Build your first collection

Requirements: a business you can work in, and a role that can create collections and fields. About ten minutes.

Your business came with a starter setup, but the real value shows when you model something specific to your work. In this tutorial you will build one collection end to end — Work Orders, the jobs a small repair shop takes in — and add the first record to it. The subject is an example; the pattern is the point. Once you have done it once, every other collection you build follows the same five steps.

Step 1 — Decide what one entry represents

Before creating anything, answer one question: what is a single entry in this collection? For our example, one entry is one job — one customer bringing in one item to be repaired. Not a customer (they might return next month with something else), and not an invoice. Say it out loud as a sentence: "one entry is one job."

This is the step people skip, and it is the one that decides whether the rest goes smoothly. Get it right and every field is obvious. Get it wrong and you will find yourself squeezing two different things into one collection.

Result: a one-sentence definition. Nothing created yet.

Step 2 — Create the collection

Create a new collection and name it for what it holds — Work Orders. Plural, because it holds many; named in your own language of work, not a generic label like "items."

You do not need to set up housekeeping fields: every collection automatically gets what it needs to identify and publish an entry. What you add next is only what is particular to your work.

Result: an empty collection, ready for fields.

Step 3 — Add typed fields

Now add the details each job carries. For our work orders, four are enough to be useful:

• Description — text: what needs doing.
• Amount — a number: what the job costs. A number, not text, so you can total and average it later.
• Scheduled date — a date: when it is booked in. A date, so it can drive a calendar.
• Priority — a set of options: normal, urgent. Fixed choices, so everyone uses the same words and you can filter cleanly.

The type is the important part of each decision, and it is worth a moment's thought: a field's type is fixed once created, so choosing "number" now saves you migrating data later. Resist the urge to make everything a text box.

Result: a collection that can describe a job — but not yet say whose it is.

Step 4 — Link it to the customer

A job belongs to someone. Rather than retyping the customer's details on every work order, add a reference field pointing at your Contacts collection, and call it Customer.

This is the step that turns a list into a model. The work order now points at a contact rather than copying it, so when that customer's phone number changes, every job of theirs is correct at once. Each job has exactly one customer, so a single reference is right here — the reasoning behind that choice is in reference vs multi-reference.

Result: a collection connected to the rest of your data.

Step 5 — Add your first entry

Create an entry and fill it in: a description of the job, the amount, the date it is booked for, its priority, and the customer it belongs to. For the customer, you pick an existing contact rather than typing a name — that is the reference doing its job.

A new entry starts as a draft, which is exactly right while you are trying things out: it is real and yours to work with, but not exposed publicly until you publish it.

Result: a real record in a collection you built.

What you built

You now have a working part of your CRM: a collection whose shape matches your actual work, with typed fields that can be sorted, filtered, and totalled, connected to your customers, holding a real record. Everything else in Sonta builds on this — a view to see these jobs as a board, an automation to email the customer when one is done, a report to total the month.

Where to go next

Add stages to your collection so jobs move from booked to done, and shape a view around them. The concepts behind what you just did are in the data model, and the fuller treatment of building collections is in working with collections and fields.