---
title: "How to Export Data from Help Scout: Methods, API Limits & Formats"
slug: how-to-export-data-from-help-scout-methods-api-limits-formats
date: 2026-04-15
author: Raaj
categories: [Help Scout]
excerpt: "Help Scout's UI only exports reporting metrics — not conversations. Learn every method to extract full threads, attachments, and customers via the API, ETL tools, or migration services."
tldr: "Help Scout's native export gives you reporting CSV/XLSX only — no message bodies. Full conversation data requires the Inbox API (OAuth, 48-hour tokens, plan-based rate limits) or an engineer-led extraction."
canonical: https://clonepartner.com/blog/how-to-export-data-from-help-scout-methods-api-limits-formats/
---

# How to Export Data from Help Scout: Methods, API Limits & Formats


Help Scout does not have an "Export Everything" button. The built-in UI export covers **reporting metrics only** — conversation volume, response times, happiness scores — delivered as CSV or XLSX. It explicitly excludes conversation thread contents, message bodies, and satisfaction rating comments. ([docs.helpscout.com](https://docs.helpscout.com/article/466-data-export-options))

If you need the actual text of your support conversations, customer records, attachments, or knowledge base articles, you must use the API or a third-party tool.

The gap between what the UI gives you and what you actually need is enormous. The UI export is fine for a quarterly board deck. It is not a migration tool.

For a **migration-ready export**, treat Help Scout as three separate data surfaces: report exports in the app, inbox data in the **Inbox API 2.0**, and knowledge base content in the **Docs API**. The Inbox API is JSON over HTTPS with OAuth 2.0. The Docs API uses per-user API keys over HTTP Basic auth. ([developer.helpscout.com](https://developer.helpscout.com/mailbox-api/))

This guide covers every extraction method available, what each one actually returns, and the exact rate limits, pagination rules, and authentication constraints you will hit along the way.

If you are exporting specifically to migrate to another platform, these destination-specific guides cover field mapping and import constraints: [Help Scout to Zendesk](https://clonepartner.com/blog/blog/help-scout-to-zendesk-migration-guide/), [Help Scout to Freshdesk](https://clonepartner.com/blog/blog/help-scout-to-freshdesk-migration-guide/), or [Help Scout to Intercom](https://clonepartner.com/blog/blog/help-scout-to-intercom-migration-guide/).

## What Data Can You Export from Help Scout?

Before choosing an export method, understand what Help Scout considers exportable and which extraction path covers each data type.

| Data Type | UI Report Export (CSV/XLSX) | Inbox API 2.0 (JSON) | Docs API (JSON) | Support Request (CSV) |
|---|---|---|---|---|
| **Conversation metadata** (subject, status, dates, assignee, tags) | ✅ | ✅ | — | — |
| **Conversation thread contents** (message bodies, replies, notes) | ❌ | ✅ | — | — |
| **Attachments** (files on threads) | ❌ | ✅ (separate download per attachment) | — | — |
| **Customers** (name, email, phone, properties) | Partial (in report rows) | ✅ Full objects | — | ✅ Names & emails only |
| **Custom fields** | ✅ (Plus plan and above) | ✅ | — | — |
| **Tags** | ✅ (included in conversation rows) | ✅ | — | — |
| **Satisfaction ratings** | ✅ (score only, no comments) | ✅ (score + comments) | — | — |
| **Knowledge base articles** | — | — | ✅ | — |
| **Users / agents** | Partial (in report rows) | ✅ | — | — |
| **Workflows / automations** | ❌ | Partial (listable, but full condition/action definitions not documented for export) | — | — |
| **Saved replies** | ❌ | ✅ (listable, `includeChatReplies=true` for chat replies) | — | — |

> [!WARNING]
> **The Free plan does not include API access.** You must be on a paid plan (Standard or above) to use the Inbox API. The Docs API uses a separate API key.

The key takeaway from this table: there is no single method that extracts everything. A complete export requires combining at least the Inbox API (for conversations, customers, and users) with the Docs API (for knowledge base articles).

## Method 1: UI Report Export (CSV / XLSX)

**What it is:** A built-in export available from the Help Scout Reports dashboard. Select a report (All Channels, Email, Company, User, or Happiness), apply date and tag filters, and click the Export/Print icon. ([docs.helpscout.com](https://docs.helpscout.com/article/849-export-reporting-data))

### Step-by-step

1. Navigate to **Reports** in Help Scout.
2. Select the report type (e.g., All Channels, Email, Company, Happiness).
3. Apply filters: date range, mailbox, tags, or custom views.
4. Click the **Export/Print** icon near the date filter and choose **Export**.
5. Confirm the date range and select the format: **CSV** or **XLSX**.
6. Help Scout queues your request and emails you a download link with the subject "Data export from Help Scout."
7. Log in with the **same user account** that initiated the export to download the file.

### What you get

- Conversation-level metadata: subject, status, assignee, created/closed dates, tags, custom fields (Plus+).
- Customer name and email associated with each conversation.
- Resolution times and response metrics.

### What you don't get

- **No conversation thread contents.** No message bodies, no replies, no internal notes.
- **No satisfaction rating comments.** You get the score, not the text.
- **No attachments.**

> [!NOTE]
> **Timezone behavior differs by format.** CSV exports use the timezone configured by your Help Scout administrator (Manage → Company). XLSX exports default to UTC. If you are merging data from multiple exports, normalize timezones first. ([docs.helpscout.com](https://docs.helpscout.com/article/849-export-reporting-data))

This export is useful for audits, board reporting, and high-level analytics. It is useless for migration — a CSV without message bodies is not importable by any target helpdesk. Multiple export requests are queued and processed one at a time, and large exports can time out.

If you are tempted to use these CSVs as your migration dataset, read [Using CSVs for SaaS Data Migrations](https://clonepartner.com/blog/blog/csv-saas-data-migration/) first. It explains exactly where flat files break down for relational support data.

## Method 2: Customer List via Support Request

Help Scout's support team can provide account owners with a CSV file containing **names and email addresses** of all customers in the account. ([docs.helpscout.com](https://docs.helpscout.com/article/466-data-export-options))

**What it includes:** Customer names and email addresses.

**What it does not include:** Conversation history, phone numbers, custom properties, or any relational data. You get no customer IDs, no conversation counts, and no way to link contacts back to their tickets.

This is useful for a quick contact list for a CRM import. It is not migration-grade.

## Method 3: The Help Scout Inbox API (Full Programmatic Export)

This is the only way to get **complete conversation data** out of Help Scout — including message bodies, internal notes, attachments, customer profiles, and custom fields. ([developer.helpscout.com](https://developer.helpscout.com/mailbox-api/))

### Authentication

The Inbox API 2.0 uses **OAuth 2.0**. Two flows are available:

- **Client Credentials flow** — for internal scripts. No user redirect required. You provide your `app_id` and `app_secret` directly.
- **Authorization Code flow** — for apps used by other Help Scout users.

To create credentials:
1. Go to **Your Profile → My Apps** in Help Scout.
2. Click **Create My App** and provide a name and redirect URL.
3. Copy your **App ID** and **App Secret**.

```bash
# Client Credentials flow — get an access token
curl -X POST https://api.helpscout.net/v2/oauth2/token \
  --data 'grant_type=client_credentials' \
  --data 'client_id={app_id}' \
  --data 'client_secret={app_secret}'
```

The response includes an `access_token` and `expires_in` of **172800 seconds (48 hours)**. ([developer.helpscout.com](https://developer.helpscout.com/mailbox-api/overview/authentication))

> [!CAUTION]
> **Access tokens expire after 48 hours.** Any export script that runs longer — or any scheduled sync — must implement automatic token refresh logic. The Client Credentials flow can simply request a new token when the old one expires. The Authorization Code flow also returns a `refresh_token` that you must persist and use to obtain new access tokens. Credentials must be tied to an active, invited user — exports will fail if the linked user has been deactivated.

### Rate Limits by Plan

All API keys on the same account share the same rate budget: ([docs.helpscout.com](https://docs.helpscout.com/article/1140-mailbox-api))

| Plan | Requests per Minute | Write Request Cost | Burst Limit |
|---|---|---|---|
| **Standard** | 200/min | Each write counts as 2 | 12 writes per 5 seconds |
| **Plus** | 400/min | Each write counts as 2 | 12 writes per 5 seconds |
| **Pro** | 800/min | Each write counts as 2 | 12 writes per 5 seconds |

Plus plans can purchase an **API Rate Limit Increase add-on** ($100/month) to bump the limit to 800 calls/min.

When you hit the limit, the API returns **HTTP 429** with a `Retry-After` header telling you how many seconds to wait. Your script **must** respect this header, or you will get stuck in a retry loop.

### Pagination

Most list endpoints return **50 items per page**, but **List Conversations returns only 25 per page**. ([developer.helpscout.com](https://developer.helpscout.com/mailbox-api/overview/pagination/)) If you are pulling years of mailbox history, that difference changes run time significantly.

One easy miss: `/v2/conversations` defaults to `status=active`. If you want full history, you must explicitly request `status=all`. The documented statuses are `active`, `all`, `closed`, `open`, `pending`, and `spam` — there is no `deleted` status filter on the normal conversation export path. ([developer.helpscout.com](https://developer.helpscout.com/mailbox-api/endpoints/conversations/list/))

### What a Full Export Looks Like

Extracting complete conversation data requires multiple API calls per conversation:

1. **List all conversations** — `GET /v2/conversations?status=all&mailbox={id}` — paginated at 25 per page.
2. **For each conversation, list threads** — `GET /v2/conversations/{id}/threads` — this is where the actual message bodies live.
3. **For each thread with attachments, download each attachment** — the API exposes both a binary file endpoint (`/file`) and a base64 data endpoint (`/data`).
4. **List all customers** — `GET /v2/customers` or `GET /v3/customers` — paginated.
5. **List all users** — `GET /v2/users` — paginated.

```python
# Pseudocode: Export all conversations with threads
import requests
import time

BASE_URL = "https://api.helpscout.net/v2"
headers = {"Authorization": f"Bearer {access_token}"}

def fetch_all_pages(url, params=None):
    results = []
    page = 1
    while True:
        resp = requests.get(url, headers=headers, params={**(params or {}), "page": page})
        if resp.status_code == 429:
            time.sleep(int(resp.headers.get("Retry-After", 10)))
            continue
        data = resp.json()
        results.extend(data.get("_embedded", {}).get(list(data["_embedded"].keys())[0], []))
        if page >= data["page"]["totalPages"]:
            break
        page += 1
    return results

# Step 1: Get all conversations
conversations = fetch_all_pages(f"{BASE_URL}/conversations", {"status": "all"})

# Step 2: For each conversation, get threads
for convo in conversations:
    convo["threads_data"] = fetch_all_pages(
        f"{BASE_URL}/conversations/{convo['id']}/threads"
    )
```

> [!WARNING]
> **Do not trust `embed=threads` for chat conversations.** Help Scout explicitly states that embedded threads are truncated for chat conversations. To get full Beacon chat thread bodies, you must call the List Threads endpoint separately for each conversation. This multiplies your API calls significantly. ([developer.helpscout.com](https://developer.helpscout.com/mailbox-api/endpoints/conversations/list/))

### Thread Types Matter

Help Scout distinguishes `customer`, `message`, `note`, `chat`, `phone`, `lineitem`, and forward thread types. If you drop `note` threads, you lose internal context. If you drop `lineitem` threads, you lose operational history like assignment and status changes. Make sure your extraction script captures all thread types, not just customer-facing messages. ([developer.helpscout.com](https://developer.helpscout.com/mailbox-api/endpoints/conversations/threads/list/))

If you need raw email fidelity, Help Scout also exposes a per-thread **RFC 822 original source** endpoint.

### The Math: API Calls for a Real Export

For an account with 10,000 conversations, an average of 6 threads each, and 2,000 attachments:

- **List conversations:** 10,000 / 25 per page = 400 requests
- **List threads:** 10,000 conversations × 1 request each = 10,000 requests (assuming most fit in one page)
- **Download attachments:** 2,000 requests
- **Total:** ~12,400 requests minimum

On the Standard plan (200/min), that is **62 minutes** of continuous, rate-limited API calls — assuming zero retries and no 429 errors. On Plus (400/min), it drops to ~31 minutes. Real-world runs take longer because of burst limits, token refreshes, and occasional timeouts.

### Useful Query Patterns

Help Scout's query language supports targeted and incremental pulls: ([developer.helpscout.com](https://developer.helpscout.com/mailbox-api/endpoints/conversations/list/))

- **Full history from one inbox:** `mailbox=123&status=all`
- **Incremental export:** `modifiedSince=2026-03-01T00:00:00Z`
- **Conversations with attachments:** `query=(attachments:true)`
- **Conversations by customer ID:** `query=(customerIds:100)`
- **Conversations by replying users:** `query=(replyUserIds:(23 OR 45))`

These filters are good enough for staged cutovers — use `modifiedSince` for delta pulls instead of re-pulling the entire account each time.

### Customer Export: v2 vs v3 Trade-Off

Help Scout offers two customer endpoints with a real trade-off: ([developer.helpscout.com](https://developer.helpscout.com/mailbox-api/endpoints/customers/list/))

- **`/v2/customers`** — page-based pagination, supports mailbox filtering. Better when you need to scope exports to specific mailboxes.
- **`/v3/customers`** — cursor-based pagination, no mailbox filtering. Recommended by Help Scout for better performance on large datasets.

If you use Help Scout's newer contact/company model, export both **customer properties** and **organizations**. Help Scout supports up to 50 contact properties and 50 company properties. ([docs.helpscout.com](https://docs.helpscout.com/article/1385-customer-properties))

### Pros
- **Complete data:** The only way to get full thread content, internal notes, and attachments.
- **JSON format:** Structured, machine-readable, preserves relationships.
- **Filterable:** Query by status, mailbox, date range, tags, and custom fields.
- **Incremental exports:** `modifiedSince` support enables delta pulls for staged cutovers.

### Cons
- Requires developer time to build, test, and maintain the script.
- Token expiration and rate limits require careful engineering.
- Attachments are returned as base64 or binary — you must decode, name, and re-associate them with the correct thread.
- No bulk export endpoint — it is always paginated, per-conversation.

> [!WARNING]
> **Do not rely on AI-generated scripts without review.** We frequently see teams use generative AI to write Python extraction scripts that fail to handle the 48-hour token refresh, the 25-item conversation pagination limit, or the `status=active` default — resulting in silent data loss. Read more in [Why DIY AI Scripts Fail](https://clonepartner.com/blog/blog/why-ai-migration-scripts-fail/).

## Method 4: Help Scout Docs API (Knowledge Base Export)

The **Docs API** is a separate API from the Inbox API, with different authentication and rate limits. ([developer.helpscout.com](https://developer.helpscout.com/docs-api/))

- **Base URL:** `https://docsapi.helpscout.net/v1/`
- **Authentication:** API key via HTTP Basic Auth (not OAuth). Find your Docs API key under **Your Profile → Authentication → API Keys**.
- **Rate limit:** Based on the number of Docs sites on your account, measured per 10-minute window.

You can export:
- **Sites** — List all Docs sites.
- **Collections** — List collections within a site.
- **Categories** — Categories within collections.
- **Articles** — Full article content (HTML body, title, slug, status, timestamps). The article list response is abbreviated, so you must fetch full article bodies from the individual article endpoint.

```bash
# List all articles in a collection
curl -u {DOCS_API_KEY}:X \
  https://docsapi.helpscout.net/v1/collections/{collection_id}/articles
```

> [!NOTE]
> **The Docs API admin UI generates requests that count toward your API rate limit.** If someone on your team is editing articles while your export script runs, you may hit 429 errors faster than expected.

**Pros:** Straightforward API key auth — no OAuth dance. Returns full HTML article content.

**Cons:** Separate API from the Inbox API — requires a second script or integration. Images embedded in articles are hosted on Help Scout's CDN; you must download and re-host them for a migration. Still paginated, no bulk export.

## Method 5: ETL / No-Code Connector Tools

Tools like **Coupler.io**, **Skyvia**, **Hevo Data**, and **Fivetran** can connect to the Help Scout API and sync data to Google Sheets, CSV, BigQuery, or a data warehouse. ([skyvia.com](https://skyvia.com/data-integration/help-scout-csv-file-import-and-export))

### What they are good for
- Scheduled reporting syncs — pull conversation metadata into a BI tool on a recurring basis.
- Building dashboards without writing custom scripts.
- Quick one-time pulls of specific data types.

### What they struggle with
- **Token expiration.** The 48-hour token limit means scheduled syncs can break silently if the tool does not handle refresh tokens properly. Coupler's own Help Scout guide calls this out as a drawback for long-running recurring exports. ([blog.coupler.io](https://blog.coupler.io/help-scout-to-google-sheets/))
- **Thread-level data.** Most connectors pull conversation-level metadata, not the individual thread bodies. You may need custom configuration to fetch threads per conversation.
- **Attachment download.** No-code tools typically cannot download binary attachment data and re-associate it with conversations.
- **Migration readiness.** The output is a flat table, not a relational dataset. Thread order, inline images, and parent-child relationships between conversations and threads are lost or flattened.

Use ETL tools when you need scheduled extracts for BI or ad hoc CSVs for stakeholders. Do not use them as your migration plan when you need thread reconstruction, attachment re-linking, or audit-grade QA.

## Method 6: Automated Migration Services

Self-service migration vendors like **Help Desk Migration (Relokia)** offer wizard-based tools that read from Help Scout's API and write directly to a target platform. They handle conversations, customers, attachments, and agents. ([help-desk-migration.com](https://help-desk-migration.com/helpscout/amp/))

### What they handle well
- Conversation metadata, thread content, and attachments.
- Agent and customer mapping to the target platform.
- A demo migration to preview results before committing.

### Where they fall short
- **Custom field mapping** is often limited or requires manual configuration.
- **Inline images** (images embedded in message bodies, not as file attachments) may not transfer correctly. Help Desk Migration's own Help Scout checklist documents cases that need personalized handling. ([help-desk-migration.com](https://help-desk-migration.com/help/helpscout-data-migration-checklist/))
- **Workflows, automations, and saved replies** are not migrated — those must be rebuilt manually.
- **Per-record pricing** makes it hard to predict total cost. For accounts with 50,000+ tickets, costs scale aggressively.

### Engineer-Led Migration (ClonePartner)

ClonePartner manages the entire extraction infrastructure — API orchestration, token management, attachment handling, and field mapping to your target platform. Unlike automated tools that charge per-ticket, we offer predictable pricing and guarantee exact preservation of inline attachments, threaded conversation history, and custom fields.

For destination-specific guidance:
- [Help Scout to Zendesk](https://clonepartner.com/blog/blog/help-scout-to-zendesk-migration-guide/)
- [Help Scout to Freshdesk](https://clonepartner.com/blog/blog/help-scout-to-freshdesk-migration-guide/)
- [Help Scout to Intercom](https://clonepartner.com/blog/blog/help-scout-to-intercom-migration-guide/)

## The Hidden Challenge: Exporting Attachments and Inline Images

Attachments are where Help Scout exports stop being easy. They are stored at the **thread level**, not the conversation level. Each thread can contain multiple attachments, and each must be downloaded individually via the API.

The List Threads endpoint includes an `_embedded.attachments` array with metadata (filename, mimeType, size) and links to the download endpoint. The process looks like this: ([developer.helpscout.com](https://developer.helpscout.com/mailbox-api/endpoints/conversations/threads/list/))

1. Parse the attachment metadata from each thread.
2. Call the download endpoint for each attachment. The API exposes both a **binary file endpoint** (`/file`) and a **base64 data endpoint** (`/data`). Use the binary endpoint when your target system expects a file stream; use the data endpoint when a downstream process expects base64.
3. Decode and store the file with a naming convention that preserves the `conversationId → threadId → attachmentId` relationship.

> [!WARNING]
> **Deprecated attachment link.** As of April 15, 2026, the old attachment `web` link is deprecated, and Help Scout says it will be removed on **May 31, 2026**. Use the documented `download` link or the `/file` endpoint instead. ([developer.helpscout.com](https://developer.helpscout.com/mailbox-api/endpoints/conversations/threads/list/))

**Inline images** are a different problem entirely. Images pasted directly into the message body are typically hosted on Help Scout's CDN and referenced as `<img>` tags in the HTML thread body. They are **not** listed in the attachments array. To preserve them:

1. Parse the HTML body of each thread.
2. Extract all `<img src="..." />` URLs pointing to Help Scout's CDN.
3. Download each image.
4. Re-host the images on your own storage (like AWS S3) and replace the URLs in the HTML body with the new location.

If you skip this step, your migrated conversations will contain broken image links once Help Scout purges or restricts access to the CDN assets.

One more nuance worth noting: **custom fields and contact/company properties are not the same thing** in Help Scout. Custom fields are inbox-specific conversation fields. Contact and company properties are global profile data. Export both if you want to preserve context cleanly. ([docs.helpscout.com](https://docs.helpscout.com/article/593-custom-fields))

## Pre-Export Checklist: Preparing Your Help Scout Data

Before pulling any data, clean up your source. It is cheaper to fix data issues in Help Scout than in your export pipeline.

### 1. Audit and consolidate tags
Remove duplicate or unused tags. Standardize naming conventions (decide between `bug-report`, `Bug Report`, and `bug_report`). Tags export as plain strings — inconsistencies carry over to your target system.

### 2. Verify custom field configurations
Custom fields are only available on **Plus plans and above**. Document every custom field name, type, and dropdown option. The API returns field IDs, not labels, in some contexts — you will need the mapping. Custom fields belong to a single mailbox; if a conversation is moved to another mailbox, the custom-field data does not carry over even if the field name matches. ([developer.helpscout.com](https://developer.helpscout.com/mailbox-api/endpoints/inboxes/list/))

### 3. Map users and agents
List all active and deactivated users. Deactivated users still own historical conversations and must be mapped to the target system — either to their corresponding new profile or to a generic "Legacy Agent" account. See [How to Migrate Users & Organizations Without Breaking History](https://clonepartner.com/blog/blog/migrate-users-organizations/) for the full playbook.

### 4. Identify your mailboxes and scope
If you have multiple mailboxes, decide which ones to export. Each mailbox has its own ID, and conversations are filtered by mailbox in the API.

### 5. Count conversations and estimate API time
Use the Reports dashboard to get a rough count of total conversations. Apply the math from Method 3 to estimate how long the extraction will take on your plan.

### 6. Check for merged conversations
Merged conversations redirect to the surviving conversation via HTTP 301 for 60 days, then return 404. If you have recently merged conversations, export sooner rather than later.

### 7. Capture saved replies
Saved replies are listable via the API (use `includeChatReplies=true` to capture chat-only replies), but importing them into another platform typically requires manual recreation. Extract them before decommissioning your Help Scout account.

### 8. Plan for Docs separately
If your knowledge base matters, set up the Docs API export as a distinct step with its own authentication and rate limit budget.

### 9. Use `status=all` and `modifiedSince` deliberately
Do not rely on the default `active` filter when you want full history. For staged cutovers, use `modifiedSince` for delta pulls instead of re-pulling the entire account.

### 10. Validate counts before calling the export complete
Compare page totals or cursor counts, spot-check attachments and threads, and confirm IDs and mappings before importing anything into the target system.

## What Cannot Be Exported from Help Scout

Some data types are either completely inaccessible or require workarounds:

| Data Type | Exportable? | Notes |
|---|---|---|
| **Workflow conditions & actions** | Partial | API can list workflows and update status, but full condition/action definitions are not documented for backup. Must be rebuilt in the target. |
| **Beacon configuration** | ❌ | No export path. Must be reconfigured in the target. |
| **Deleted conversations** | ❌ | Once hard-deleted, they are gone. May be recoverable if still in a soft-delete state — contact Help Scout support. |
| **Satisfaction rating comments** | API only | Not included in UI CSV/XLSX exports. Available through the Inbox API. |
| **Reporting aggregates (Standard plan)** | ❌ | Standard plan cannot access reporting API endpoints. Requires Plus or Pro. |

## Choosing the Right Export Method

| Scenario | Recommended Method | Why |
|---|---|---|
| Quarterly board report | UI Report Export | Quick, no code, gives you the metrics. |
| Feed data into a BI tool | ETL Tool (Coupler.io, Fivetran) | Scheduled syncs without custom code. |
| Quick customer contact list | Support Request CSV | Fastest path to names + emails. |
| Knowledge base migration only | Docs API | Separate, simpler auth. Returns full HTML. |
| Full platform migration | Inbox API + Docs API (custom script or service) | Only way to get thread content + attachments. |

For a full migration — moving to Zendesk, Freshdesk, Intercom, or another helpdesk — the UI export and ETL tools are not enough. You need the Inbox API for conversations and threads, the Docs API for knowledge base articles, and a plan for attachments and inline images. That is a minimum of two API integrations, token refresh logic, rate-limit handling, and a data pipeline that preserves relational integrity.

## When to Bring in Engineers

If your Help Scout account has more than a few thousand conversations, the API-based export becomes a legitimate engineering project. Token management, rate-limit throttling, attachment download and re-association, inline image handling, and data validation all take time to get right — and they all have failure modes that silently corrupt your data.

If you only need a report export, do it yourself. If you only need a BI feed, use an ETL tool. But if you need a defensible dataset that another system can actually ingest without losing context, get engineers involved.

At ClonePartner, we have run hundreds of helpdesk migrations, including many from Help Scout. We handle the full extraction pipeline — API orchestration, attachment management, thread-level fidelity, and field mapping to your target platform — with [zero downtime guaranteed](https://clonepartner.com/blog/blog/zero-downtime-data-migration/). If you would rather not spend two weeks writing and debugging export scripts, we will have your data out in days.

For a deeper look at how we approach helpdesk moves, see [How We Run Migrations at ClonePartner](https://clonepartner.com/blog/blog/how-we-run-migrations-at-clonepartner/).

> Need your Help Scout data exported cleanly — conversations, threads, attachments, and all? Our engineers handle the full extraction so you don't have to wrestle with token expiration, rate limits, and broken inline images. Book a free 30-minute call.
>
> [Talk to us](https://cal.com/clonepartner/meet?duration=30&utm_source=blog&utm_medium=button&utm_campaign=demo_bookings&utm_content=cta_click&utm_term=demo_button_click)

## Frequently asked questions

### Can you export full conversations from Help Scout as CSV?

The UI report export produces CSV/XLSX files with conversation metadata (subject, status, dates, tags) but explicitly excludes message thread contents, reply bodies, and internal notes. To get full conversation text, you must use the Help Scout Inbox API.

### What are Help Scout's API rate limits?

The Inbox API allows 200 requests/min on Standard, 400/min on Plus, and 800/min on Pro. Write requests count double. All API keys on the same account share the same budget. Plus plans can add a rate limit increase add-on for $100/month. List Conversations pages at 25 records; most other list endpoints page at 50.

### How do I export customers from Help Scout?

There is no self-service customer export in the UI. You can request a CSV of names and emails from Help Scout support (no conversation history included), or use the Inbox API's /v2/customers or /v3/customers endpoints to programmatically extract full customer profiles with phone numbers, properties, and conversation counts.

### Do Help Scout API access tokens expire?

Yes. Help Scout OAuth access tokens expire after 48 hours (172800 seconds). You must implement token refresh logic for any export that takes longer or runs on a schedule. The Client Credentials flow lets you request a new token directly; the Authorization Code flow requires persisting and using a refresh token.

### Can you export Help Scout knowledge base articles?

Yes, using the separate Docs API (not the Inbox API). It uses API key authentication via HTTP Basic Auth, returns article content as HTML, and is rate-limited per 10-minute window based on your number of Docs sites. Images within articles are hosted on Help Scout's CDN and must be downloaded separately.