Logo & Favicon Extractor: Company Logos from a List or CSV

Give it a list of websites. Get back each company's logo, its best square icon and its favicon, as working image links with their real size and format.

Building a CRM, a directory, a comparison page or a lead list and need a logo next to every company? Since Clearbit's free logo service was switched off, that usually means hunting through each site by hand. This does it for the whole list in one run: paste websites, or point it at an Apify dataset, a CSV, an Excel file or a Google Sheet, and every row comes back with its original columns plus the logo columns.

$0.002 per website where something is found ($2 per 1,000). Sites that can't be read or have nothing usable are free.

What you get for each website

Column	What it is
logoUrl	The company's own logo file: from its structured data (schema.org), or the image in its header's home link
logoFormat, logoWidth, logoHeight	Measured from the downloaded file (svg, png, webp, jpeg...), not read from the HTML
iconUrl	The best square icon, largest first: apple-touch-icon, web app manifest icon or favicon
iconWidth, iconHeight, iconFormat	Measured, so you can drop anything too small for your layout
faviconUrl	The classic small browser icon
siteName	The brand name the site declares (or its page title)
themeColor	The brand colour the site declares for browsers, when it has one
ogImageUrl	The social share image, kept separate because it is a banner, not a logo
storedLogoUrl, storedIconUrl	Your own copy of each image, if you switch on "Save a copy" (no extra charge)
status, statusDetail	ok, or a plain reason why not (blocked, unreachable, robots.txt, nothing found...)

Every image link is checked. Each logo and icon is downloaded and confirmed to be a real image before it is returned, so you never get a link that turns out to be a "not found" page. Broken candidates are skipped and the next one is tried.

Example

Input: websites: ["stripe.com", "https://www.ikea.com", "notion.so", "github.com"]

website	siteName	logoUrl	logo size	iconUrl	icon size
stripe.com	Stripe	...stripeassets.com/.../favicon.svg	svg 512x512	.../apple-touch-icon.png	180x180
ikea.com	IKEA	https://www.ikea.com/global/assets/logos/brand/ikea.svg	svg 98x40	.../favicon.ico	48x48
notion.so	Notion	https://www.notion.com/front-static/logo-ios.png	png 512x512	.../apple-touch-icon.png	512x512
github.com	GitHub	(none: drawn as inline SVG)		.../manifest icon	512x512

When a site draws its logo as inline code (GitHub, Airbnb, The Guardian) there is no logo file to link to; logoIsInlineSvg says so and the best icon is still returned.

Input

• Websites: type or paste websites or bare domains, one per line.
• Dataset: the output of any Google Maps, directory or lead scraper. The website column is found automatically (website, domain, url, homepage...).
• File or Google Sheet URL: a public CSV, Excel, JSON file, or a Google Sheet shared as "Anyone with the link can view".

Options: keep only found rows or only problems, save a copy of every image into the run's storage, list every icon the site declares, export CSV or Excel, append to a named dataset.

What it costs

Event	Price
Website where a logo or icon was found	$0.002
CSV or Excel export file	$0.01

1,000 websites cost about $2. Unreachable, blocked, robots.txt-refused, duplicate and invalid rows, and sites with nothing usable, are not charged. If the home page is blocked but the site's icon is at its standard address, you get the icon and that row is charged once.

Good to know

• Robots.txt is respected by default. A site that asks tools not to read its home page is skipped (not charged).
• Logos are trademarks. This Actor tells you where each company publishes its own logo. Whether you may display it is between you and the owner.
• No browser, no proxy needed for most sites, which is why it is fast and cheap. An optional proxy is there for sites that block by country.

For AI agents and automations

Pay per event, no subscription, callable through the Apify API, MCP (nerolabs/logo-favicon-finder) and x402. Send {"websites": ["stripe.com"]} and read logoUrl and iconUrl from the dataset.

The rest of the toolkit

Pair it with Website Contact Finder and Domain Availability Checker & WHOIS on the same list, then shape the result with Dataset Filter & Transform, Merge Datasets & Remove Duplicates or Dataset Cleaner & Exporter.

---

If this saved you hunting for logos one site at a time, a review on the Store page helps a lot. If something looks wrong, open an issue on the Issues tab and I will answer personally.