kickstarter-scraper

Scrape Kickstarter as clean, structured JSON or CSV. You get project search with category, state, location, goal and funding filters, full project details with every reward tier and add-on, live funding counters, and a funding tracker that reports what changed since the last run. It ships as a CLI and a Node library.

Powered by ScrapeUnblocker. Every request goes through the ScrapeUnblocker getPageSource API with the official Node.js SDK, so you don't run browsers, manage proxies or get rate-limited from your own IP. The scraper reads the JSON Kickstarter itself uses (its discover search feed, the project object embedded in each project page, and small stats and location endpoints) instead of the visual markup, so there are no CSS selectors to maintain.

Features

• 🔎 Project search: Kickstarter's discover index with free-text terms, any of its 15 categories and 154 subcategories, project state (live, successful, failed, upcoming), location (a city or a country, by name or ID), "Projects We Love" only, goal, pledged and percent-raised ranges, and all six sort orders (magic, popularity, newest, ending soon, most funded, most backed). You can fetch up to 200 pages of 12 projects per query. Rows are deduped, and you can pass several search terms at once.
• 📊 Clean project records: goal, pledged, percent funded, backers, average pledge, the USD conversion Kickstarter fixed for the campaign, launch date, deadline, duration, days left, category with its parent, creator, location with its ID, staff pick and late-pledge flags, and the image and video URLs.
• 🎁 Project details and rewards: comments and updates counts, tags, and every reward tier and add-on with its price, backers, quantity limit and remaining stock, estimated delivery month, shipping summary, availability window and included items.
• 📈 Funding tracker: stats reads a tiny JSON endpoint with live backers, pledged amount and comments. With --history it appends snapshots to a JSON Lines file and reports the change since the previous run, so a cron job can build funding curves for a watchlist.
• 🗺️ Location lookup: turns "Berlin" into the location ID the search filter takes.
• 🗂️ Category list: ships offline, or you can refresh it live with categories --live.
• 📄 JSON, JSON Lines or CSV output. Nested values are flattened for CSV.
• 🔁 Retries with backoff on blocks, timeouts and incomplete responses. A real 404 is never retried.

Install

Requires Node.js 18 or newer.

git clone https://github.com/ScrapeUnblocker/kickstarter-scraper.git
cd kickstarter-scraper
npm install
npm link   # optional: puts `kickstarter-scraper` on your PATH

Get an API key at scrapeunblocker.com and export it (or copy .env.example to .env and load it your usual way):

export SCRAPEUNBLOCKER_KEY=your_key_here

CLI usage

kickstarter-scraper <command> [arguments] [options]

  search [term...]       Projects from Kickstarter's discover search (12 per page)
  project <url...>       Full project details: funding, creator, rewards, add-ons
  rewards <url>          One row per reward tier and add-on of a project
  stats <url...>         Live backers / pledged / comments (cheap, for tracking)
  locations <place>      Find the location ID for a city or country
  categories             List categories (no API call; --live to refresh)

<url> is a project URL (any tab, with or without ?ref= parameters) or a creator/project-slug path. Run kickstarter-scraper --help to see every option.

# The most-funded live board games, 3 pages, as CSV
kickstarter-scraper search -c "Tabletop Games" -s live --sort most_funded -p 3 -o games.csv

# Successful campaigns for two keywords with a $10k-$100k goal, most backed first
kickstarter-scraper search solar "e-bike" -s successful --goal 10k-100k --sort most_backed

# Staff-picked technology projects from Berlin
kickstarter-scraper search -c technology --location Berlin --staff-picks

# Campaigns ending soon that are already over 100% funded, with 500+ backers
kickstarter-scraper search -s live --sort end_date --raised over-100 --min-backers 500 -p 5

# Prelaunch pages in Video Games, newest first
kickstarter-scraper search -c "Video Games" -s upcoming --sort newest

# Full project details, and its reward tiers as CSV
kickstarter-scraper project https://www.kickstarter.com/projects/brotherwise/war-for-roshar
kickstarter-scraper rewards brotherwise/war-for-roshar -o rewards.csv

# Track a watchlist (one URL per line); each run reports the change since the last
kickstarter-scraper stats --from watchlist.txt --history funding.jsonl -f csv

# Location IDs and categories
kickstarter-scraper locations "New York"
kickstarter-scraper categories -o categories.csv

Option	Values
-c, --category	A name (Gadgets), a slug (technology/gadgets, games/tabletop-games) or an ID (337). Names that exist in several categories, such as "Comedy", ask you to pick a slug.
-s, --state	live, successful, failed, upcoming or all (default). Upcoming projects are prelaunch pages, with the state submitted or started.
--sort	magic (default), popularity, newest, end_date (alias ending-soon), most_funded, most_backed
--goal, --pledged	under-1k, 1k-10k, 10k-100k, 100k-1m, over-1m (USD)
--raised	under-75, 75-100, over-100 (percent of goal)
--location	A city or country name (the best match is used) or a location ID from locations
-p, --pages / -l, --limit	Pages per term (1-200) and maximum projects per term
--min-backers, --min-percent	Local filters applied after fetching
-o, --out / -f, --format	Output file and json, jsonl or csv (taken from the file extension by default)

Library usage

import { KickstarterScraper, toCsv, trackStats } from "kickstarter-scraper"; // or "./src/index.js" in a clone

const ks = new KickstarterScraper(); // reads SCRAPEUNBLOCKER_KEY

// Search: friendly filters, paging, dedupe
const games = await ks.search({
  category: "Tabletop Games",
  state: "live",
  sort: "most_funded",
  pages: 2,
});
console.log(games[0].name, games[0].percent_funded, games[0].days_left);

// Several terms with the same filters; each row gets the `term` that found it
const rows = await ks.searchTerms(["solar", "e-bike"], { state: "successful", goal: "10k-100k" });

// Full details with reward tiers and add-ons (null if the project does not exist)
const project = await ks.project("brotherwise/war-for-roshar");
for (const r of project.rewards) console.log(r.title, r.minimum, r.backers_count, r.remaining);

// Live counters for a watchlist, and the change since the previous snapshot
const changes = await trackStats("funding.jsonl", await ks.stats([project.url]));

// How many projects match a query, from one request
const failed = await ks.count({ category: "Gadgets", state: "failed" });

// Location IDs
const [berlin] = await ks.locations("Berlin"); // { id: 638242, name: "Berlin, Germany", ... }

console.log(toCsv(rows));

Constructor options: apiKey, proxyCountry (default "US"), retries (default 2), backoffMs (default 2000, doubles each retry), concurrency (parallel requests for multi-project calls, default 4), log (a progress callback) and client (inject a mock for tests).

Example output

kickstarter-scraper search -c "Tabletop Games" -s live --sort most_funded -l 1 (URLs shortened):

[
  {
    "id": 1878275683,
    "name": "Stormlight: War for Roshar",
    "blurb": "Command the forces of Honor and Odium in this strategic board game based on Brandon Sanderson's epic fantasy series!",
    "state": "live",
    "url": "https://www.kickstarter.com/projects/brotherwise/war-for-roshar",
    "category": "Tabletop Games",
    "category_slug": "games/tabletop games",
    "category_id": 34,
    "parent_category": "Games",
    "creator": "Brotherwise Games",
    "creator_id": 237961243,
    "creator_slug": "brotherwise",
    "creator_url": "https://www.kickstarter.com/profile/brotherwise",
    "location": "Los Angeles, CA",
    "location_id": 2442047,
    "location_type": "Town",
    "location_state": "CA",
    "location_country": "US",
    "country": "US",
    "currency": "USD",
    "currency_symbol": "$",
    "goal": 100000,
    "pledged": 2521276,
    "percent_funded": 2521.3,
    "backers_count": 10194,
    "average_pledge": 247.33,
    "usd_rate": 1,
    "goal_usd": 100000,
    "pledged_usd": 2521276,
    "staff_pick": true,
    "spotlight": false,
    "late_pledges": false,
    "created_at": "2026-05-17T00:19:05Z",
    "launched_at": "2026-10-13T13:00:19Z",
    "deadline": "2026-11-06T04:59:00Z",
    "state_changed_at": "2026-10-13T13:00:23Z",
    "duration_days": 23.7,
    "days_left": 9.2,
    "image_url": "https://i.kickstarter.com/assets/055/489/461/..._original.png?...",
    "video_url": "https://v2.kickstarter.com/.../video-1467688-h264_high.mp4"
  }
]

project returns the same fields plus comments_count, updates_count, tags, reward_count, add_on_count, rewards and add_ons. One reward tier looks like this:

{
  "id": 11415772,
  "type": "base",
  "title": "First Ideal",
  "minimum": 90,
  "currency": "USD",
  "description": "Speak the words and receive the retail version of the base game along with all unlocked base game stretch goals.",
  "backers_count": 211,
  "limit": null,
  "remaining": null,
  "available": true,
  "estimated_delivery": "2027-11",
  "shipping": "Only certain countries",
  "shipping_type": "multiple_locations",
  "starts_at": null,
  "ends_at": null,
  "items": ["Stormlight: War for Roshar"]
}

stats ... --history funding.jsonl -f csv (a second run adds the deltas):

id,state,backers_count,pledged,comments_count,url,fetched_at,backers_delta,pledged_delta,comments_delta,previous_at,hours_since
1878275683,live,10194,2521276,1111,https://www.kickstarter.com/projects/brotherwise/war-for-roshar,2026-10-28T00:47:48Z,,,,,
570152316,live,11838,1763869,968,https://www.kickstarter.com/projects/dicethrone/dice-throne-nexus,2026-10-28T00:47:48Z,,,,,

Good to know

• Money and currencies. goal, pledged, average_pledge and reward prices are in the project's own currency. goal_usd and pledged_usd use the USD rate Kickstarter fixed for the campaign (usd_rate), so you can compare projects across countries. stats reports pledged in the project currency.
• Paging. The discover search returns 12 projects per page and serves at most 200 pages (2,400 projects) per query. To collect more, split the query, for example by subcategory, state or goal range. Each response carries a ranking seed, which the scraper sends with every later page so the default magic order stays consistent while you page. Projects are deduped by ID.
• The goal and pledged ranges are Kickstarter's own filter buckets and are in USD.
• Upcoming projects have no launch date or deadline yet (both null).
• Reward limits. limit and remaining are only set for tiers with a quantity cap. starts_at and ends_at are only set for time-boxed tiers.
• Campaign story text is not included: it is not part of the project data the page ships.
• Counts from count() (and the success-rate-by-category example) are the totals Kickstarter's discover search reports. They cover the projects that search lists, so use them to compare categories, not as Kickstarter's official statistics.

Runnable examples

Script	What it does
examples/live-campaigns-to-csv.js	The most-funded live staff picks in a category, saved to CSV
examples/reward-tiers.js	Which reward tiers backers pick, by estimated share of pledges
examples/funding-tracker.js	A cron-friendly watchlist tracker built on stats and trackStats
examples/success-rate-by-category.js	Successful vs failed project counts per category

node examples/live-campaigns-to-csv.js Technology 2
node examples/reward-tiers.js brotherwise/war-for-roshar

Project layout

kickstarter-scraper/
├── bin/cli.js              # CLI entry point (argument parsing, output formats)
├── src/
│   ├── index.js            # Public exports
│   ├── scraper.js          # KickstarterScraper: fetching, retries, paging, dedupe
│   ├── parse.js            # Discover JSON, project pages, stats, locations -> records
│   ├── urls.js             # URL builders and the search filter vocabulary
│   ├── categories.js       # Category tree and name/slug/ID resolution
│   ├── history.js          # Funding tracker snapshots and deltas
│   └── util.js             # Numbers, dates, entities, CSV, concurrency pool
├── examples/               # Runnable scripts
├── test/                   # Offline tests with a mocked client (no API calls)
├── .github/workflows/ci.yml
├── .env.example
└── package.json

Development

npm install
npm test        # node:test, fully offline: the ScrapeUnblocker client is mocked
npm run lint    # eslint

The tests build small Kickstarter-shaped responses in test/fixtures.js and cover the parsers, the retry and paging logic, the funding tracker and the CLI argument parser, so they spend no API credit.

How it works

1. search builds a kickstarter.com/discover/advanced?...&format=json URL from your filters and fetches it with client.getPageSource(url). Kickstarter answers with a JSON feed of the matching projects.
2. project fetches the project page and decodes the project object the page embeds (window.current_project), which holds the funding numbers, creator, location, reward tiers and add-ons.
3. stats fetches <project>/stats.json, a small JSON document with the live counters. That makes it a cheap way to poll many campaigns.
4. Blocks, timeouts and responses without the expected data are retried with exponential backoff. A real 404 (an unknown project, or a page past the last one) is returned as "not found" and never retried.

Links

• 🌐 ScrapeUnblocker
• 📚 Documentation
• 📦 Node.js SDK

License

MIT © 2026 ScrapeUnblocker

---

This is an unofficial, educational example. "Kickstarter" is a trademark of Kickstarter, PBC. This project is not affiliated with or endorsed by Kickstarter. Scrape responsibly and respect each site's Terms of Service.