. If you don't have any matters yet, the **Create a New Matter** dialog is already open — go straight to the next step.
3. In **Input Matter Reference**, type `Practice — Tutorial`.
4. Click 
.
BriefTech opens the new matter on its 
tab. You should now see an empty document area with a **Drop files or folders here for upload** zone in the centre — your blank practice matter is ready.
***
## Step 2 — Load the sample files
Instead of hunting for documents to upload, we'll let BriefTech load a ready-made set for us.
On the empty upload screen, click the text link below the 
button: **Just exploring? Load sample files to try BriefTech now.**
BriefTech begins loading a sample set of around twenty documents of different types — PDFs, emails, and text notes. You should see the file list fill with documents as they load.
***
## Step 3 — Open the 
tab
As each document finishes processing, BriefTech automatically extracts a date and a description — no action required from you.
Open the tab. You should see a **Date** and **Description** filling in for each row as extraction runs in the background.
***
## Step 4 — Correct one detail
The AI is a strong start, but you're in charge. Let's refine a description so you can see how corrections work.
1. In the tab, pick any document row.
2. Double-click its **Description** cell. An inline editor opens.
3. Edit the text to improve or clarify it, then click outside the cell to save.
The cell updates immediately — there's no Save button. You should now see your revised description in that row, and it will stay that way: re-running the AI's analysis won't overwrite a value you've set by hand.
Open the 
tab to continue.
***
## Step 5 — Build the Document Index
The **Document Index** is the ordered index that becomes the spine of your bundle. The fastest way to start one is to apply a preset.
1. In the tab, you'll see your documents listed in a table.
2. Click 
in the toolbar.
3. Under **Apply bundle/index preset**, select **Bundle of Documents**.
4. Under **Generate index headings**, select 
.
5. Under **Sort all sections**, select **Date**.
The table updates instantly. You should now see tidy columns — a serial-number column labelled **Tab**, then Date, Description, Vol, and Page — with headings generated from your folder structure and documents sorted by date within each section. That's your Document Index taking shape.
{% hint style="success" %}
**Tip:** **Bundle of Documents** is a general-purpose preset — one of several. When you prepare a real matter you can pick the one that matches your court and document type. See [Document Index presets](/reference/index-presets.md) for the full list and what each configures.
{% endhint %}
***
## Step 6 — Generate the court bundle
Now we'll turn that index and the documents into a paginated PDF.
1. In the toolbar, click 
. The **Create Bundle** panel opens on the right.
2. Leave every bundle setting at its default.
3. Click 
.
A **Job Logs** window opens and shows the bundle being built. You can close it at any time — the bundle keeps building. When it's ready, BriefTech shows a **Bundles created successfully** message. While it runs, the 
icon in the top bar pulses; click it to open **Background Jobs** and check the job's status.
***
## Step 7 — Download your bundle
1. In the **Create Bundle** panel — the same panel you opened in Step 6 — find the completed bundle, listed below the button.
2. On the bundle card, click the 
download icon.
The bundle downloads to your computer. You'll see a card for each volume PDF and a **Complete Bundle (ZIP)** card containing them all — download whichever you prefer, then open it.
You should now see a finished court bundle: a table of contents listing the documents, and the documents themselves with continuous page numbers stamped across them and bookmark links to each document.
That's a real, court-style bundle — built by BriefTech from scratch, for your review and approval, in seven steps.
***
## What you've accomplished
You took an empty matter and:
* loaded a set of documents and let them process,
* watched the AI automatically extract dates and descriptions, then corrected one yourself,
* built a structured Document Index from a preset — with folder-based headings and documents sorted by date, and
* generated and downloaded a paginated court bundle.
You now know the shape of the whole BriefTech pipeline. Everything else is a variation on these steps.
***
## Where to go next
| Review an AI-Analysed Matter | A guided run through the Review stage — take a matter the AI has analysed and confirm, correct and refine its details until it's ready to index. | Review an AI-Analysed Matter |
| How a matter moves through BriefTech | The staged pipeline and the two workstreams that run alongside it. | How a matter moves through BriefTech |
| How AI analysis works: agents and playbooks | What agents and playbooks are, and what they extract. | How AI analysis works: agents and playbooks |