NewCost control: a token allowance a month for each model choice — when one runs out, the cheaper one answers.See how it works
memtro

AI for hospitality, across bookings, takings and reviews

A restaurant group or hotel runs on takings in Square or Stripe, reviews on Google and Trustpilot, bookings and events in a calendar and Eventbrite, a newsletter in Mailchimp and an inbox that never empties. Memtro connects to them and answers the questions an operator asks every morning, with each figure cited to the transaction or review it came from, and watches the reviews so nobody finds out a week late.

7-day free trial · card on file, nothing charged until it ends · cancel any time

Bookings, guests, reviews, staff rotas and the ledger answered from one memory, with every figure cited to its record.

The systems these questions draw on, connected to one memory. Every connector reads by default; the tools that change something are granted by name, per team.

The problem

What it looks like today

  • 01Takings by site are in one tool and the reviews that explain a bad week are in two others.
  • 02A one-star review is noticed when the next guest mentions it.
  • 03Event bookings, deposits and dietary notes arrive by email and live there.
  • 04The weekly owners' report is assembled by hand on Sunday night.
  • 05Next week's staffing is set from last week's impression rather than last week's covers.
  • 06A regular's preferences are known by one member of staff, who is off on Saturday.
How it works

What Memtro does about it

One memory in front of the systems you already pay for, reading by default and acting only where an administrator has granted the tool by name.

The stack

System by system

What Memtro reads in each part of the stack, and what it may change there once the tool has been granted.

Takings

Square or Stripe: transactions, refunds, payouts and the day's takings by site, read against the week before rather than typed into a sheet.

Reviews and social

Google Business Profile, Trustpilot, Meta Pages and Instagram, X and YouTube: what was said in public, when, by whom and about which site.

Bookings and events

Google Calendar, Outlook Calendar and Eventbrite: what is in the diary, who booked it, what has been paid and what the organiser last said.

Mail and forms

Gmail or Outlook with Typeform: enquiries, dietary notes, deposits and the attachment that came with them, read as text.

Marketing

Mailchimp, Klaviyo or Brevo: campaigns, sends, opens and clicks, joined to the bookings and takings that followed.

The back office

Xero or QuickBooks Online for the ledger, and a booking or EPOS system over Postgres or MySQL where it exposes a reporting database, read-only with the query shown.

There are 144 connectors in all, each with its own tools, its own permissions and a page of its own. Browse every connector →

Who it is for

Who asks what

Operator

Asks the Monday question once across every site: takings, covers, reviews and what changed, with the site named on every figure.

General manager

Starts the day with last night's reviews and today's bookings, and can see what a guest has said or spent before without asking the team.

Events manager

Sees which events are short of a deposit, what was promised about dietaries and layout, and what the organiser last wrote.

Marketing

Traces a campaign to a week's trade rather than to an open rate, and sees what the reviews said about it afterwards.

The work that comes round

The week in a venue

  1. 01

    Every morning

    New reviews under four stars arrive with the reviewer's own words, so a bad night is answered the same day rather than discovered by the next guest.

  2. 02

    Through the week

    Enquiries and events are answered from the diary, the deposit and the thread together, and the reply is drafted where you have granted the mail action.

  3. 03

    Monday

    Takings and covers by site, last week against the week before, with the reviews and campaigns of the same week beside them.

  4. 04

    Over the season

    Guests, events and venues are anchored into memory, so what a regular likes and what went wrong at last year's Christmas party outlive the person who remembered them.

In practice

Ask it like this

Plain English, from chat, your editor, a scheduled job or your own agent. In agent mode each question also checks the other systems that know about the same people, companies and references, and the answer ends with a line per system consulted.

Memtro · agent mode
How did the Kings Road site do last week against the week before, and what did reviewers say?
Reading 5 systems…
  • Square
  • Stripe
  • Google Business Profile
  • Trustpilot
  • Meta Pages and Instagram

Every figure in the answer cites the record it came from, and the run ends with a line per system consulted.

  • ›Which events in the next 14 days have no deposit paid?
  • ›What have guests said about service at the Harbour hotel this month, across Google and Trustpilot?
  • ›Which newsletter campaigns this quarter brought the most bookings?
  • ›Which days last month took the most, and what was on at each site?
  • ›What do we know about the party booked for Saturday: who booked it, what is paid and what did they ask for?
  • ›Which enquiries this week have had no reply, and what did each want?
  • ›How do this month's takings compare with the same month last year, by site?
On a schedule

Work that runs itself

A job is a prompt the agent runs on a cron schedule as its owner, with their tools and connections. Say any of these in chat and the job exists; every run's report is kept in a chat you can question.

  • schedule:Every day 09:00: new reviews under four stars across every site, with the reviewer's words
  • schedule:Every Monday 08:00: takings and covers by site, last week against the week before
  • schedule:Every day 10:00: events in the next 14 days with a deposit outstanding, and the organiser's last email
  • schedule:Every Monday 09:00: enquiries received last week with no reply, by site
  • schedule:First of the month: takings and review scores by site against the same month last year
How jobs work
Questions

Questions people ask

Can it reply to reviews?+

Not on the review sites today. It finds them, summarises them and drafts the reply in Gmail where you grant that action; as connectors gain actions, each is granted by name per team and runs only when a person asks.

Does it work for a group with several venues?+

Yes. Each site's connections are granted to the people who run it, and a group question is answered across all of them with the site named on every figure.

Does it connect to our booking system?+

Google Calendar and Eventbrite are covered. A booking system with a reporting database is reachable over Postgres or MySQL, and bookings that arrive by email are read from the inbox.

Can staff ask from their phone?+

Yes. The chat is a web page, and Memtro answers in Slack and Microsoft Teams where those are connected, so a question from the floor gets the same cited answer the office gets.

Does it make sense for one site?+

Yes, and it is the same product: one connection per system rather than one per venue. A group simply has more sites named in the answer.

What about guest data?+

Facts are classified personal or shared as they are extracted, connections are granted to named teams, credentials are encrypted at rest, and a request to be forgotten removes a person from every store with a receipt.

Next

The same work, by team

The same connectors and the same memory, described through the job somebody does rather than the business they do it in.

Memtro for hospitality

Ask it about your own business

Connect the first two systems, ask the question you opened this page with, and see where the answer comes from. About five minutes.

7-day free trial · card on file, nothing charged until it ends · cancel any time