of this argument is not a statistic. It is that the countable cost of inconsistency is the part that is easy to measure, and there is no good reason to believe the easy to measure part is the whole of it.
The third cost: the people
There is a cost between the two, and it lands on the team.
Working in an inconsistent bar is harder than working in a consistent one. The handover is unreliable, so you start every shift finding out what happened. The standard depends on who is on, so you are either carrying people or being carried. Problems recur, because nothing that was noticed on Tuesday reached anyone by Friday.
That is the environment good staff leave. And when they leave, they take the institutional knowledge that was holding the place together, which makes the inconsistency worse, which makes the next person more likely to leave. One venue in our case studies reported a 73 percent reduction in turnover, attributed to LearnTAP. That is a client reported figure from one venue, not a benchmark, and we would not present it as one. What it does illustrate is that the staffing cost and the consistency problem are the same problem seen from two ends.
How to size it for your own bar
You cannot measure the invisible cost directly, but you can bound the visible one and stop guessing about the rest.
Start with variance at the level of the shift rather than the period. Which products, which nights, which sections. Then look at whether the pattern is stable or drifting, because a steady 3 percent and a 3 percent that was 1 percent six months ago are completely different problems with the same number attached.
That is a reasonable afternoon of work if the transaction record is available to you in that shape, and effectively impossible if it is not, which is the actual reason most bars have never done it.
If you want it done against your own trading rather than described, that is what a CoreTAP demo is.