Makuake Project Scraper

Find and collect structured public Makuake project data for crowdfunding market research, competitor tracking, launch benchmarking, and recurring monitoring.

Use keywords when you want the Actor to discover projects for you. Keep startUrls empty, or combine keywords with project, Discover, ranking, category, or tag URLs when you need a controlled seed set.

The Input tab is prefilled with one public project URL so a first test run stays small and finishes quickly. Clear or replace that sample URL before a keyword-only run. The prefill affects only the Console input form; Saved Tasks and API calls keep their supplied inputs and defaults.

What this Actor does

The Actor uses a two-stage workflow:

1. Discover candidate projects from Makuake keyword searches and optional Start URLs.
2. Merge duplicates, filter and rank listing cards, select the requested Top N, and visit only those project pages for full details.

The default dataset contains one row per successfully collected project. Page and run audits are stored separately in the run's key-value store, so audit records do not become dataset results.

Keyword search examples

Find up to 100 car-refrigerator-related projects and rank them by the funding amount shown on Makuake listing cards:

{
  "keywords": ["車載冷蔵庫", "ポータブル冷蔵庫"],
  "excludeKeywords": ["マグネット", "ステッカー"],
  "matchMode": "any",
  "keywordMatchScope": "title",
  "status": "all",
  "sortBy": "amount_raised",
  "resultsWanted": 100,
  "candidateLimit": 500,
  "maxPages": 10,
  "includeStory": true
}

For broader discovery, set keywordMatchScope to makuake_search. This keeps projects returned by Makuake even when the literal keyword does not appear in the listing title, but it can also retain more false positives.

What "popular" means

Makuake keyword search does not expose a reliable official popularity sort. In this Actor, sortBy: "amount_raised" uses the funding amount shown on each discovered listing card as a transparent popularity proxy. It is not an official Makuake popularity score.

sortBy: "relevance" preserves Makuake's result/listing order instead. Funding values and order are snapshots from the collection time.

Start URL examples

startUrls is optional and accepts:

• A project URL, such as https://www.makuake.com/project/example/
• A Discover URL, such as https://www.makuake.com/discover/all
• A ranking URL, such as https://www.makuake.com/discover/most-funded
• A category URL, such as https://www.makuake.com/discover/categories/product/
• A tag URL, such as https://www.makuake.com/discover/tags/94

URL-only run:

{
  "startUrls": [
    { "url": "https://www.makuake.com/discover/categories/product/" }
  ],
  "resultsWanted": 50,
  "maxPages": 10,
  "includeStory": false,
  "includeImages": true
}

Mixed keyword and URL run:

{
  "keywords": ["車載冷蔵庫"],
  "startUrls": [
    { "url": "https://www.makuake.com/project/example/" },
    { "url": "https://www.makuake.com/discover/categories/product/" }
  ],
  "sortBy": "amount_raised",
  "resultsWanted": 100
}

Direct project URLs are always kept ahead of discovered listing candidates and bypass keyword, exclusion, and status filters. They still count toward the global resultsWanted limit. If both keywords and Start URLs are omitted, the Actor falls back to https://www.makuake.com/discover/all for backward compatibility.

Input controls

Field	Purpose
keywords	Up to 20 unique Makuake search terms or phrases. One search listing is opened per keyword.
excludeKeywords	Excludes non-direct candidates when any term appears in the listing title.
matchMode	any keeps a candidate matching at least one keyword; all requires every keyword.
keywordMatchScope	title uses literal listing-title matching; makuake_search trusts membership in Makuake search results.
status	Keeps all, active, ended, or coming-soon listing candidates.
sortBy	amount_raised ranks by the listing funding snapshot; relevance preserves discovery order.
startUrls	Optional project or public listing URLs. Multiple sources can be combined.
resultsWanted	Global maximum after source merging, filtering, and ranking.
candidateLimit	Maximum unique listing candidates retained before filtering and ranking.
maxPages	Maximum scroll/load-more cycles per listing page.
includeStory	Adds long project-detail text. Off by default.
includeImages	Adds project-detail image URLs. On by default.
maxConcurrency	Parallel browser pages, limited to 1-10.
requestDelayMs	Polite delay before each page is processed.
proxyConfiguration	Apify Proxy settings. Cloud requests without a proxy may be blocked.

candidateLimit is automatically raised to at least resultsWanted. A run can return fewer than requested when the discovered pool is smaller, candidates fail filters, or project pages fail to load.

How it compares with Kickstarter scrapers

The input follows the familiar Kickstarter Discover workflow, while adding keyword discovery tailored to Makuake's public search pages.

Kickstarter scraper pattern	Makuake equivalent in this Actor
Kickstarter keyword or Discover search	Makuake keyword search plus optional Discover, ranking, category, or tag URLs
Project ID or slug deduplication	Stable Makuake project slug deduplication across all sources
Goal, pledged, backers, funded percentage	Amount raised in JPY, supporters, achievement rate
Creator and category metadata	Makuake creator and category metadata
Project deadline and state	Available start/end fields plus normalized active, ended, or coming-soon status
Result and page limits	resultsWanted, candidateLimit, and maxPages safety caps
Proxy configuration	Apify Proxy configuration; recommended for cloud runs

Makuake does not expose the same field set or navigation model as Kickstarter. Empty fields mean the public page did not expose a reliable value at collection time; the Actor does not invent missing goals, currencies, locations, or dates.

Output

Storage	Record	Meaning
Default dataset	project rows	One structured row per unique successfully collected Makuake project.
Key-value store	PAGE_AUDITS	Listing/search audit objects with query keyword, links found, new candidates, expansion cycles, and success status.
Key-value store	RUN_AUDIT	Inputs, candidate/filter/selection totals, exclusions, collection totals, and failures.

All three outputs are linked from the run's Output tab.

Project records can include:

• Project ID, slug, title, description, canonical URL, and main image
• Creator and category names and URLs
• Amount raised in JPY, supporter count, achievement rate, remaining text, and normalized status
• Start/end timestamps when reliably exposed
• Tags, project image URLs, and optional story text
• Matched keywords and all source listing URLs
• Discovery rank, final selection rank, listing funding snapshot, selection sort, and scrape timestamp

Validation checklist

1. Confirm the RUN_AUDIT.failures list is empty or review each failed URL.
2. Compare unique_projects_discovered, projects_after_filters, projects_selected, and projects_collected.
3. Check each PAGE_AUDITS item for success: true and a plausible project count.
4. Keep matched_keywords, source_list_urls, and selection_rank when ranking or sampling matters.
5. Treat amount and supporter values as a snapshot at scraped_at, not final campaign results unless the project has ended.

Limits and responsible use

• Collects public Makuake pages only; it does not log in or access account-specific data.
• Makuake can block cloud datacenter traffic. Apify Proxy is recommended, but no proxy guarantees access.
• Search and listing pages load dynamically, so maxPages is a safety cap rather than a guaranteed page count.
• Empty or failed pages are retried once. Final failures include compact page diagnostics in RUN_AUDIT.failures.
• Keyword title matching is literal after whitespace and Japanese Unicode normalization; it is not semantic search.
• Listing-card status and amount are used for pre-detail filtering and ranking. The detail-page fields remain the final collected snapshot.
• Long story text and image arrays increase storage and transfer usage.
• Page changes can require parser maintenance; always inspect audit records.
• Use reasonable concurrency and comply with applicable laws, Makuake terms, and your organization's data-handling rules.