Loading
Loading
Loading writing
Operational writing rather than content marketing. Where a piece makes a claim about how bars behave, it is either traceable to our published research or it is labelled as opinion.
Action31 Aug 20264 min read
Every bar tool ends at a number, a report or a dashboard. The interpretation and the follow up are left to the operator. That gap is where the money goes.
Read itAction3 min read
A worked example, and the four causes one variance number cannot tell apart.
Action2 min read
One number, averaged across everything that could have caused it.
Action2 min read
A short buying test you can run in any demo, including ours.
Action2 min read
What the register genuinely gives you, and the three questions it cannot answer.
Action2 min read
The cost of a finding rises with its age. Here is roughly how fast.
Consistency4 min read
Four things that actually hold a recipe when nobody is watching.
Consistency3 min read
The expensive half of inconsistency is invisible by construction.
Consistency3 min read
Both sites end up acceptable, and different. That is the failure nobody braces for.
Consistency3 min read
Same numbers, opposite outcomes, depending entirely on who the finding is addressed to.
Consistency3 min read
Averaging two Fridays into a weekly total destroys the only signal worth having.
Action3 min read
If you bought inventory software and nothing moved, the tool is not underperforming. You had a different question.
Consistency3 min read
Most operational improvement fails at repetition, not at diagnosis.
Longevity3 min read
Three costs, in order of size. Nobody counts the third one.
Longevity3 min read
Losing a fact is cheap. Losing a pattern is not, and patterns are made of facts nobody recorded.
Longevity3 min read
The test takes a week off and looks at the same numbers.
Longevity3 min read
Three questions worth asking of any tool in this category.
Action3 min read
If you already know your variance and it is bad, counting again will confirm it rather than change it.
Culture3 min read
Visibility is upstream of the other two. That is a claim about sequence, not about pain.
Action2 min read
A vendor who answers all four plainly is telling you something. So is one who answers with a logo grid.
Learning3 min read
They retain whatever they used in week one. Everything else gets taught again later.
Learning3 min read
The obstacle is not disagreement. It is retrieval.
Learning3 min read
Speed finishes improving. Judgement does not.
Culture3 min read
The measurable half is the half with limited upside.
Culture3 min read
Ask a vendor what their product cannot see. It is the most informative question available.
Action3 min read
Three alerts a night is less useful than three a week, because the three a night get dismissed unread.
Longevity2 min read
Revenue is not a lever. It is the output of every lever.
Culture3 min read
Not is this person doing their job, but did this thing happen and what followed.
Culture2 min read
Specificity decays with time, which is why the two failures are really one.
Culture2 min read
Checking bundles three things together. Only one of them needs the owner.
Culture3 min read
A venue that chases has accountability that varies with a manager's mood.
Culture3 min read
Culture is what happens by default when nobody senior is watching.
Culture3 min read
The tell is precision. A sentiment score with a decimal point is a design decision.
Longevity2 min read
Fix the recurring irritations first, because they are the fastest to demonstrate.
Consistency3 min read
Setting a par from the average guarantees the variable products run out.
Action3 min read
Four clauses. Each one rules something out.
Action2 min read
The disappointment arrives around month three, with a beautiful live view and nothing running differently.
Consistency3 min read
Most venues have a live view and a monthly review, with the useful interval missing in between.
Action3 min read
Seven questions, and an afternoon long test you can run on every vendor at once.
If you want the underlying study rather than the commentary, that is at research: 13,063 guest reviews across 41 Texas venues, classified twice by separate models and checked against state alcohol tax filings.