> ## Documentation Index
> Fetch the complete documentation index at: https://docs.hizura.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Quickstart
> Build your first prospect list in Hizura — from an ICP described in plain language to a workable table.
## Before you start
You need a Hizura account. The Free plan includes a one-time grant of 100 credits and no card, which is enough to build a real list.
## Build your first list
### Step 1: Create a table
Open Hizura and click **Create new table**, or use the keyboard shortcut:
* **Mac:** `Cmd` + `;`
* **Windows/Linux:** `Ctrl` + `;`
### Step 2: Name your table
An editable title field appears at the top. Give the table a name that describes the segment — for example, "EU fintech ICP". Press **Enter** to confirm and move to the input field.
### Step 3: Describe your ICP
Type it in plain language. State who you're targeting, any criteria they must meet, and every field you want as a column:
```text theme={null}
Find 10 B2B fintech companies.
Criteria:
1. Headquartered in the EU (compulsory)
2. Preferably Series A or later (optional)
For each company, extract:
- Website(e.g. https://example.com)
- Headquarters location(e.g. Berlin, Germany)
- What they sell
- Employee count
- LinkedIn company page(link)
```
Press `Cmd` + `Enter` (Mac) or `Ctrl` + `Enter` (Windows/Linux) to submit.
You don't have to write a full brief. Send a one-liner like "find me some fintech startups to sell to" and Hizura will interview you instead — see [Conversation](/features/conversation).
### Step 4: Confirm scope and price
If your request leaves anything open, the [clarifier](/features/conversation) asks a short batch of questions before spending credits — which fields you want, whether it should follow specific steps, and which criteria are compulsory.
The last question is always how many rows you want, with the real credit cost next to each option:
```text theme={null}
Roughly how many rows are you looking for?
5 35 credits
10 70 credits
As much as possible (~40 rows, up to 280 credits)
```
Hizura then shows you the assembled brief. **Edit it if anything's off** — reword a criterion, drop a column, add a `Steps:` block — because what you approve is what runs.
No credits are spent until you approve the brief, and even then you're only billed for cells the run actually delivers.
### Step 5: Watch the run
Hizura streams its progress while it works — planning, searching the web, visiting pages, and a running count of entities completed. Runs take several minutes.
You can safely reload the page or close the tab. The run continues on the server, and Hizura reconnects you to it when you come back.
### Step 6: Read the results
When the run finishes, a summary strip reports what you got and what it cost:
```text theme={null}
Delivered 18 rows · 47 credits spent · Found everything you asked for
```
If you'd rather not see it, the `/summaryVisibility` command hides it — and keeps it hidden for future runs until you turn it back on. It's a per-person setting, so hiding it doesn't take it off a teammate's screen.
Your [table](/concepts/tables) has a row per entity, a column per field you requested, and sources on every researched value. Click a cell, or press `Cmd` + `0`, to open the citation panel and see which pages the value came from.
### Step 7: Work the list
A list you can't action isn't worth much, so finish the loop:
* **Add contacts.** Select the rows worth pursuing and run [`/enrichEmails`](/features/emails) to get verified work emails for them. You're billed only for addresses actually found, so enrich the shortlist rather than the whole table.
* **Move them through stages.** Switch to [the pipeline](/features/pipeline), set up steps that match how your team sells, and drag leads across as they progress.
* **Take it elsewhere when you need to.** Export to CSV or XLSX, or read the table through the [API](/api) or [MCP](/mcp).
#### Bring in your team
Share the table from the copy menu in the header — pick **read-only** or **can edit**, then send the link. Every plan includes unlimited team members, so this isn't metered.
Your teammates share the work, not your screen: everyone sees the same rows, the same pipeline steps, and the same card placements, so moving a lead to "Contacted" is visible to the whole team. Column widths, which view you're in, and card density stay yours alone — nobody's layout moves when someone else reorganises theirs.
Credits stay personal too. A teammate running their own research on a shared table spends **their** balance, not yours, and doesn't block you from running something at the same time.
See [Sharing & exporting](/features/sharing) for access levels and how to revoke one.
#### Hand it to an agent
Everything above is something an agent can do for you. Connect the [MCP server](/mcp) and your assistant works the same table you're looking at — topping the list up with accounts you don't have yet, qualifying them against your criteria, and advancing them through your stages.
That's the point of keeping the pipeline here rather than in a spreadsheet: your agents can reach it. Give one the rest of its toolbelt and the work between stages gets handled too — the outreach sent, the call booked, the card moved — while you review what happened rather than doing it.
Only `create_table` spends credits. Reading, qualifying, boarding, and re-sorting are free and instant, so there's no cost to letting an agent keep a list organised.
Boards an agent builds are shared, so you and your team see the result.
## Next steps
How the interview works, and how to edit the brief before it runs.
Understand per-cell billing and why a run stopped where it did.
Turn the table into a kanban board and work the rows.
Add verified corporate emails to selected rows.
Access levels, collaborators, and exporting your lists.
Let an assistant build and sort your lists from the chat.