AI assistant · N4 Executable plan
Slow-Day Campaign Builder for Restaurants
Decides whether your slow day deserves a fight or plain administration: runs the day's money math, diagnoses the real cause and builds the 4-week campaign that gives that day a reason of its own.
N4Executable plan
A plan with owners, deadlines and the metric you review. Not advice: an agenda someone can execute on Monday.
◆ This assistant 23 of the 164 published assistants are at this levelIs this for you?
What it does and what you get
- Almost every restaurant carries a Monday or Tuesday that barely covers payroll, and the usual reaction is copying the neighbor's promotion without doing the math.
- This assistant first tells whether that day has a possible audience or is structurally slow, computes what opening it costs and, only if the fight is worth it, designs the named campaign with calendar and measurement to actually move it.
What you end up with
The slow day's math in your currency, the cause diagnosed with evidence, an argued fight-or-administer decision and the 4-week campaign with calendar and measurement.
EXAMPLE This is what you getAn example built with Sazón de Origen, the house's fictional restaurant.Not your data: your result is built with YOUR restaurant's.See the full example →Sample result
The owner asked this: "Monday is my worst day: about 36 tickets across the whole shift, 11:30 to 8:30, and my own delivery barely rings. Do I fight that day or cut the hours?"
The slow day is Monday, all of it, and two candidates were ruled out before saying so. Saturday does not appear among your strong moments either, but it is not the slow day: neighbourhood families already choose it and its sales sit behind Sunday, not beside Monday; check it against old order slips before closing the question. And Tuesday-to-Friday nights are genuinely hollow, but that is a window inside a good day and it belongs to the daypart sheet: pushing the lunch that already fills into the evening spreads sales, it does not add them.
The baseline is 6 working Mondays out of the last 8 weeks. The other 2 were public holidays and drop out: in Colombia almost every holiday falls on a Monday, and a holiday Monday looks nothing like a working one. The math is for the main location; the second one repeats the exercise on its own baseline, which is not the same.
| Item | Estimated value | Source |
|---|---|---|
| Average Monday sales | 1,512,000 COP | 6 working Mondays, the 2 holidays removed |
| Tickets on the day | 36 · 27 land before 3:00 p.m. and 9 after 6:00 | Shift order slips |
| Monday average check | 42,000 COP | 1,512,000 ÷ 36 |
| Food cost, 34% | −514,000 COP | House food cost |
| Shift payroll: 7 of the location's 9 operations people plus 1 from admin, 8 hours | −780,000 COP | Assumed until you check it against real payroll |
| Utilities for the day | −96,000 COP | Prorated bills |
| Waste from a 3-day mise en place | −74,000 COP | Assumed |
| Left to pay the day's rent | 48,000 COP | Subtraction |
| Rent, insurance and admin for the day | −620,000 COP | Assumed until you check it against your P&L |
| Monday result | −572,000 COP | Subtraction |
Monday leaves 48,000 COP —one average check— to pay for a day of rent that costs 620,000. Every open Monday takes 572,000 COP with it: 2,288,000 in a month of four Mondays, and that is one location only. The declared house labour cost is 28%; that same shift, on a Monday, takes 52% of what the day sells. Illustrative figures, built from the data you gave.
Three possible causes, and only one survives the evidence.
Medellín's calendar is ruled out. The 2 Mondays removed from the baseline were holidays and sold 3,870,000 and 3,640,000 COP: two and a half times a normal Monday, with the same room and menu. The city does go out on a Monday when it has a reason.
Half speed explains the check, not the covers. Short menu, no dessert of the day, nobody at the door: that is where the 6,000 COP sit between the Monday check, 42,000, and the house check, 48,000. It is real, cheap to fix, and it brings in not one new table.
Habit is what is left. The 30-to-45 professional who lunches here Tuesday to Friday is still at the office on Monday: the dining room does 24 tickets against a Wednesday's 64 (38%), and your own delivery and the platform hold half their volume. The people are nearby and they order; they have no reason to sit down.
| Low cost of opening | High cost of opening | |
|---|---|---|
| High possible audience | A notice is enough. No slot of yours sits here today. | Monday lunch. 27 tickets in three hours with the full kitchen on. You fight it with a reason of its own, not with a discount. |
| Low possible audience | Hold while it costs nothing: short menu, one person on. | Monday night. 9 tickets after 6:00, 7 of them in the room, with the whole shift inside. You administer it: hours cut by 2. |
The call: fight the lunch and administer the night, in the same week. With the whole day open and today's check, Monday needs 57 covers to pay shift, utilities, waste and its day of rent. Closing the room at 6:30 —the kitchen dispatches own delivery until 8:00, where the other 2 tickets come from— the door drops from 950,000 to 680,000 COP and break-even falls to 40. The cut is not free: 7 covers are lost, 194,000 COP of margin, against 270,000 saved. It leaves 76,000 net per Monday and lowers the bar by 17 covers, more than any promotion moves.
The judgement, declared and arguable: first you make the door cheaper, then you call people in. The concession: if your second location has a neighbourhood crowd on Monday nights, that quadrant reads the other way round and there you do not cut.
Monday Pot. Four editions, first Monday to fourth, with the review date written before the run starts.
| Row | What gets written down |
|---|---|
| Concept | The three-meat sancocho, today a Sunday dish, comes out on Mondays from 12:00 to 3:00 in a pot to share, served at the table. Mondays only, lunch only, main location only. |
| Reason | Your guest comes for a craving for home cooking, and Monday is the day nobody cooked. No new reason is invented: the one that already brings people in on Sunday is moved one day over. |
| Offer | Pot for two, 96,000 COP, with corn on the cob, avocado, rice and panela lemonade. The price per person does not move with the size of the pot: 48,000, exactly the house average check and inside the 28,000 to 72,000 menu range. Cost per cover 15,100 COP, food cost of 31%, under the 32% ceiling (assumed until you check it against your costing sheet). Not one peso of discount: Monday gets a format, not a markdown. After 60 days a markdown is simply your price. |
| Announcement | The same text for the room card, the delivery insert, the Instagram story and the platform listing. One line on price and one on hours. |
| Gesture in the room | The pot arrives uncovered and is served in front of the guest: the next table watches it go by, and that is the whole advertising budget for the second Monday. The server tells it in one sentence, not three. Next Monday's date is printed on the check. |
The announcement, to paste as is:
The sancocho does not end on Sunday. Monday at midday the three-meat pot comes to the table to share: corn on the cob, avocado, rice and panela lemonade. 48,000 per person, from 12:00 to 3:00, while the day's pot lasts. No bookings: there is pot until it runs out.
The day's pot has a bottom, and that is on purpose. You set the limit before waste sets it for you, since waste is what eats a slow day's margin: 44 portions are cooked, four above break-even, and Monday's leftovers are Tuesday's staff lunch.
Illustrative example generated with a fictional restaurant. Not a promise of results, not a client case.
This is 3 of 5 parts. Behind the wall sit the two pieces that get executed. The four-Monday plan, row by row, with the columns Week, Action, Channel and Owner: the costing sheet signed and the three meats bought in week zero, the card inside the lunch check, the delivery insert with the Sunday 6:00 p.m. story, the one-to-one WhatsApp message and the platform highlight switched on only from 11:30 to 3:00. And the cut-off, written before the run: the fourth Monday at 3:30 in the afternoon, till closed, where under 36 covers sends the day to administration, 36 to 39 buys four more Mondays, and 40 or more keeps it on the calendar. The measurement block carries the tracking table for the four editions —covers, average check, pot mix and margin— and the cannibalisation line that checks Tuesday of that same week.
See the full example — free accountWorks with these AIs
How to use it
Which of your data it uses
This assistant works with 14 of the 23 fields in «My Restaurant», among them Strong and slow dayparts, Business type and City and country. It does not ask out of curiosity: these are what make the answer speak about YOUR scale instead of an industry average — the same calculation on a business of another size returns a number you cannot decide with. You fill them once and they apply across the library, so the second assistant you open already starts with them in place. Whatever you leave blank, the assistant asks for one thing at a time instead of inventing it, and says so before answering, so you know what it is working from.
Always up to date
Every assistant carries a visible version and date. When an AI changes how it works, the assistant's version goes up and you see what changed.
v1.2.02026-08-20Grounding: the task stops describing a generic restaurant. The slow day is now deduced by subtracting the declared strong moments; the cost of opening it is judged with the house team size and labour cost; the hook comes from the star dishes inside the price range without breaking the average check, and distribution happens only through the sales channels and active social profiles the restaurant already runs. The boundary with the daypart sheet is stated in step one. Nine business fields working outside the header; role, method and data block untouched.v1.2.02026-08-20Grounding: the task stops describing a generic restaurant. The slow day is now deduced by subtracting the declared strong moments; the cost of opening it is judged with the house team size and labour cost; the hook comes from the star dishes inside the price range without breaking the average check, and distribution happens only through the sales channels and active social profiles the restaurant already runs. The boundary with the daypart sheet is stated in step one. Nine business fields working outside the header; role, method and data block untouched.v1.2.02026-08-20Grounding: the task stops describing a generic restaurant. The slow day is now deduced by subtracting the declared strong moments; the cost of opening it is judged with the house team size and labour cost; the hook comes from the star dishes inside the price range without breaking the average check, and distribution happens only through the sales channels and active social profiles the restaurant already runs. The boundary with the daypart sheet is stated in step one. Nine business fields working outside the header; role, method and data block untouched.v1.1.12026-08-14Spelling: Spanish accents restored in the prose. No field marker, link or meaning changed.v1.1.02026-08-13Body rewritten under the margin mandate: 6 steps to decide and build the slow day, 5 named output blocks with columns, 8 business fields, named trade traps, verification checklist and cost-of-opening and measurement figures with units.Same method, another variant
Inside the app this is a single control: you press it and the assistant reframes itself, including the variants that don't appear here because they don't have their own page yet.
More assistants in this category
Builds the campaign that shortens the gap between one visit and the next: plants the next visit inside the cur…
N4 · Executable planPicks the consumption reason that moves the biggest check among those already bringing people to your tables, …
N5 · Complete systemBuilds your campaign's landing page — bookings, delivery, an event or an offer — as a publish-ready HTML file,…
Included in the always-growing library
Access to every published assistant, adapted to your AI and personalised with your restaurant's data, updates included.
