Write a Technology Brief for Your First Booking Tool or Client Portal
Tom
Digital Business & Systems Consultant · Singapore

A requirements document for a booking system or client portal can fit on one page: who does what today, who will use the tool, the three to five tasks it must handle first, who can see or change what, what data it keeps, how you will test it, and what is left out. Use that page to check whether a manual method or a ready-made tool already covers the need before paying for custom work. In a service business, the first version is the smallest workflow that removes your main obstacle.
Key takeaways
- Start from today's workflow and its main obstacle, not a feature list.
- Limit the first version to three to five tasks; the rest goes on a "later" list.
- Write acceptance criteria as "a customer can …" or "staff can …", testable with a yes or no.
- Test the brief against a manual method and a ready-made tool before asking for a custom quote.
Describe who does what today
Write the current workflow in plain steps: who asks, who answers, where details are recorded and where things slip. Include people a feature list forgets, such as the part-time staff member who checks the diary.
Then name the main obstacle in one sentence, from what you have seen: "[Who] loses [what] because [where it slips]." Not sure customers want the service yet? Test the idea first; a brief cannot fix missing demand. Unsure which steps cost the most time? Log them for a few weeks with this task-volume and error worksheet.
Check manual and ready-made options first
Custom development is the last of three options. A form, chat and a shared sheet can be enough at low volume, and a ready-made booking or portal tool often covers standard slots and reminders. This manual, off-the-shelf or custom checklist shows where you are.
Your brief makes the comparison fair. Try each acceptance criterion below in the trial or demo of one or two ready-made tools, where one is offered. Note which pass, which need a workaround, and what each workaround costs in weekly time.
If a ready-made tool passes every criterion that matters, use it. The brief still showed you what you need.
A custom tool is worth discussing when a criterion that matters keeps failing and the workaround costs time every week. I build that as a web app development project. If the problem is an internal workflow scattered across WhatsApp and spreadsheets, operations digitalisation may fit better.
Define the smallest useful workflow
In a service business, an MVP (minimum viable product) is the smallest workflow that removes the main obstacle, used by real customers or staff. It is not a demo, and not version one of everything.
List every feature you want, then ask of each: does the main obstacle remain without it? If not, it goes on the "later" list. Aim for three to five tasks; payments, reminders and reports wait unless they are the obstacle.
The one-page requirements brief
Copy the eight sections into a document and fill in the middle column. The third column is a fictional example: Jasmine runs a small pet grooming shop with one part-time groomer, Shafiq. Bookings arrive by WhatsApp, Instagram and phone.
| Section | What to write | Example (fictional): a grooming shop |
|---|---|---|
| Problem today | Today's steps, where they slip, the main obstacle | Bookings from three channels go into a paper diary; slots get double-booked; care notes live in Jasmine's head |
| Users and roles | Everyone who will log in | Pet owners, groomers, and Jasmine as owner |
| First-version tasks (3 to 5) | Tasks without which the obstacle stays | Customer adds pets and books; customer cancels; groomer sees the day's list with care notes; owner sets hours and blocks dates |
| Permissions | Who can see or change what | Customers see only their own pets and bookings; groomers edit care notes, not prices or hours; the owner changes everything, including staff logins |
| Data kept and who owns it | What is stored, why, and who can export it | Contact details, pet details, care notes, booking history; no card details; the shop owns and can export the records |
| Acceptance criteria | Yes-or-no tests: "a customer can …", "staff can …" | See the list below |
| Not in this version | Exclusions, so nobody assumes them | Online payment, SMS reminders, loyalty points, an app-store app, a second outlet |
| Handover | What you receive; who holds each login | Owner and staff logins, a how-to guide, a data export, and written support terms |
Sample acceptance criteria
Each criterion should be testable by someone who did not write it. From the fictional brief:
- A customer can add two pets and book a slot for one of them on a phone, and sees a confirmation.
- A customer cannot see another customer's pets, notes or bookings.
- A customer can cancel up to 24 hours before; after that, the screen shows the shop's phone number.
- Two customers cannot book the same groomer for the same slot.
- Staff can see today's appointments with care notes, but cannot change prices or hours.
- The owner can block a public holiday, and it disappears from the booking screen.
- The owner can export customers, pets and bookings to a spreadsheet.
Decide the data row carefully
A portal that holds individual customers' details holds personal data. PDPC's advisory guidelines on key concepts in the PDPA (revised 29 April 2026) say an organisation "must obtain the consent of the individual before collecting, using or disclosing his personal data for a purpose". Note why you need each field and leave out the rest. This is not legal advice; check your own obligations.
Agree evidence that the tool works
Acceptance criteria turn "is it finished?" into a list you can tick. Agree them before the quotation, so price, build and final check all refer to the same page.
- 1Attach the criteria to the scope, and agree who runs the final check.
- 2Test with sample data, never real customer records.
- 3Mark each criterion pass or fail. A fail against the agreed scope gets fixed; a new idea becomes a separate request.
- 4Decide on go-live yourself once the criteria pass and you hold the handover items.
For who handles what, and what you hold at handover, see working with BusinessWithTom as your technical delivery partner.
How the fictional brief played out
(Fictional example) Jasmine wrote her brief over three evenings; asking Shafiq what he needed on a busy morning added the care-notes task. Two ready-made booking tools handled slots, cancellations and holiday blocks, but neither showed per-pet care notes on the groomer's day list, so she kept a shared notes sheet as a daily workaround.
The upside: she knows exactly what a custom tool must do, and can compare quotes line by line. The effort: about ten hours of writing, trials and logging. The uncertainty: one failing criterion may not justify a custom build, so she will run the workaround for two months and weigh its weekly time against any quotation.
Your next step
Write the brief this week and test it against your current method and one ready-made tool. If a criterion that matters keeps failing, message me the brief. I start each web app project with a requirements workshop, and scope and quote it individually, with a breakdown by milestone.
Sources (checked 31 October 2026)
Frequently Asked Questions
What is an MVP in a service business?
The smallest working version of a tool that removes your main obstacle for real customers or staff. For bookings, that might mean booking, cancelling and a daily list, with payments and reminders later.
How do I avoid asking for every feature at once?
Ask whether the main obstacle remains without it. If not, add it to a written later list, and review that list once the first version is in use.
Do I need technical knowledge to write a requirements brief?
No. Describe the people, tasks and rules in plain language. Technical choices are worked out during scoping.
What if a ready-made tool meets most of my criteria?
Use it if the missing criteria matter little or the workaround is quick. Keep the brief; it shows when the workaround stops being worth it.
Can the brief change after I get a quote?
Yes, but changes can affect the scope and price. Agree how changes are handled before work starts, and treat new ideas as separate requests.
Written your one-page brief, and a criterion that matters still fails with your current method and a ready-made tool? Message me the brief, with sample data rather than real customer records. We can talk through what a first version would cover and how the project would be scoped. It is a conversation with no commitment, and if a ready-made tool fits, I will say so.
No obligation. No sales pitch. Just an honest conversation.