
A restaurant manager does not need another dashboard: they need to know what to do now, with little reading and no data entry for things someone else already knows. A useful tool removes admin work before it asks for any, and every piece of information it shows leads to an action or to its source in one step.
3pm, between two services
A manager walks into the office, opens a computer and finds a screen with eight charts: completion rate, average check trend, response time, score by station. They have twenty minutes before the team meeting. They look at the first chart, search for what to do with it, find nothing, close the tab and write by hand a list of three things to sort out before dinner service.
What would have helped fit in three lines: yesterday's fish delivery is short, the new line cook has not yet signed off on the hot station, a temperature log was not done this morning. All three were known to someone. None was in front of them.
Why the dashboard misses its target
It answers the owner's question, not the manager's. The owner wants to compare. The manager wants to act. The same screen does not serve both.
It shows results, not next actions. Knowing an indicator is low does not tell you where to start.
It adds data entry. For the screen to be alive, it has to be fed. The manager becomes a data entry operator on top of running the team.
It opens the door to monitoring. A per-person indicator quickly becomes a way to compare, then to judge. Teams feel it and protect themselves.
It makes people forget the memory. Today's numbers without the history (why we decided this, what had been promised) do not guide action.
What a manager is really looking for
| Moment | What they ask themselves | What helps |
|---|---|---|
| Before opening | What do I check, who is in? | Their typical day for the role, absences, what was left over from yesterday |
| During service | Who needs help, what is missing? | A short instruction, in the right place, with no login |
| After service | What do I need to hand over? | Facts already entered, pre-filled in the handover |
| In a one-to-one | What do I remember? | The person's dated commitments and facts |
| In the evening | What is still open? | A list of items, with their age |
Five rules for a tool that helps
1. Count what is removed, not only what is displayed
Before installing anything, note what the manager enters today: how many times the same information, in how many places. The tool should shorten that list.
2. Bring the day's priorities up without navigation
Three lines at most at start-up, each with its reason. The manager does not have to search.
3. Pre-fill the handover with known facts
Tasks done, incidents, logs, absences: already known, so not re-entered. The manager adds only what exists nowhere else.
4. Let people ask a question instead of digging
"What was not done this morning?", "Who has signed off on the hot station?" A short answer, with the items behind it.
5. Never rank people
A manager should be able to see facts about a station or a task without them turning into an individual score. This is a design choice, not a setting.
A simple test
Ask a manager to show you, without preparation, what they check before their shift. If they open three tools and a notebook, the issue is not analysis: it is that the useful facts are scattered. Repeat the test after any change of tools.
Common mistakes
- Buying an analytics tool before reducing the existing data entry.
- Judging the success of a rollout by the number of logins.
- Displaying every indicator "just in case."
- Asking the manager to explain gaps the tool could document.
- Using the data to compare managers with each other.
What to measure
- Entries eliminated, counted per week and per manager.
- The time between arriving at the office and the first useful action of the day.
- The number of follow-ups the owner still has to make to get a piece of information.
Where Tsuno comes in
Tsuno is designed to remove data entry: the tasks of the typical day are written once, facts already known are carried into the handover, and the manager can ask a question in plain language, from Tsuno or from their own AI. The answer is short and rests on dated facts, each with its provenance. Tsuno prepares the action, a person confirms, everything is traced. It does not grade people. Details on the features page.
The companion articles
This article sets the framework; the next five build on it:
- Reports: why weekly manager reports often teach you nothing.
- The one-to-one conversation: how to prepare an employee conversation with facts, not impressions.
- Goals: restaurant employee goals: why they often fail.
- A new hire's independence: how to know when a new restaurant employee is actually ready.
- The limits of the tool: why management software can quickly feel like surveillance.
Key takeaways
The manager needs their next action, not an analytics screen. Judge a tool by the data entry it removes, the clarity of its priorities and its refusal to rank people.
Frequently asked questions
Does a restaurant manager need a dashboard?
Mostly they need to know what to do right now, without navigating. A useful dashboard shows very little and leads every line to an action or its source. Charts nobody looks at do not help.
What should a manager see at the start of their shift?
The day's priorities, the tasks of their typical day, what was left open from the day before, and absences or team changes. The rest can wait for a voluntary lookup.
How do you stop a tool from becoming one more admin burden?
Measure what the tool removes before what it asks for: entries eliminated, chasing avoided, searches shortened. If the manager enters more data than before, something is badly designed.
Can you question your data instead of reading screens?
Yes, with a tool that keeps a memory of the facts. You ask a plain-language question and get a brief answer with the items behind it, without digging through menus.
How can a tool help without monitoring teams?
By connecting dated facts whose source everyone can see, and by refusing to grade or rank people. The software recalls, people decide.