The room was full.Nobody wrote it down.
Hospitality runs on recency: recent reviews, recent posts, a recent reason to come back. All three are jobs that lose every time to service.
Open source projects
Public on github.com/JIGGAI
AI plugins
Published on npm, free to install
AI plugin installs
65,301 to date, per npm
AI monthly plugin installs
2,811 in August 2026, per npm
What we hear in the first hour.
Not a generic list. These are the ones that come up in businesses shaped like yours.
We reply to reviews in a batch, weeks later, if at all.
A reply written the same shift reads as the same business that served them. Three weeks later it reads as admin.
The feed goes quiet exactly when we are busiest.
Which is when it matters most. Planned a month ahead and queued, it keeps posting through the rush.
One bad Saturday review is still the top result on Wednesday.
Recency cuts both ways. Answering fast is the only lever you control.
We collected emails for a year and never sent anything.
A list you never message is a list that will not remember opting in. Sooner is both better and safer.
Every venue in the group does their own thing.
One voice, several rooms — the same drafting, with what changes per venue kept to a handful of settings.
Nobody reads your menu first. They read what last week’s customers said about it.
So the marketing that matters most is the part that happens after service — and that is exactly the part nobody has energy for at closing.
The software already in the building.
We do not arrive with a platform to sell you. We arrive expecting these, and the work is usually in the gaps between them.
Bookings
OpenTable, ResDiary, SevenRooms, or the phone and a book. Where the covers and the repeat guests are.
Google Business Profile
Where most decisions are made, and where photos and hours go stale first.
TripAdvisor and delivery platforms
More review surfaces, each with different reply rules and none of them watched daily.
The de facto menu for a large share of new custom, and the first thing to lapse in a busy month.
The POS
Lightspeed, Square, Toast. Where what actually sold is recorded, and what to post about.
The mailing list
Collected diligently at the door, then rarely used, and often without a clear record of consent.
Answering the weekend before Monday.
Roughly what one entry in your report looks like — a real shape, with the numbers changed.
- What we saw
- Reviews from a busy weekend going unanswered into the working week, with no one owning the reply, and negative ones surfacing only when a manager happened to scroll.
- What it costs today
- On a profile where recency dominates, an unanswered Saturday complaint shapes the impression of the whole of the next week — and quiet Tuesdays are set by exactly that.
- What we would build
- Monitoring across Google and the review platforms you actually appear on, drafted replies in the venue’s own voice waiting for approval, immediate text escalation on anything negative, and a Monday summary of the weekend.
- How you would know it worked
- Time from review to reply, share answered, and the trend in reviews mentioning service recovery. If replies are still going out late, the approval step is with the wrong person.
The apology and anything it promises. A comped meal or a returned booking is a commercial decision, and a generated reply that implies one is a promise nobody in the building agreed to.
What coordination work costs you.
Three numbers you already half-know. Move them until they look like your business.
re-keying, chasing status, producing the same document again
on that work specifically, not their whole job
salary, tax, benefits, desk
A 75% capture rate. The other twenty-five per cent is judgement, exceptions, and not wanting to look greedy.
This is arithmetic, not a finding. It rests on three numbers you guessed. The assessment replaces all three with numbers we observed — and tells you which of those hours are actually worth automating.
See what a real finding looks like →What the six days look like.
The most common question we get is not about AI. It is what these people will actually do in my building.
- Day 1Walk the floorWhoever is on shift
We start where the work happens, not in a meeting room. Nobody prepares anything, and the first day is mostly watching.
- Day 2Sit with the people doing itOps, admin, finance
Conversations with the roles that touch the work most, and a real task followed end to end — including the parts that happen in a group chat.
- Day 3Systems, then a read-backWhoever holds the logins
What you run, what talks to what, and where a person is currently the integration. We tell you what we saw before we leave, while it is still cheap to correct.
- Days 4–6Research and discoveryOur desks, not yours
Away from your building. We cost the work we watched, model the alternatives, and test the shortlist against your own numbers rather than a framework.
- +1 weekOne recommendationPresented in person
The single change worth making first, specified precisely enough to build — with the ranked analysis behind it and the list of what we would not automate.
The questions this raises.
We are one site. Is this worth it?
The reviews team usually is; the social team depends on whether you have anything to say weekly. On a call we will tell you which of the three is worth your money and which is not.
Can it post food photos?
It schedules and drafts around what you supply. It will not generate images of food you did not serve — a picture of a dish that does not exist is the fastest way to earn the review you were trying to avoid.
What about TripAdvisor and the delivery apps?
Monitoring covers the surfaces you actually appear on, and each has its own reply mechanics. We confirm which are worth watching for your venue rather than assuming all of them.
Our group has six venues with different voices.
Voice is configuration per venue, and that stays inside the fixed price. What changes is drafting inputs, not the machinery — that is exactly the case a starter handles well.
Start with half an hour.
We'll tell you whether an assessment would pay for itself in an operation like yours. Sometimes the answer is no.