Product Hunt Launches Scraper – Daily Top, No Login

Get Product Hunt daily leaderboards for any day or period up to 31 days: one row per launch with name, tagline, votes, comments, daily rank, topics, website and public makers. No login and no API token.

Add how many days to read (or a date range), click Start, and you get the launches in a dataset you can export to JSON, CSV or Excel. A typical run (yesterday's top 20) takes about 10–25 seconds and costs about $0.02.

Table of contents

• What it does
• Quick start
• Input
• Output
• Pricing
• Use cases
• Scheduling
• API, export and integrations
• Limits and partial results
• Moving from another Product Hunt scraper
• Troubleshooting
• FAQ
• Support

What it does

The Actor reads the public Product Hunt daily leaderboard for each day you choose. Days follow Product Hunt's calendar (Pacific Time). For each launch it saves one dataset row. With details on (default), it also opens the launch page and adds the product website, makers, hunter, description and pricing type.

You can keep only some launches: by topic (for example artificial-intelligence or Marketing) and by a minimum number of votes.

Data comes from Product Hunt. This Actor is not made by or connected to Product Hunt.

Good fit

• Marketers: a daily list of new tools in your niche, with website and tagline, for newsletters, competitor watch or partner outreach.
• Investors: the top launches of a week or month with votes, comments and rank, to spot early products.
• Sales: new products with their website and public maker names, as a starting list for lead research.

Not a fit / limits

• The Actor does not collect emails, phone numbers or other private contact data. Makers have only public name, username and profile URL.
• No comments text, reviews, followers or maker profiles.
• Up to 31 days in one run. For a longer period, run it again with the next dates.
• Votes and ranks are live numbers. They can change after the launch day. Today's leaderboard (daysBack: 0) is not final.
• Only Product Hunt launches. No other launch sites.

Quick start

1. Open the Input tab.
2. Keep Days back (daysBack) = 1. This reads yesterday's final leaderboard.
3. Keep Max launches (maxItems) = 20, or set your own number.
4. Optional: add Topics (topics), for example artificial-intelligence.
5. Click Start.
6. Open the Output tab and check the Launches view: rank, name, votesCount, website.

Expected result: up to 20 rows from yesterday's featured leaderboard, best rank first.

Input

Input fields

Field	Plain name	Type	Required	Default / prefill	Allowed values	Effect	Example	Recommendation
daysBack	Days back	integer	no	default 1, prefill 1	0–31	Reads N full days before today. 0 = today (live, not final). Ignored when dateFrom or dateTo is set.	7	Best for daily schedules: no dates to edit.
dateFrom	From date	string	no	none	YYYY-MM-DD, not in the future	First day of the period. Alone, it reads one day.	2026-10-01	Use for a fixed past period.
dateTo	To date	string	no	none	YYYY-MM-DD, ≥ dateFrom, period ≤ 31 days	Last day, included. Alone, it reads one day.	2026-10-07
featuredOnly	Featured launches only	boolean	no	true	true / false	true: the featured launches (usually 10–60 a day). false: all launches of the day (often several hundred).	false	Keep true unless you need small launches too.
topics	Topics	array of strings	no	empty	up to 20 names or slugs, any case	Keeps launches with at least one of these topics. Empty = all topics.	["artificial-intelligence", "Marketing"]	Use the slug from the topic page URL.
minVotes	Minimum votes	integer	no	0	≥ 0	Keeps launches with votesCount ≥ this number.	100	Makes runs faster: the Actor stops a day early when votes drop below it. Product Hunt sorts by votes, so a very low value (under 10) can rarely skip a launch.
includeDetails	Include website and makers	boolean	no	true	true / false	true: one extra request per launch for website, makers, hunter, description, pricingType. false: leaderboard fields only.	false	Turn off when you need only votes and ranks: much faster for big runs.
maxItems	Max launches	integer	no	default 100, prefill 20	1–10,000	Stops after this many rows in total. Newest day first.	500	Each row is one paid result.

Empty input ({}) reads yesterday with up to 100 rows. Wrong dates (future date, dateFrom after dateTo, more than 31 days) stop the run at once with INVALID_INPUT, before any source request and without row charges.

Interacting settings

When	Input A	Input B	Effective behavior	Example	Cost/result effect
Both set	daysBack	dateFrom / dateTo	Dates win, daysBack is ignored	daysBack: 7, dateFrom: 2026-10-01 → only 2026-10-01	none
Filter + cap	topics / minVotes	maxItems	Filters first, then the cap counts saved rows	7 days, topics: ["saas"], maxItems: 50	You pay only for saved rows
All launches + filter with few matches	featuredOnly: false	topics	The Actor still reads every leaderboard page of every day	31 days, a rare topic	Many pages: see Pricing
Vote filter	minVotes	any	The leaderboard is sorted by votes, so the Actor stops a day after a full page below minVotes	minVotes: 50	Fewer pages, lower cost
No details	includeDetails: false	topics	Topic filter uses the topics shown on the leaderboard (up to 3 per launch)		Faster; website, makers are null

Input examples

Yesterday's top 20 (prefill)

{
  "daysBack": 1,
  "maxItems": 20
}

Expected behavior: reads one leaderboard page and opens each launch page. Expected output: up to 20 launches with rank, name, votesCount, website, makers. Cost note: about $0.02.

AI launches with 100+ votes in one week

{
  "dateFrom": "2026-10-05",
  "dateTo": "2026-10-11",
  "topics": ["artificial-intelligence"],
  "minVotes": 100,
  "maxItems": 200
}

Expected behavior: reads 7 days, keeps AI launches with at least 100 votes. Expected output: launches with topics containing "Artificial Intelligence" and votesCount ≥ 100.

Output

Each dataset row is one launch from one leaderboard day. Rows go from the newest day to the oldest, and by rank inside a day.

Output example

{
  "name": "ngram 2.0",
  "tagline": "Create explainers, screencasts, and AI avatar videos",
  "url": "https://www.producthunt.com/products/ngram/launches/ngram-2-0-2",
  "productUrl": "https://www.producthunt.com/products/ngram",
  "website": "https://www.ngram.com",
  "votesCount": 269,
  "launchDayVotesCount": 248,
  "votesHidden": false,
  "commentsCount": 42,
  "rank": 2,
  "weeklyRank": 3,
  "monthlyRank": 35,
  "topics": ["Education", "Artificial Intelligence", "Video"],
  "topicSlugs": ["education", "artificial-intelligence", "video"],
  "makers": [{ "name": "…", "username": "…", "profileUrl": "https://www.producthunt.com/@…" }],
  "hunter": { "name": "…", "username": "…", "profileUrl": "https://www.producthunt.com/@…" },
  "description": "Turn ideas, documents, and recordings into narrated videos…",
  "pricingType": "free_options",
  "featured": true,
  "launchDate": "2026-10-12",
  "launchedAt": "2026-10-12T07:01:00.000Z",
  "thumbnail": "https://ph-files.imgix.net/ff626e36-21c0-49e6-9937-f794c9825256.png",
  "postId": "1266889",
  "slug": "ngram-2-0-2",
  "productSlug": "ngram",
  "detailsStatus": "ok",
  "source": "Product Hunt (https://www.producthunt.com/)",
  "fetchedAt": "2026-10-14T03:40:18.334Z"
}

This is one row from a run for 2026-10-12. The description is shortened and maker names are hidden here. null means Product Hunt did not give this value or details were off. It is never used instead of 0.

Output fields

Field	Meaning	Type	Can be empty?	Example / enum	Notes
name	Launch name	string	no	ngram 2.0
tagline	One-line tagline	string	yes	Create explainers, screencasts, and AI avatar videos
url	Product Hunt launch page	string	no	https://www.producthunt.com/products/ngram/launches/ngram-2-0-2
productUrl	Product Hunt product page	string	yes	https://www.producthunt.com/products/ngram
website	Product website	string	yes	https://www.ngram.com	null when details are off or there is no link
votesCount	Upvotes now	integer	yes	269	null when Product Hunt does not return it; never 0 for missing
launchDayVotesCount	Upvotes at the end of the launch day	integer	yes	248
votesHidden	The maker hides the vote count	boolean	no	false
commentsCount	Number of comments	integer	yes	42
rank	Daily leaderboard rank	integer	yes	2	null for launches without a rank
weeklyRank, monthlyRank	Weekly and monthly rank	integer	yes	3, 35
topics	Topic names	array	no (can be [])	["Education", "Artificial Intelligence", "Video"]	All topics with details, up to 3 without
topicSlugs	Topic slugs	array	no	["education", "artificial-intelligence", "video"]
makers	Makers: public name, username, profile URL	array	yes	[{"name": "…", "username": "…", "profileUrl": "…"}]	[] = none listed; null = details not read
hunter	Who posted the launch	object	yes	{"name": "…", "username": "…", "profileUrl": "…"}
description	Launch description	string	yes	Turn ideas, documents, and recordings into narrated videos…	HTML entities decoded; email addresses replaced with [email removed]
pricingType	Pricing set by the maker	string	yes	free, free_options, payment_required
featured	Featured by Product Hunt	boolean	no	true
launchDate	Leaderboard day (Pacific Time)	string	no	2026-10-12
launchedAt	Launch time, UTC	string	yes	2026-10-12T07:01:00.000Z
thumbnail	Thumbnail image URL	string	yes	https://ph-files.imgix.net/….png
postId, slug, productSlug	Product Hunt IDs	string	productSlug yes	1266889, ngram-2-0-2, ngram	Use postId to remove duplicates between runs
detailsStatus	Did the details load?	string	no	ok, not_found, skipped	not_found = the launch page no longer exists; the row keeps leaderboard fields
source	Data source	string	no	Product Hunt (https://www.producthunt.com/)
fetchedAt	When the row was read, UTC	string	no	2026-10-14T03:40:18.334Z

Output entities and surfaces

Situation	Where	Result shape	User action
Normal run	Dataset, view Launches; RUN_SUMMARY status COMPLETE or CAPPED	Launch rows	Export or use the fields
No launches on these days	RUN_SUMMARY status NOT_FOUND, run SUCCEEDED	Empty dataset	Check the dates (Product Hunt started in 2013)
Filters removed everything	RUN_SUMMARY status NO_RESULTS, run SUCCEEDED	Empty dataset	Use a wider topic or a lower minVotes
Some data could not be read	RUN_SUMMARY status PARTIAL with reason	Saved rows; some rows may have website/makers = null	Read message and warnings; run again for missing days
Source error, nothing saved	RUN_SUMMARY status ERROR with errorCode, run FAILED	Empty dataset	Run again later; if SOURCE_QUERY_CHANGED, open an issue

RUN_SUMMARY (in the run's key-value store) has status, reason, errorCode, message, the normalized input, counts (days, pages, rows saved, filtered out, cut by maxItems), one line per day and request statistics.

Pricing

Pricing unit

You pay per event:

Event	When	Price
Actor start	Once per run (per GB of memory; default 256 MB = 1 event)	$0.001
Leaderboard page	Each leaderboard page read (up to 20 launches)	$0.0005
Launch	Each row saved to the dataset	$0.001 (FREE plan) → $0.00085 (GOLD and higher)

Launch price by Apify plan: FREE $0.001, BRONZE $0.00095, SILVER $0.0009, GOLD and higher $0.00085. Website and makers (includeDetails) do not cost extra.

A run with no rows still pays for the start and the pages it read. An empty day or a filter with no matches on one day costs about $0.0015. Invalid input costs only the $0.001 start event: no pages or rows are charged. If you set a maximum charge for the run, the Actor stops at that limit and keeps the saved rows (PARTIAL, reason CHARGE_LIMIT).

Cost examples (FREE plan prices)

Scenario	Input scope	Rows	Pages	Approx. charge	Stop/cap behavior
10 launches	daysBack: 1, maxItems: 10	10	1	$0.012	Stops at 10 rows
Yesterday's top 20 (prefill)	daysBack: 1, maxItems: 20	20	1	$0.022	Stops at 20 rows
One week, featured	daysBack: 7, maxItems: 100	100	up to 14	$0.11	Stops at 100 rows
One busy day, all launches	featuredOnly: false, maxItems: 1000	1,000	~50	$1.03	Stops at 1,000 rows
Empty day or no matches	1 day	0	1	$0.0015	SUCCEEDED, NOT_FOUND / NO_RESULTS
31 days, all launches, rare topic	featuredOnly: false, 31 days, topics	0–few	~1,400–1,600	$0.70–0.80 + rows	Reads every page; set a charge limit

Use cases

Buyer job	Input path	Output entity	Key fields	Next integration
Daily digest of new tools	daysBack: 1, topics	Launch	name, tagline, website, topics	Schedule + export to Sheets or Slack
Top launches of a week for deal flow	dateFrom, dateTo, minVotes	Launch	rank, votesCount, commentsCount, weeklyRank	CSV / Excel
Lead list of new products	daysBack: 7, topics	Launch	website, makers, productUrl	Your CRM via API
Full day, also non-featured	featuredOnly: false	Launch	featured, rank, votesCount	Dataset API

Scheduling

To get new launches every day, create an Apify schedule with this input and run it once a day, after midnight Pacific Time:

{
  "daysBack": 1,
  "maxItems": 100
}

Each run reads the last full day, so you do not edit dates. The Actor does not compare runs: each run is a full snapshot of that day. Use postId to remove duplicates if your periods overlap.

API, export and integrations

• Export: from the Output tab as JSON, CSV, Excel, XML or HTML.
• API: start the Actor with the Apify API, then read the dataset items. postId is a stable key.
• No-code: connect the dataset to Google Sheets, Slack, Make, Zapier or n8n with Apify integrations.

Use with AI agents (MCP)

• MCP server with only this Actor: https://mcp.apify.com?tools=cliqtomedia/product-hunt-launches (see Apify MCP server).
• Smallest useful input: {"daysBack": 1, "maxItems": 20}.
• Key output fields: name (launch), rank (daily place), votesCount (current votes), website (product site), topics (topic names).
• A typical call takes 10–25 seconds and costs about $0.02. Limits: 31 days and 10,000 rows per run; INVALID_INPUT explains a wrong date in RUN_SUMMARY.message.

Limits and partial results

• Up to 31 days and 10,000 rows per run; up to 150 leaderboard pages per day.
• Speed: about 1 leaderboard page per second; details add about 0.5 seconds per row.
• Default timeout is 30 minutes. 31 days with all launches and details may need more: set timeout to 3600 seconds. When time runs out, the Actor saves what it has and ends with PARTIAL, reason DEADLINE.
• PARTIAL means some data was not read, but the saved rows are correct. RUN_SUMMARY.reason says why: DEADLINE (time limit), CHARGE_LIMIT (your maximum charge), REQUEST_CAP (internal request limit) or SOURCE_ERROR (Product Hunt failed after retries).
• Rows from a PARTIAL run can have website and makers = null when details were not read. detailsStatus shows this for each row.

Moving from another Product Hunt scraper

If you already use a dataset from another Product Hunt scraper, these renames are often enough:

Common field elsewhere	Field here	Note
upvotesCount	votesCount	Same number (live votes)
dayRank	rank
weekRank, monthRank	weeklyRank, monthlyRank
thumbnailUrl	thumbnail
tags	topics	Here: all topics with details, not only one
website.url	website	Here: a plain string
team	makers + hunter	Here: only the launch makers and the hunter, public fields only

name, tagline, url, commentsCount and launchDate have the same names.

Troubleshooting

The dataset is empty

Symptom: the run SUCCEEDED but there are no rows. Likely reason: RUN_SUMMARY.status is NO_RESULTS (filters removed all launches) or NOT_FOUND (no launches on these days). Check: RUN_SUMMARY.counts.launchesListed and filteredOut. Action: use a topic slug from the Product Hunt topic page, lower minVotes, or choose other dates.

Ranks have gaps (for example 1, 2, 4)

Symptom: a rank number is missing in a day. Likely reason: Product Hunt removed or hid that launch, or the place is an ad. The Actor skips ads and counts them in RUN_SUMMARY.skippedFeedTypes. Action: nothing to fix; the other rows are complete.

Some rows have website and makers = null

Symptom: leaderboard fields are filled, details are empty. Likely reason: includeDetails is off (detailsStatus: skipped), the launch page no longer exists (not_found), or the run was PARTIAL. Action: check detailsStatus and RUN_SUMMARY.warnings; run that day again if needed.

The run ended with PARTIAL and reason DEADLINE

Symptom: not all days were read. Likely reason: a long period with all launches and details needs more than the run timeout. Action: set timeout to 3600 seconds, or split the period into smaller runs.

FAQ

Do I need a Product Hunt account or API token?

No. The Actor reads public leaderboard pages. You do not add a login, cookie or token.

Why are votes different from the Product Hunt page?

Votes are live. They change during the day and a little after it. launchDayVotesCount keeps the number from the end of the launch day. fetchedAt shows when the row was read.

Does it give maker emails?

No. Only public name, username and profile URL. Emails in descriptions are replaced with [email removed].

Are all launches of a day included?

With featuredOnly: true, you get the featured launches. With featuredOnly: false, you get all launches on the daily leaderboard, except ads.

Support

For bugs, questions or feature requests, open an issue in the Issues tab. Include the run ID, your input (without private data) and the RUN_SUMMARY message. Do not send tokens or private data.