Use cases · examples by sector
Use cases that hold up in daily work
This is what AI automation can look like in four sectors: services, hospitality, retail and practices. Each example comes with what stays with people.
Examples, not client projects·Practices: organisational workflows only
01Sectors
Use cases by sector
Pick your sector and see what can be taken over and what stays with your team.
Services
Have enquiries, quotes and appointments prepared for you
Agencies, consultancies and offices: much of the work on enquiries and commitments follows the same pattern and can be prepared before a person looks at it.
Stays with peoplePrices, commitments and the tone towards customers. Quotes only go out after approval.
Example ideas, not client projects
- Sort enquiries from email and forms and prepare a draft reply
- Create quote drafts from a template and a short briefing
- Send appointment confirmations and reminders automatically
- Reconcile timesheets and invoice data
Hospitality
Reach guests even when the house is full
Calls, bookings and the same questions again and again often arrive when nobody has time. An assistant takes over the routine.
Stays with peopleSpecial requests, complaints and the decision on tables, rooms and prices.
Example ideas, not client projects
- Take booking requests by phone or chat outside opening hours
- Answer questions about opening hours, directions and the menu
- Review ratings and prepare draft replies
- Send confirmations and information before arrival
Retail
Orders and customer questions without the typing
In retail, data moves between suppliers, inventory systems and customer service. The less of it runs by hand, the fewer typing errors creep in.
Stays with peopleGoodwill decisions, complaints and price approvals.
Example ideas, not client projects
- Pre-sort questions about delivery status and returns and answer them with prepared replies
- Read delivery notes and invoices and enter them into the inventory system
- Draft product texts from supplier data
- Reconcile order confirmations with stock
Practices (no health data)
Relieve the organisation without touching sensitive content
Here we automate organisational workflows only. Health data stays out, and a person handles professional content.
Stays with peopleProfessional judgement, advice and anything that concerns the content of individual cases.
Example ideas, not client projects
- Organise appointment requests and rescheduling
- Send appointment reminders
- Capture callback requests with name and phone number and pass them on
- Request documents and track their receipt
02Scenarios
Three workflows, before and after
They show where AI prepares and where a person decides.
Example scenario
- Before
- Appointment requests arrive by email in many wordings. Someone reads them, checks the calendar and replies by hand.
- After
- The request is recognised, free slots are suggested and a draft reply is ready.
- Stays with a person
- Your team checks the draft and sends it; special requests are decided by a person.
- Before
- Calls and chat messages outside opening hours are left waiting, and the next morning each has to be answered individually.
- After
- An AI assistant answers standard questions, takes down requests and files them sorted.
- Stays with a person
- For anything beyond standard questions, the assistant hands over to your team.
- Before
- A quote is built by hand from an enquiry: find the template, transfer the data, adapt the text.
- After
- Outlook, SharePoint and Power Automate create a quote draft with the right data from the enquiry.
- Stays with a person
- Sales checks the draft, changes it if needed and approves it.
03Criteria
How to recognise a good use case
In the AI check we examine four things before we recommend anything.
01It recurs regularly
A workflow that runs daily or weekly carries the effort better than a rare one.
02It follows recognisable rules
Where people decide in similar ways, preparation works. Where every case is different, it stays manual.
03The inputs are digital
Email, form, document or record can be processed. Handwritten notes are harder.
04A person can check the result
We only build what your team can see, change and approve before it goes outside.
Frequently asked questions
Questions about use cases
The four sectors are our focus. Whether a specific workflow fits your business, we clarify in the AI check using your own workflows.
AI check
Which workflow should come first?
Tell us briefly where the work gets stuck. We will get back to you and clarify whether an AI check makes sense.