Who is this assistant for?

You post all month and at close you cannot tell which sales came from it. Every promotion starts strong and dies in the second week. You filled the room with an offer and the month closed with less margin than before. The team hears about the campaign the day it starts and runs half of it.

Posting for the sake of posting is not a campaign: it is noise. A real campaign has a money objective, a defined audience, an offer with a closing date and a calendar the team can execute without dropping the operation. This assistant builds that whole system around the goal you choose — launching a dish, filling a date, moving a channel — and makes clear from day one how you will know it paid off.

AI assistant · N3 Analysis with a verdict

◆ Included in the annual plan✍️ Copywriting and High-Impact Campaignsv1.3.0 · 2026-08-25

High-Impact Campaign Builder for Restaurants

Designs a complete campaign end to end: money objective, audience, message, offer, channels and a 4-week calendar, with the way to measure whether it worked.

Depth

9/23

Your data

Compatible AIs

Steps to use it

Depth of the resulttap a level to see what you get

◀ Put out today's fireRun the year ▶

📊

N3Analysis with a verdict

Crosses YOUR numbers and takes a stand, with a table, units and currency. Not options: a recommendation you can defend.

◆ This assistant 24 of the 164 published assistants are at this level

Is this for you?

What it does and what you get

Posting for the sake of posting is not a campaign: it is noise.
A real campaign has a money objective, a defined audience, an offer with a closing date and a calendar the team can execute without dropping the operation.
This assistant builds that whole system around the goal you choose — launching a dish, filling a date, moving a channel — and makes clear from day one how you will know it paid off.

What you end up with

A 4-week campaign in tables: objective with a figure, core message, offer with conditions, week-by-channel calendar, and the measurement board with the 3 numbers that decide whether it runs again.

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

Sazón de OrigenFictional contemporary Colombian restaurant in Medellín, 2 locations, 6 years in business. Every sample result in the library comes from this same case. Fictional restaurant

The owner asked this: "I want to lift the operating margin without touching menu prices: Tuesday-to-Friday lunch runs at 79% of what the kitchen can handle and the delivery platform is eating what our own delivery leaves behind. Where do I strike, and with what figure?"

The window is Tuesday-to-Friday lunch, 12:00 to 14:30, across both locations. Two were ruled out. Sunday midday runs at 92% of what the kitchen can serve without dropping the standard —52 covers per location against a ceiling of 56— and six more covers per location stretch the wait for the family table, which is the one that comes back. Friday and Saturday nights are not among your peak moments: the house has no practised hand there, and you strike where it already knows how to respond.

Row	What gets written down
Objective	9,200,000 COP of extra sales between Tuesday the 1st and Friday the 25th of September. That is 2,300,000 COP a week; divided by the 48,000 COP average check, 48 more covers a week: 6 per location per lunch day.
Audience	The 30-to-45 professionals who already eat lunch on weekdays within 3 km of each location. Weekend families stay out of it: that slot is not touched.
Message	"The house lunch, served and paid for before your hour runs out." One idea, and it is his, not yours.
Offer	House lunch, 44,000 COP: three-meat sancocho or garlic trout in the lunch portion, house lemonade and dessert of the day. Tuesday to Friday, 12:00 to 14:30, four weeks, 30 plates per location per day.

The house bandeja stays out of the offer on purpose: it sells on its own at 42,000 COP, and putting it in the set would have given up margin without bringing a single new cover.

The sum against the team. The 18 operations people split 9 to a location. With those 9, the window serves 44 covers per location per day today and holds 56 before the second course starts running late. The 6 the campaign asks for take it to 50, with 6 of headroom left. If your baseline over the previous 8 weeks is not 44, the objective is recalculated before anything is printed. Illustrative figures, built from the data you gave.

The Channel column only takes what already brings money in: the dining room, your own delivery and the delivery platform. Instagram and WhatsApp move people but collect nothing, so they live inside the action and not in the column.

Week	Channel	Action	Owner	Done when
1 · Sep 1-4	Dining room	Card on all 30 tables and the server's line, said in one sentence before the menu is handed over	Floor manager	All 30 tables set on Tuesday at 11:55
1 · Sep 1-4	Own delivery	Insert in every package, with the house lunch and the location's WhatsApp number	Delivery assistant	400 inserts out by Friday
2 · Sep 8-11	Own delivery	One-to-one message to everyone who ordered in the last 60 days; no broadcast list	Admin	12 orders carrying the ALM code
2 · Sep 8-11	Dining room	Instagram ads within 3 km of each location, Tuesday to Thursday only, between 10:00 and 13:00	Admin	160,000 COP spent of the 640,000
3 · Sep 15-18	Delivery platform	Note on the listing: the house lunch is served in the dining room. The plate is not listed there	Admin	Listing updated on Tuesday
3 · Sep 15-18	Dining room	Card inside the check with the closing date written on it: Friday the 25th	Floor manager	25 extra covers that week
4 · Sep 22-25	Own delivery	Call the 20 who ordered once in September and did not come back	Delivery assistant	20 calls made by Wednesday
4 · Sep 22-25	Dining room	Last day announced at the table and at the door; the card comes down Friday at close	Floor manager	Cards collected on the 25th at 15:00

The offer is withdrawn on Friday the 25th and does not get stretched. At 60 days an offer stops being a campaign and becomes the new price, and by then nobody thanks you for it.

The costing sheet first, the poster after. The three lines come from the standard recipe with the lunch portion weight, which is not the evening one.

Line	Cost per cover
Three-meat sancocho, lunch portion	9,800 COP
House lemonade	2,100 COP
Dessert of the day, house cake	2,100 COP
Total	14,000 COP

Against the 44,000 COP price, food cost lands at 31.8%, under the 32% ceiling the campaign set itself. The whole house runs at 34%: this plate was built backwards, from the cost towards the price, and that is why it fits. Payroll and rent stay off the plate and sit in the budget.

The margin given up. A normal 48,000 COP cover, at 34% raw material, leaves 31,680 COP of contribution. The house lunch leaves 30,000 COP. The difference is 1,680 COP per cover, and it is only lost on the guest who was coming anyway and switches to the set.

The covers that give it back. Assumed, until you check it against your own tickets: 4 in 10 covers from the baseline switch, which is 141 of the 352 in a week. That gives up 236,880 COP a week. At 30,000 COP of contribution each, 8 new covers a week give back everything conceded, and the objective asks for 48. Put the other way round: one new cover pays for what almost 18 regulars give up.

Why the plate is not listed on the platform. With the 27% commission the same lunch leaves 18,120 COP instead of 30,000. Those 11,880 COP of difference are, almost to the peso, what this campaign is trying to win back. Illustrative figures.

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 used every week. The board carries three numbers with their columns —Number, How it is computed, Reading day and Traffic light—: the extra contribution margin of the window, which is the one measuring the year's goal; extra covers against the previous 8-week average; and house lunch take-up, which carries two reds, because going past 60% is bad news too. It comes with the decision rule of Monday the 14th at 9:00, switching the ads off below 60 accumulated covers and raising the daily cap from 30 to 40 plates above 96. The budget lists the five expenses with their total, the six weekly hours it takes from the team, and the 11:40 checklist: fifteen minutes, five boxes, taped to the kitchen door.

See the full example — free account

Works with these AIs

Claude con Claude Design — the best for this assistant The only one that COMPOSES the piece instead of describing it: it opens its canvas, builds one artboard per unit and sets the text in place. Ask it for the style contract before the first piece, and have screenshots of your reference account at hand.

How to use it

Define what you want to achieve and by when: a figure and a date beat an intention.

Paste the assistant into your AI with your My Restaurant data loaded and state your objective in one sentence.

Receive the full campaign, adjust the calendar to your team's reality and run it, measuring what was agreed every week.

Which of your data it uses

of 23

This assistant works with 9 of the 23 fields in «My Restaurant», among them Main goal this year, Your main guest and Strong and slow dayparts. 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.

Main goal this yearYour main guestStrong and slow daypartsActive sales channelsActive social channels (with the @ or the link)Average monthly salesAverage check (with currency)Team sizeBest-selling dishes

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.

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.

What you want to achieve

attract new guests win back guests raise the check fill slow days build loyalty repeat purchase reservations delivery event

On what criterion

the slow hours the reason for eating out the occasion