> For the complete documentation index, see [llms.txt](https://help.filed.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.filed.com/guides/getting-started/your-first-client-end-to-end.md).

# Your first client, end to end

This page walks through a complete return for one client using all the Filed products in sequence. By the end, the client's documents are organized, questions are answered, the draft is reviewed, and data is entered into tax software.

Prerequisites: a workspace is set up and a client has been created with at least one document uploaded. See [Adding your first client](/guides/getting-started/adding-your-first-client.md) if you haven't done this yet. Allow 30-60 minutes for a first run on a typical 1040.

## Before you start

The client hub should be open, with the product tiles visible: Binder, AI Tax Prep, AI Review, AI Tax Planning, and Chat. If Filed is still processing the uploaded documents, the Binder tile shows a processing state; wait for it to complete before starting AI Tax Prep.

## Step 1: upload the client's documents

Source documents reach Filed through one of three paths:

* **Manual upload**: drag-and-drop in the Add Client flow or directly into the Binder.
* **Document management integration**: Google Drive, SharePoint, SmartVault, Box, or Dropbox.
* **Filed Printer**: a virtual printer that sends documents to Filed from any application on your desktop.

A typical 1040 document set includes the prior-year return (required), W-2s, 1099s, K-1s, organizer or workpapers, and any supplemental schedules. See [Best practices for preparing your documents](/guides/getting-started/best-practices-for-preparing-your-documents.md) for type-specific guidance.

**Walkthrough: upload documents in the Add Client flow**

1. In the **Add Client** flow, on **Step 2 (Upload their source documents)**, drag the client's files onto the drop zone (or use **Browse files**). Each file shows an upload progress indicator, and ZIP archives are expanded automatically. Click **Create client** once at least one upload completes.

![The Add Client upload step with the document drop zone](https://storage.googleapis.com/filed-prod-public-assets/walkthroughs/adding-first-client/image2.webp)

2. Filed opens the new client's hub and begins building the Binder. The **Binder** tile first shows a processing state, then updates to reflect the organized documents - a missing-documents count and an observations count once binder creation completes.

![The client hub with the Binder tile showing document signals after processing](https://storage.googleapis.com/filed-prod-public-assets/walkthroughs/source-files/image1.webp)

## Step 2: review the organized Binder

The Binder is the per-client document workspace. Filed AI splits each uploaded file into individual subdocuments, classifies each by document type, issuer, and tax year, and places them in the Binder. A single uploaded PDF can produce multiple Binder entries.

The Binder has three tabs:

* **Docs**: all classified documents grouped by category.
* **Forms**: tax-prep values extracted by AI Tax Prep (empty until AI Tax Prep has run).
* **Leads**: AI Review leadsheets (empty until AI Review has run).

The layout is a three-panel view: document list on the left, document viewer in the center, and a Notes & flags panel on the right.

The Docs tab shows two key signals: a missing-documents checklist (document types expected for this return type that are not yet in the Binder) and Filed AI observations (notes and flags on documents needing attention).

**Walkthrough: review the organized Binder**

1. Open the **Binder** and stay on the **Docs** tab. The left panel lists the classified source documents grouped by category (for example, Income and Payments). Click a document to open it in the center viewer, with the **Notes & flags** activity panel on the right showing the AI's notes anchored to that document.

![The Binder Docs tab: document list, center viewer, and Notes & flags panel](https://storage.googleapis.com/filed-prod-public-assets/walkthroughs/binder-docs/image1.webp)

2. Click **Missing documents** at the top of the left panel to see the checklist of document types expected for this return type that are not yet in the Binder. Each item explains why it is expected and offers **Upload**, **Dismiss**, and **Comment** actions.

![The missing-documents checklist with per-item explanations](https://storage.googleapis.com/filed-prod-public-assets/walkthroughs/binder-docs/image2.webp)

## Step 3: ask Chat about the client

Per-client Chat is scoped to one client's full document set and return data. Every answer cites its source document so you can trace it back to the Binder. Chat is available from the Chat tile on the client hub and from within the Binder.

Chat is useful throughout the return, not only at the start. Common uses include looking up a specific figure, summarizing income sources before prep, checking what documents are still missing, and explaining complex documents like K-1s.

### Example 1: look up a specific figure or document

Prompt: "What were the total wages on the W-2?"

Chat returns the value with a citation to the specific W-2 in the Binder. You can click the citation to open the document at the relevant page.

### Example 2: summarize the client's tax situation

Prompt: "Give me a summary of this client's income sources this year."

Chat returns a structured overview of income types and amounts drawn from the classified documents in the Binder - useful as a pre-prep orientation before running AI Tax Prep.

### Example 3: check what is missing

Prompt: "What documents are still missing for this return?"

Chat reads the Binder's missing-documents checklist and lists the gaps, so you know exactly what to chase from the client before proceeding.

### Example 4: compare against the prior year

Prompt: "How does this year's income compare to last year?"

Chat compares figures across the current and prior-year documents in the Binder. This requires a prior-year return to be present, which is strongly recommended for every client (see [Before you start checklist](/guides/getting-started/before-you-start-checklist.md)).

### Example 5: explain a complex document

Prompt: "Explain the Box 20 codes on this K-1."

Chat interprets the document, explains each code's tax treatment, and points to where the relevant entries appear in the Binder.

### Example 6: find and jump to supporting evidence

Prompt: "Where is the charitable contribution recorded?"

Chat identifies the relevant document, states the amount, and provides a deep link into the Binder document at the exact page.

## Step 4: run AI Review

AI Review cross-checks a draft return against the source documents already in the Binder and surfaces findings ranked by severity: Critical, High, Medium, and Low. Findings appear as structured leadsheets in the Binder's Leads tab.

Before running AI Review, make sure the Binder contains all available source documents. AI Review uses whatever is in the Binder at the time of the run, so a more complete Binder produces better findings.

**Walkthrough: run AI Review**

1. Launch **AI Review** from the client hub tile. On **Step 1**, upload the draft return PDF, then click **Next: Confirm source documents**.

![AI Review step 1: upload the draft return](https://storage.googleapis.com/filed-prod-public-assets/walkthroughs/ai-review/image1.webp)

2. On **Step 2**, confirm the binder source documents look complete, then click **Start review**.

![AI Review step 2: confirm the binder source documents](https://storage.googleapis.com/filed-prod-public-assets/walkthroughs/ai-review/image2.webp)

3. AI Review runs (an in-progress loader is shown) and, when complete, opens a results summary with the total findings and a breakdown of Critical, High, Medium, and Low counts. Click **Open leadsheets** to work through each finding in the Binder's **Leads** tab. For the full results walkthrough, see [AI Review](/reference/products/ai-review.md) and [Leadsheets](/reference/products/binder/leadsheets.md).

![The AI Review in-progress loader](https://storage.googleapis.com/filed-prod-public-assets/walkthroughs/ai-review/image3.webp)

## Step 5: run AI Tax Prep

AI Tax Prep reads the organized Binder documents, extracts figures, maps them to the correct forms, and flags items for attention in Pre-Entry Notes. If a tax software integration is connected (CCH Axcess, Drake, Lacerte, or UltraTax CS), Filed can enter the prepared data directly into the return, with no re-keying.

Extraction and Pre-Entry Notes work whether or not a tax software integration is connected. The connection is required only for automated data entry.

**Walkthrough: run AI Tax Prep**

1. From the client hub, open the **AI Tax Prep** tile. Filed processes the Binder documents (a processing state shows while it runs) and produces a **Task result** page summarizing what it extracted: the forms detected, counted under **Forms**, **Documents**, and **Gaps**, with a **Forms extracted** breakdown (for example, W2, Taxpayer Info, Estimated Payments, Charitable Contribution).

![The completed AI Tax Prep task result with extracted form counts](https://storage.googleapis.com/filed-prod-public-assets/walkthroughs/ai-tax-prep-result/image1.webp)

2. Click **Review results in binder** to open the **Forms** tab, where you verify each extracted value against its source. The AI's Pre-Entry Notes appear in the Binder's **Observations** view (see [Pre-Entry Notes notifications](/reference/products/ai-tax-prep/pre-entry-notes-notifications.md)).

![The Observations list of AI Pre-Entry Notes](https://storage.googleapis.com/filed-prod-public-assets/walkthroughs/observations/image1.webp)

If a tax software integration (CCH Axcess, Drake, Lacerte, or UltraTax CS) is connected, the result page also offers a data-entry step to push the prepared values into the return.

## The final value

After completing this walkthrough, the firm has:

* An organized, classified Binder with all client documents accessible in one place
* Quick answers from Chat, each traced to a source document
* A reviewed return with findings ranked by severity and resolved or noted
* Prepared data entered into tax software, ready for final review and filing

Next steps worth exploring: the [Playbook](/reference/products.md) for encoding firm-level preferences, and [Plugins](/reference/products.md) for adding or adjusting integrations.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://help.filed.com/guides/getting-started/your-first-client-end-to-end.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
