Booking gets harder the moment a second person joins your team. Customers may want a specific stylist, therapist or adviser, or they may not mind who they see. Each person has a different calendar, skills, working days and speed. A business with five practitioners is not five times as complex as a business with one; it is a different problem: who is free, who can do this service, who does the customer prefer, and how do you avoid double booking anyone. Chatbot booking can handle some of this and not all of it, depending on the booking tool behind it. This guide explains the models for multi-staff scheduling, what each needs, which fit which kinds of business, how Crescelo's booking fits today and how to choose an approach that will not create chaos.
Crescelo's built-in booking picker offers slots from your opening hours and one slot length, hides taken slots and locks a slot while it is being booked. It does not assign bookings to individual staff or manage separate staff calendars. This article explains how to handle multi-staff businesses using per-person links and other approaches.
The scheduling models
| Modell | Så fungerar det | Good for | Watch for |
|---|---|---|---|
| Single pool of slots | One shared set of times; whoever is free serves the booking | Small teams where anyone can do the job | Customers cannot choose a person; capacity must be managed by you |
| Booking link per person | Each practitioner has their own booking page or link | Customers who choose a specific person | The chat must send visitors to the right link |
| Round robin | The system assigns the next available person | Teams serving similar customers | Requires a tool that supports it |
| Skills-based routing | The system matches service to qualified staff | Businesses with specialities | Needs service and staff data in the tool |
| Resource-based | Rooms, chairs or equipment are the limiting factor | Salons with chairs, clinics with rooms | Needs a tool that models resources |
Choose the simplest model that matches how your customers actually book.
What customers want
- A specific person: regulars often want the same stylist or practitioner.
- The earliest time: new customers usually want the first free slot, whoever it is.
- A skill: 'someone who does curly cuts' or 'a specialist in sports injuries'.
- Continuity: returning to the same person for follow-up.
Ask in the chat: 'Do you have a preference for who you see, or shall I find the earliest time?' The answer decides which route to use.
How to handle it with a simple booking tool
With a single-pool tool like Crescelo's built-in picker, you can still serve a team by combining approaches:
- Use the single pool for customers who do not mind. Set booking hours to times when at least one person is available, and assign the booking to whoever is free when it is made, using your dashboard and calendar feed.
- Use per-person links for customers who care. Give each practitioner a Calendly or Cal.com link, and have the assistant offer the right link when a customer names a person.
- Describe your team in the knowledge base. Who does what, when they work, and what each specialises in, so the assistant can recommend and route.
- Keep the rules in one place. A shared document with names, days and services prevents confusion.
This is manageable for small teams. For larger teams with complex rules, a dedicated scheduling system is the right home for the calendar, and the assistant's role is to answer questions and hand over to it. See AI website assistant vs Calendly and Acuity and configuring availability rules.
Stylist and practitioner preferences
In salons and clinics, preference is personal. The assistant can capture it and route accordingly. 'I would like to see Priya' leads to Priya's link. 'Anyone is fine' leads to the shared pool. 'Someone who does balayage' leads to the specialists' links. Store specialities in the knowledge base in plain words, and update them when staff change. See stylist preferences and chair allocation.
Avoiding double booking a person
- Each person's availability must live in one system, not two.
- If personal calendars are separate from the booking system, decide which is the source of truth.
- A read-only feed shows bookings in a personal calendar but does not make personal events block slots; adjust booking hours to reflect real availability. See AI bookings in Google and Outlook Calendar.
- Two-way sync tools reduce clashes; check what your tool offers.
Double bookings are avoided by locking a slot at the moment of booking. See how online booking avoids double bookings.
Capacity and fairness
- Decide how bookings are distributed: by preference, by rotation or by availability.
- Protect breaks and admin time in booking hours.
- Balance workloads so no one is overloaded.
- Handle absence: when someone is off, their link should close or redirect.
Fairness matters to staff, especially where income depends on bookings. Agree rules in advance and apply them openly.
What the assistant should say
| Customer message | Response |
|---|---|
| I want to book with Priya. | Sure. Priya has Tuesday at 10:00 and Thursday at 2:30. Which would you like? |
| Anyone is fine, earliest please. | The earliest available is tomorrow at 9:30. Shall I book it? |
| Who does colour? | Priya and Sam do colour. Priya works Tuesday to Friday and Sam on Saturdays. Would you like to see times with either? |
| Priya is not available. | I can offer Sam, who does the same service, or show Priya's next opening on Tuesday week. |
A rollout plan
- Write who does what and when, in a short document.
- Choose your model: shared pool, links per person or a dedicated tool.
- Set up each person's link or booking hours.
- Load the team information into the assistant.
- Test: a preference request, an 'anyone' request, a skill request and an unavailable person.
- Watch the first weeks and adjust.
Worked example: a five-person salon
A salon has five stylists. Regulars nearly always ask for their own stylist; new customers usually want the first available time. The owner gives each stylist a Calendly link, sets the shop's shared booking hours for anyone-will-do bookings and loads a short team document into the assistant: who does colour, who does barbering, who works which days. When a visitor asks for a person, the assistant sends that stylist's link. When the visitor says 'anyone', it offers the earliest shared slot, and the front desk assigns the booking each morning. Bookings for a named stylist go straight into that stylist's diary, and shared bookings are handed out by the front desk based on workload. The setup took an afternoon and reduced phone calls about who is free. It is not fully automatic, but it is reliable and simple to maintain.
Common mistakes
- Two systems holding the same person's calendar.
- Offering a person's link when they are on leave.
- Letting the assistant promise a specific person when only a pool exists.
- Forgetting to update the team document when staff change.
- Overloading the same stylist because of a popular link.
Frequently asked questions
Can a chatbot book with a specific staff member?
It can send visitors to the right person's booking link, or book from a shared pool. Crescelo's built-in picker does not assign staff, so use per-person Calendly or Cal.com links for preferences.
How do I handle a customer with a preferred stylist?
Capture the preference and give the stylist's link, and store each person's schedule and specialities in the knowledge base.
Can it do round-robin?
Only if the booking tool supports it. Crescelo's built-in picker uses a single pool.
How do I stop double booking staff?
Keep availability in one system and use a booking tool that locks slots.
What if someone is off sick?
Close or redirect their link and update booking hours.
When should I use a dedicated scheduling system?
When you have many staff with complex rules, resources or capacity. The assistant can still answer questions and hand over.
See the appointment booking feature, read AI appointment booking and try a booking conversation in the playground.
Further reading
- Google Calendar API overview: How calendar events are read and written by software.
- Google Search Central: creating helpful, people-first content: What Google looks for in useful pages.
SVARA ALLA BESÖKARE. BOKA DEM SOM ÄR REDO.
Lägg in Crescelo på din webbplats på 15 minuter
Träna din assistent på dina prislistor, dokument och tjänstevillkor. Aktivera välgrundade svar dygnet runt och bokning av tider direkt i chatten utan krockar.