Every screen in the app, and what each one does.

Two people use this app, and they use the same one.

The owner runs the shop. She gets four tabs: Desk, Shop, Reports and Account. If she also holds a chair, she can switch to her own day and back without signing out.

A stylist works a chair. She gets four tabs too: Today, Schedule, Reports and Account. Same app, fewer tabs, and nothing about anybody else's chair.

You never choose which of these you are. The shop you were added to decides it, and the app draws the tabs to match.

Each screen gets a picture and a short explanation. If you came here to do one particular thing rather than to read, `tasks.md` walks the common jobs end to end. The pictures come from a real shop running on a phone, with a real day in it: nine stylists, three services, a queue at the counter and three appointments on the book. Names and numbers are made up; everything else is the app as it ships.

Where a screen does something that is easy to get wrong, there is a line beginning Watch out. Those are the ones worth reading twice.

A walk-in is somebody standing in the shop with no appointment. They join the queue, they get a ticket number, and somebody seats them when a chair frees up.

An appointment is a booked time. It has a client, a service, a start and an end, and the client gets a text before it.

The counter holds both, in one list, in time order. That is the whole idea.

This is the first thing the app shows, and it is where signing out returns you.

Sign in is for anybody who already has a chair at a shop on Chaise. That is almost everybody: the owner adds you, then tells you to install the app.

Request to join is for somebody with no shop yet. It does not create an account.

The EN pill in the top right sets the language before you have signed in, which matters because the next screen is in whichever language you pick.

English and French, and the choice sticks to this phone. Signing out does not undo it, so you meet this screen once.

You can change it later under Settings without touching your sign-in.

Tapping Request to join opens this. Three fields, no password, no account: it is a message asking somebody to call you back.

No email address is collected because the reply comes by phone.

Send request stays grey until all three fields are filled and the number is a complete one. A half-typed number cannot be called back, so the button does not pretend otherwise.

You get a reference. Keep it: quoting it is faster than re-describing the shop when somebody rings.

Nothing was created. There is no account behind this and no password to remember.

Your mobile number is your account. There is no password on this product at all.

The small print under the field is the consent line for the one text you are about to receive. Marketing messages are a separate opt-in and never ride on a sign-in.

The field formats as you type, so `5145550001` becomes `(514) 555-0001` on its way in. Send code wakes up once the number is complete.

The code lands by text in a few seconds. Change goes back to the number if you fat fingered it.

The resend counter stops you asking for a second code before the first has had time to arrive.

Type the six digits and tap Verify. Five wrong guesses kills the code and sends you back to the number, so take the extra second.

If you hold a chair at more than one shop, a picker appears here asking which one you are signing into. With one shop the app skips it.

The first time you sign in after an owner adds you, the app asks who you are before it shows you anything. Three steps, and the bar at the top says where you are.

This name is what your shop sees and what clients see on a booking.

Not you? Sign out is there for the case where somebody typed the wrong number when they added you.

Date of birth and gender. Your shop keeps these with your record; clients never see them.

A wheel rather than a typed date, because it cannot produce the 31st of February.

Continue lights up once both are set.

Your home address, not the shop's. Clients never see it.

Start typing and the app suggests real addresses. Picking one fills the city, province and postal code for you, so there is nothing left to mistype.

City, province and postal code are greyed because the address was verified and is kept exactly as matched. Enter it by hand is the way out for an address the lookup does not know.

Finish lands you on your own day. The registration screens never come back.

Everyone still waiting. Appointments show their booked time; walk-ins show when they arrived and carry a ticket number on the right.

The number is an identifier, not a rank. Ticket 4 is not necessarily fourth in line, and the list order is the answer to who is next.

Everyone in a chair right now. Empty here means exactly what it says.

Somebody lands on this tab when the stylist taps Start, not when the owner does anything.

How the day ended for the people who have finished with it: served, left, no-show. Rosalie came in, waited, and went, so her row says Left.

Tapping a row restates it: the time, the service, and how long they have been waiting. The row truncates that last part; the sheet has room for it.

Reserve is the only way into the queue from a screen. Both fields are optional, so an unnamed person who has not decided what they want is still a row on the counter.

The shop's catalogue, with how long each one takes. The duration is what the wait estimate and the timeline are built from.

Name and service are set. Next asks the one question that decides what this becomes.

Now puts them in the queue: they are standing at the counter.

Anything else on the timeline makes it a booking. The grey blocks are the day's existing appointments and the red line is the current time, so you can see what a new slot would sit between.

Tap the timeline and the app lays the service's own length on it. Here a 45-minute cut lands at 5:26 PM and ends at 6:11 PM.

The sentence under it says plainly what is about to happen: this will be a booking, not a place in the queue.

You land back on the Desk with the sentence at the top and the new row in the list. The client gets a confirmation text with the time and the link they can change it with.

Five things, and the first four are the shop's own record: who works here, what you sell, when the door is open, and the code on the wall.

The fifth is red and at the bottom, on its own, away from anything you might tap by mistake.

Everybody with a chair here, by name, with what each of them offers underneath. Nothing on this screen ranks anybody.

The whole record for one member of staff on one page: when they started, whether they are bookable, what they offer, the week they work, and how to reach them.

Remove from staff at the bottom opens a screen of its own rather than doing anything here.

A name and a mobile number is the whole form. They get a text, they install the app, and the three screens in Part 1 ask them the rest.

Set a last day and the screen tells you what is standing behind it: how many days are booked after it, how far out the furthest one is, and how many appointments and clients are affected.

If those numbers are not zero, deal with them before you confirm.

Everything the shop sells, how long each takes, how many of the team offer it, and the price. Colour is offered by 5 of 9; everybody cuts.

Names in both languages, minutes in the chair, and the price. Leave the French name blank and French pages show the English one.

One price per service, and everybody who offers it charges it.

Switch a person on or off for this service. Camille is greyed with a reason: somebody has already booked this service with her, so it cannot be switched off underneath them.

A name and a length. Nobody offers a new service until you switch them on, which is the line under the fields.

The room's week. Each person sets their own week inside these, so a stylist can never be booked at an hour the shop is shut.

Closed right now takes the shop off the market immediately without changing the week. Appointments already on the book still stand.

A day can hold more than one window, which is how you draw a lunchtime close. Switch the day off entirely and it reads Closed on the week.

Print the poster and put it where people wait. Somebody scanning it joins the queue from their own phone instead of asking at the counter.

Allow "No preference" lets them say they will take whoever is free first. Switch it off and every self-serve walk-in has to name a stylist.

One screen, today, and it says what it will cost before it asks you to confirm: future bookings that will need explaining to clients.

You type the shop's name to confirm. Past visits are kept; the page goes and nobody can book.

Today first: what is in your chair, and what is in the shop. Then a period, and what happened in it.

Bookings, services completed and cancellations are counted on the day of the appointment, not the day somebody booked.

The grid is the shop's hours with the booked ones shaded, so a quiet Tuesday morning is visible without reading a number.

How full the day was measures against the shop's hours as they are set today. Change the opening hours and this figure moves, which is worth knowing before you read too much into it.

Listed by name. Nothing here is ranked, and the figures are against this shop's own past rather than against other salons.

Clients who came back stays silent until there is enough history to make a rate out of, rather than printing a number built on four visits.

Your own record, not the shop's. The card at the top is the switch between the shop and your own chair, and it is only there for an owner who holds one.

Everything you entered when you first signed in, each on its own row.

Clients see your name. Nothing else on this page reaches them.

One field to a screen, with a line saying who sees it. Changing your name changes it everywhere, including on bookings already made.

The days and hours you work. Clients only ever see the overlap between this and the shop's own hours.

I'm off right now stops anybody booking you from this moment. Appointments you already have stay exactly where they are.

Same shape as the shop's day screen, pointed at your week. Add a second window for a split day.

What you offer and what it costs. The shop sets these, so the screen shows them and does not let you change them. Ask the owner.

Owners only. Every shop the business runs, with whether each is open.

Rename it or re-address it. A verified address keeps its city, province and postal code as matched, and Enter it by hand is the escape hatch.

A second location, with its own address, its own time zone and the option to copy the first one's hours.

Language, what your clients are told, and signing out.

Sign out of this phone says whether anything is still waiting to send before you do it.

The reminder the day before is the one you control. The four beneath it always go, and each line says why.

A cancellation text is the one message a cancelled booking must still send, which is why it is not a choice.

The language the app answers in. It does not change what your clients are texted; that follows the language each client chose.

Your day and nobody else's, in the same three tabs the counter uses. The line above them answers the question you opened the app to ask: is anybody waiting for me.

Reserve is here too, so a stylist can put somebody in the queue without going to the counter.

Unlike the counter's sheet, this one has buttons, because this is your chair.

Start begins the service. Reschedule opens the timeline. Open the appointment goes to the full page.

Once started, they move here and the box on the left shows the booked length. The row says when you started.

The sheet now offers one thing: finish. It also shows how long they have actually been in the chair, which is the number the report compares against the booked one.

Served, with the time you finished. The client's history and your report both pick it up from here.

Everything about one booking: the service, the day, the time and the price. The badge at the top says this is a first visit.

Move this appointment keeps the same booking and the same reference, so the client's link still works.

Cancelling is free and texts the client. If you only want to change the time, move it instead.

How many times they have been to you, when they last came, what they usually have, and your own notes. A first-timer reads exactly like this one.

Book them again starts a new booking with them already filled in.

The week across the top, the day underneath. The red block is the appointment you are moving; the grey ones are what else is on that day.

Drag it or tap a new time, then save. The client is texted so nobody turns up at the old time.

Your week, one day at a time. The dot under a date means something is on it, and the line across the timeline is now.

Days the shop is closed say so rather than drawing an empty grid.

Your own numbers, over whichever period you pick. A booking lands here once it is complete, and each row shows what was booked against what was recorded.

The figure is the whole slot, processing time included.

The same Today screen, with the same three tabs, drawn for somebody with nothing on yet.

Four rows instead of five. No Shops, and no switch to the counter, because there is no counter on this side of the door.

The line under your name says Your chair rather than Owner. That is the whole difference in the app, and it is decided by the shop, not by you.