HTML Tables to Excel Exporter

Convert HTML tables to Excel from anonymous public webpages. Supply one or more URLs and receive a formatted XLSX workbook with one worksheet per page, plus structured dataset records for automation.

The Actor preserves table order, expands rowspan and colspan cells into a rectangular grid, detects headers, converts plain numeric cells to Excel numbers, and can add sum or average formulas. Multiple tables from one page are stacked in their original order on that page's worksheet.

What does HTML Tables to Excel Exporter do?

The Actor downloads server-rendered HTML and finds native <table> elements using a configurable CSS selector. For each page it:

1. validates that the URL resolves to a public HTTP(S) address;
2. follows a bounded number of safe redirects;
3. extracts matching captions, headers, rows, and cells;
4. creates one uniquely named worksheet for that page;
5. writes every matching table in document order;
6. saves the complete workbook as OUTPUT.xlsx;
7. returns source metadata and extracted rows in the default dataset.

It processes URLs independently. A failed URL or a page without matching tables produces a bounded status record while other URLs continue.

Who is it for?

• Analysts turning public statistical tables into reporting workbooks
• Researchers combining tables from several reference pages
• Operations teams replacing repeated copy-and-paste work
• Developers feeding ordered web-table rows into data pipelines
• Finance and reporting teams that need optional sum or average formulas
• Apify users scheduling snapshots of server-rendered public tables

What data is exported?

Field	Meaning
sourceUrl	URL supplied in the input
finalUrl	URL after safe redirects
pageTitle	Downloaded page's HTML title
status	exported, no_tables, or error
sheetName	Unique worksheet name for that page
tableCount	Number of matching tables exported
rowCount	Total extracted data rows on the page
columnCount	Largest table width on the page
tables	Ordered captions, headers, and cell-row arrays
workbookKey	Key-value store record name, OUTPUT.xlsx
error	Page-level error or no-table explanation
processedAt	ISO timestamp for the page result

The Excel workbook is the primary file deliverable. The dataset provides a JSON-friendly representation of the same extracted tables and a result for every attempted URL.

How are worksheets organized?

Each successfully exported webpage gets one worksheet. The page title becomes the worksheet name, shortened to Excel's 31-character limit and made unique when titles repeat.

The first rows contain the page title and clickable source URL. Each table then contains:

• its caption, or a deterministic Table N label;
• a styled header row;
• ordered data rows;
• an optional formula row;
• a blank separator before the next table.

Column widths are adjusted for readability. Large cell text is capped visually by a practical maximum width but remains in the workbook.

How much does it cost to export HTML tables to Excel?

The Actor uses pay-per-event pricing:

• $0.005 start fee once per run
• $0.0010688 per successfully exported page at the Bronze tier
• lower per-page rates at higher platform tiers
• no per-page fee when a webpage is not successfully exported

Example Bronze-tier costs with the current pricing are approximately:

Successful pages	Estimated Actor charge
1	$0.0061
10	$0.0157
25	$0.0317
100	$0.1169

Actual platform billing can also include Apify compute and storage according to your plan. The input is capped at 50 pages per run; the 100-page example represents two full runs and includes two start events.

Getting started

1. Open the Actor in Apify Console.
2. Add one or more public webpage URLs under Webpage URLs.
3. Keep table as the selector to export all native tables, or enter a narrower selector such as table.wikitable.
4. Choose whether numeric columns need sum or average formulas.
5. Adjust page, table, row, or timeout limits if necessary.
6. Click Start.
7. Download Excel workbook from the run's Output tab.
8. Inspect Page results for source mapping, ordered JSON rows, or bounded errors.

The prefilled W3Schools URL returns a real customer table and is suitable for a first run.

Input parameters

startUrls

Required array of public HTTP(S) webpage URLs. The Actor supports anonymous pages only. URLs containing credentials and addresses resolving to local or private networks are rejected.

tableSelector

CSS selector applied to each page. The default table matches every native HTML table. Useful examples include:

• #customers
• table.wikitable
• main table.data

The selector must resolve to <table> elements. A valid selector with no matching native tables returns no_tables.

summaryMode

• none: no summary formula
• sum: SUM formula below numeric columns
• average: AVERAGE formula below numeric columns

Formulas are added only to columns containing at least one parsed numeric cell.

Processing limits

• maxPages: 1–50, default 10
• maxTablesPerPage: 1–50, default 20
• maxRowsPerTable: 1–5,000, default 2,000
• requestTimeoutSecs: 5–120, default 30

These explicit limits keep workbook size, runtime, and failure behavior predictable.

Input example

{
  "startUrls": [
    { "url": "https://www.w3schools.com/html/html_tables.asp" },
    { "url": "https://en.wikipedia.org/wiki/List_of_chemical_elements" }
  ],
  "tableSelector": "table",
  "summaryMode": "average",
  "maxPages": 2,
  "maxTablesPerPage": 3,
  "maxRowsPerTable": 2000,
  "requestTimeoutSecs": 30
}

Output example

A successful page produces a record like this:

{
  "sourceUrl": "https://www.w3schools.com/html/html_tables.asp",
  "finalUrl": "https://www.w3schools.com/html/html_tables.asp",
  "pageTitle": "HTML Tables",
  "status": "exported",
  "sheetName": "HTML Tables",
  "tableCount": 1,
  "rowCount": 6,
  "columnCount": 3,
  "tables": [
    {
      "tableIndex": 1,
      "caption": null,
      "headers": ["Company", "Contact", "Country"],
      "rows": [["Alfreds Futterkiste", "Maria Anders", "Germany"]],
      "rowCount": 6,
      "columnCount": 3
    }
  ],
  "workbookKey": "OUTPUT.xlsx",
  "error": null,
  "processedAt": "2026-01-15T12:00:00.000Z"
}

The actual rows array contains every accepted row up to maxRowsPerTable.

Download the XLSX workbook

The run stores the workbook in its default key-value store under:

OUTPUT.xlsx

Use the Output tab's Excel workbook link or request the record directly:

https://api.apify.com/v2/key-value-stores/<STORE_ID>/records/OUTPUT.xlsx

The MIME type is the standard Office Open XML workbook type, so the file opens in Microsoft Excel, LibreOffice Calc, Google Sheets, and compatible tools.

Schedule recurring table exports

Apify schedules can run the same input hourly, daily, or weekly. A useful recurring workflow is:

1. schedule the Actor with stable source URLs;
2. download each run's OUTPUT.xlsx through an integration;
3. store the file with the run timestamp;
4. compare dataset rows or workbooks downstream.

The Actor creates snapshots. It does not itself diff old runs, send change alerts, or merge historical workbooks.

Use the Actor through the Apify API

cURL

curl -X POST \
  "https://api.apify.com/v2/acts/automation-lab~html-table-to-excel-exporter/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "startUrls": [{"url":"https://www.w3schools.com/html/html_tables.asp"}],
    "tableSelector": "#customers",
    "summaryMode": "none",
    "maxPages": 1
  }'

JavaScript

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('automation-lab/html-table-to-excel-exporter').call({
  startUrls: [{ url: 'https://www.w3schools.com/html/html_tables.asp' }],
  tableSelector: '#customers',
  summaryMode: 'none',
  maxPages: 1,
});
console.log(run.defaultKeyValueStoreId, run.defaultDatasetId);

Python

import os
from apify_client import ApifyClient

client = ApifyClient(os.environ['APIFY_TOKEN'])
run = client.actor('automation-lab/html-table-to-excel-exporter').call(run_input={
    'startUrls': [{'url': 'https://www.w3schools.com/html/html_tables.asp'}],
    'tableSelector': '#customers',
    'summaryMode': 'none',
    'maxPages': 1,
})
print(run['defaultKeyValueStoreId'], run['defaultDatasetId'])

Connect with webhooks and integrations

After a successful run, connect the dataset or workbook to:

• Google Drive or Dropbox for workbook archiving
• Slack or email for completion notifications
• Make, Zapier, or n8n for downstream workflows
• Python, pandas, or BI tools through the dataset API
• Apify webhooks for event-driven processing

Use the dataset status field to distinguish exported pages from bounded errors before downstream ingestion.

Use with Apify MCP

Add the Actor to Claude Code:

claude mcp add --transport http apify \
  "https://mcp.apify.com?tools=automation-lab/html-table-to-excel-exporter"

Claude Desktop, Cursor, and VS Code setup

Use this equivalent MCP JSON configuration in Claude Desktop, Cursor, or VS Code:

{
  "mcpServers": {
    "apify": {
      "url": "https://mcp.apify.com?tools=automation-lab/html-table-to-excel-exporter"
    }
  }
}

Example prompts:

• “Export the native tables on these three public URLs into one Excel workbook.”
• “Use table.wikitable and create average formulas for numeric columns.”
• “Run the HTML table exporter and summarize any URLs that returned no tables.”

Limits and supported pages

The Actor intentionally supports native, server-rendered HTML tables. It does not render JavaScript-only grids, sign in, solve CAPTCHAs, click pagination controls, or reconstruct visual layouts made from <div> elements.

Other limits:

• 5 MB maximum downloaded HTML response per page
• 5 safe redirects per URL
• 50 pages, 50 tables per page, and 5,000 data rows per table
• 100 expanded columns per source cell's colspan
• one shared CSS selector for all supplied pages
• one worksheet per successful page, not one worksheet per table

If a site blocks anonymous HTTP access, the URL returns an error record and does not stop other pages.

Tips for reliable exports

• Inspect the page source to confirm the data uses actual <table> elements.
• Start with a narrow selector when a page contains navigation or layout tables.
• Use maxTablesPerPage: 1 when only the primary table matters.
• Keep source pages grouped by a selector they share.
• Use none when text that resembles numbers must remain exactly textual.
• Review formulas before using a generated workbook for financial decisions.
• Schedule at a respectful frequency and follow the source website's terms.

HTML table to CSV, JSON, and Markdown

The default dataset can be downloaded from Apify as JSON, CSV, Excel, XML, or other platform-supported formats. The Actor's custom file deliverable is XLSX.

It does not generate a bespoke Markdown file or preserve visual CSS styling from the source page. If your job is specifically HTML table to CSV, the dataset's CSV export may be useful, but nested multi-table records are best consumed as JSON or through OUTPUT.xlsx.

Legality and responsible use

Only process public pages you are authorized to access. Respect website terms, robots policies where applicable, intellectual-property rights, database rights, privacy law, and rate limits. Avoid collecting personal or sensitive information without a lawful purpose.

The Actor blocks local and private network destinations to reduce server-side request forgery risk. It does not bypass authentication or access controls.

FAQ and troubleshooting

Why did I receive no_tables?

The page downloaded successfully, but no native <table> matched tableSelector. Confirm the selector in browser developer tools and check whether the visible grid is rendered later by JavaScript.

Why did one URL fail while the run succeeded?

Errors are bounded per URL by design. Check that record's error field for HTTP status, timeout, content type, private-address rejection, response-size limit, or redirect details.

Why are some numbers still text?

Values containing units, footnotes, dates, ranges, or nonstandard separators remain text to avoid destructive guessing. Plain currencies, percentages, and conventional numbers are converted conservatively.

Why does a merged source cell repeat?

rowspan and colspan are expanded into rectangular workbook cells so row and column positions remain machine-readable. The repeated value represents the source cell's covered grid area.

Can it export pages requiring login?

No. The supported scope is anonymous public webpages with server-rendered native HTML tables.

Related automation-lab Actors

• CSV & Excel Data Quality Cleaner — normalize, validate, and deduplicate tabular files after export.
• Dataset Dedup — remove duplicate records from downstream datasets.

Use this Actor for webpage-table extraction, then apply a related utility only when the downstream workflow requires additional cleanup.

Changelog

See the Actor's Changelog tab for customer-facing release notes.