James Beard Awards Scraper - Chef & Restaurant Winners

Scrape the complete James Beard Foundation Awards archive (1990-present): every winner, nominee and semifinalist across Restaurant & Chef, Book, Journalism, Broadcast Media, Leadership and Achievement categories. 16,000+ records. Refresh annually or on semifinalist announcement.

Built for restaurant PR firms, hospitality investors, chef recruiters, food media journalists, culinary schools, restaurant lead-gen platforms.

What this jamesbeard.org scraper extracts

Every run produces one row per record, with change annotations against your last snapshot, with the following fields:

• object_id (string), Algolia record id (Salesforce origin)
• year (integer)
• parent_category (string), Restaurant & Chef, Book, Journalism, Broadcast Media, Leadership, Achievement
• category (string), Specific award category (e.g. Emerging Chef, Outstanding Restaurateur)
• status (string), Winner, Nominee, or Semifinalist
• name (string), Honoree or entity name
• restaurant (string)
• city (string)
• state (string)
• type (string), individual or organization
• summary (string), Pre-formatted nomination display string

Quick start

1. Click Try for free.
2. Leave the input at defaults, or paste the recommended input from lower on this page.
3. Click Start. The default run finishes in under a minute with a small sample.
4. Open the Dataset tab to inspect the shape.
5. Once you like the output, click Schedule and pick a cadence. Daily is typical.

How a run works

1. Scan, walks jamesbeard.org from the seed URL.
2. Fetch, visits each detail page and extracts the fields listed above.
3. Diff, compares each record against the snapshot in your named key-value store, labelling it new, changed or unchanged.
4. Persist, the new snapshot is saved for the next run.

What is a delta feed and why does it save money?

A delta feed only pays for what changed since your last run. Instead of re-scraping every record each night, this scraper compares against your stored snapshot and only fully processes rows that added, changed or first appeared. Enable onlyChanged to skip unchanged rows entirely (they are billed at a discounted rate).

What differentiates this scraper?

Definitive US culinary award archive with 30+ years of history and zero direct competitors on the Apify Store. Uses the same public Algolia index that powers the JBF search UI, so records are always in sync with what appears on jamesbeard.org.

Common use cases

• Annual monitoring of new winners + semifinalist announcements
• Historical status-progression analysis (semifinalist to finalist to winner)
• Restaurant PR firm prospecting by award category and geography
• Chef recruiter shortlists by category and year
• Regional culinary trend analysis (which cities/states rack up which categories)

How to interpret the fields

What does Award status mean?

The James Beard Awards work in three tiers: Semifinalist (about 20-30 per category, announced January), Nominee/Finalist (5 per category, announced March), Winner (1 per category, announced June). This scraper's status field reflects the current level.

What does Parent category vs category mean?

Parent categories are the top-level buckets (Restaurant & Chef, Book, Journalism, Broadcast Media, Leadership, Achievement). Category is the specific award within (e.g. Emerging Chef, Outstanding Restaurateur, Best Chef Northeast). Filter on parent_category first, then category, for cleanest slice.

How this compares to alternative scrapers

James Beard scraper vs Michelin Guide scraper

Different signals. Michelin is European/international with star ratings and Bib Gourmand tiers. James Beard is US-centric with a chef-industry-jury lens. PR firms and investors typically track both; feed the winner lists into your CRM as separate cohorts.

James Beard scraper vs Eater 38 scraper

Eater 38 tracks essential-restaurant lists per city, refreshed by local editors. JBF is national/regional and jury-selected. Complementary: Eater signals current-year buzz; JBF signals sustained excellence.

James Beard scraper vs the JBF site's own search

The jamesbeard.org search UI is fine for one-off lookups but rate-limited, non-exportable, and hard to snapshot. This scraper hits the same Algolia index that powers that UI, returns structured JSON at scale, and supports delta mode so you can track status progression year over year.

James Beard scraper vs a Michelin + Zagat + Eater combo

Each of these is a different lens. Michelin = European-style star rating (data-driven for a European buyer). Zagat = crowd rating (US mainstream). Eater = editorial buzz. JBF = industry jury. Enterprise CRM cohorts typically merge two or three signals; JBF is the "sustained-excellence" lens.

How much does it cost to scrape jamesbeard.org?

Pay-per-event pricing:

• Actor start: $0.00500 per GB of memory, one event per run.
• Award record: $0.00200 per record delivered to the dataset.
• Award checked, unchanged: $0.00050 per row when onlyChanged is on and the row was unchanged. Roughly 25% the cost of a full record.

There are no per-request charges and no page-count charges. Costs come only from records you actually receive. A scheduled monitor in delta mode typically costs cents per day; a first-time full crawl is the expensive case that the delta model is designed to avoid on subsequent runs.

Concrete cost examples

Rule-of-thumb cost per run (assumes primary event only, no delta discount):

• 100 records = about $0.20 in records + one actor-start fee
• 500 records = about $1.00
• 1,000 records = about $2.00
• 10,000 records = about $20.00

With delta mode + onlyChanged: true, a daily monitor of 1,000 properties where 5% change costs roughly $0.58 per run (50 full charges + 950 delta-priced skips).

Common questions

Does this scraper need a James Beard account?

No. All records are pulled from the public Algolia index that powers the awards search on jamesbeard.org.

How often does the archive update?

New records land on the semifinalist announcement (typically January) and the winner ceremony (typically June). A monthly schedule catches both cycles; weekly is sufficient for real-time coverage during award season.

What is the status progression?

Semifinalist to Nominee (finalist) to Winner. This scraper returns the current status; run on a schedule to track which nominees advance each year.

Are chef bios and photos included?

No. The public Algolia index does not carry bio text or photo URLs. Only structured award metadata (year, category, name, restaurant, city, state, status).

How many total records?

About 16,240 records spanning 1990-2026. Restaurant & Chef is the largest bucket at 10,939 records; Book, Journalism and Broadcast are each about 1,700-1,800.

How do I use JBF winner data for restaurant PR pitching?

Pull the annual Winner cohort, filter by parent_category = 'Restaurant & Chef', then split by state/city. PR firms typically pitch trade press and local dailies within two weeks of the June ceremony. Semifinalist announcements (January) and Finalist announcements (March) are separate pitching windows.

Can I track the same restaurant across multiple years?

Yes. Restaurants often appear as Semifinalists for several years before winning, or win multiple categories over their history. Group records by the restaurant field, sort by year ascending, to see the full arc.

Are semifinalist announcement dates predictable?

Yes. Semifinalists are announced in late January, Finalists in late March, Winners at the June ceremony. Run the scraper on a schedule around each of those three windows for real-time coverage.

What's the difference between the JBF Awards and the JBF Cookbook Awards?

The Book category (parent_category='Book') covers cookbooks, food writing, and reference works. Winners and finalists are announced on the same June date but the shortlist logic is separate from the restaurant-chef side.

Can I detect newly-added award categories year-over-year?

Yes. Group by year + parent_category + category, diff against the prior snapshot. JBF occasionally introduces new categories (Emerging Chef, Best New Restaurant, Outstanding Bakery, etc.) which show up as brand-new category values for a given year.

Does the scraper cover the JBF Impact Awards and Leadership Awards?

Yes. Both Impact Awards (organizational recognition for humane immigrant rights, food workers, etc.) and Leadership Awards (individual honorees for lifetime advocacy) are in the Achievement / Leadership parent categories.

How much does it cost to scrape the full 16,240-record archive?

Under $1 in most cases — Algolia is fast, memory footprint is tiny, and there is no proxy cost because the endpoint is public. A weekly incremental refresh with delta mode enabled costs pennies per run.

Does this scraper need an account with jamesbeard.org?

No. The scraper reads only public pages. You do not need credentials.

Why do I need to enable Apify Proxy?

Most sites at scale trigger rate limits or challenges against datacenter IPs. Apify Proxy (residential recommended) handles this. The default input keeps useApifyProxy: true.

Why does the state store need a name?

Delta mode compares each run against the previous snapshot. Apify wipes the default (unnamed) key-value store on every run, which would erase the snapshot. A NAMED store (stateStoreName: "james-beard-scraper-state") persists across runs.

Can I run in delta mode without paying for unchanged rows?

Yes. Set onlyChanged: true. Unchanged rows are billed at the discounted record-checked rate; changed rows at the full award rate. New rows are always full price.

Recommended daily-schedule input

{
  "yearFrom": 2024,
  "yearTo": 2026,
  "parentCategories": [],
  "statuses": [],
  "maxItems": 5
}

For a full-catalogue first run, raise maxItems to 0 (no cap) and leave onlyChanged: false. For daily monitoring, set onlyChanged: true and keep maxItems at 0.

Turn it into a live alert with Schedules and Webhooks

Combine this scraper with an Apify Schedule and a Webhook to get "new listing in my buy box" alerts without writing code.

1. Configure the input with your buy-box filters (yield floor, price ceiling, markets, etc).
2. Create a Schedule (Actor page, right rail, Schedule) at your cadence. Daily 06:00 UTC is typical.
3. Add a Webhook (Integrations tab on this actor):
   • Event: ACTOR.RUN.SUCCEEDED
   • URL: your Slack / Discord / Zapier / n8n endpoint
   • Payload template: pass resource.defaultDatasetId so the receiver can pull only the delta rows

For Slack, wrap the payload in a text field and push through an incoming webhook. For Zapier or n8n, use the "Webhook by Apify" trigger and map dataset fields to your downstream tool of choice.

Pair this with onlyChanged: true and the actor becomes a pure alert feed. Delta-priced rows keep the bill flat on quiet days.

Related scrapers from this publisher

See other actors under growth-installations for related consumer-intelligence scrapers.

Video walkthrough

Video walkthrough placeholder. Add a YouTube URL to readme.video_url in the yaml to auto-embed once one is published.

Notes and limits

• Keep concurrency modest to stay polite to jamesbeard.org. Default is 5.
• Field schemas can shift when the target site restyles. If a field starts returning null, report it and it will be patched.
• The default input is engineered to complete in under 5 minutes so the platform's daily automated test passes and the actor does not get flagged for maintenance.

Legal

Data is pulled from the public Algolia index that powers jamesbeard.org's awards search. No login or scraping of gated content. You are responsible for compliance with applicable law and JBF's terms of service.