Caviar Email Scraper
Pricing
from $1.19 / 1,000 results
Caviar Email Scraper
Caviar Email Scraper SD - Caviar Email Scraper is a lead generation tool that extracts leads with public contact emails, account names and profile URLs from Caviar results by keyword, location and email domain - Caviar email extractor.
Pricing
from $1.19 / 1,000 results
Rating
0.0
(0)
Developer
Neuro Scraper
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
5 days ago
Last modified
Categories
Features
Share
Caviar Email Scraper
The Caviar Email Scraper finds restaurant owner email list entries and catering contact addresses that Google has already indexed alongside Caviar store pages on trycaviar.com.
Be clear on expectations before you spend a run. Caviar now runs on DoorDash infrastructure and exposes few distinct indexed pages, so expect very low volume; the DoorDash Email Scraper usually returns more.
Our own test run says the same thing. In a sample of about 10 parsed results, 6 rows carried an account identity, only 4 carried a profile URL, and just 3 unique emails were found.
That makes the Caviar Email Scraper the weakest of the non-zero Actors in this food-delivery batch. We are not going to dress those numbers up.
There is still one genuine reason to run it, and it is not volume. Caviar's catalogue skews toward premium and fine-dining restaurants, and its default keywords are restaurant and fine dining.
So the handful of rows you do get are a narrow, upmarket segment rather than a mass list. Position the Caviar Email Scraper as a small, high-end supplement to a bigger source, never as your primary restaurant contact database.
It fits suppliers selling premium ingredients, tableware, glassware or upscale hospitality services. Twenty chef-led rooms beat two thousand random takeaways when your product costs real money.
The Caviar Email Scraper does not log in, does not use any platform API, and does not open the Caviar website. There is no browser and no JavaScript rendering anywhere in it.
What the Caviar Email Scraper collects on this platform
Caviar store pages exist so diners can order, book and arrange large orders. Restaurants that want catering and event enquiries publish a business address on them.
That published address is the whole target. Because it is a business contact point, this is B2B restaurant outreach data rather than personal data harvesting.
The platform-specific catch is coverage. Since Caviar moved onto DoorDash infrastructure, a large share of its URLs resolve into the DoorDash catalogue instead of distinct trycaviar.com documents.
Fewer distinct documents means fewer snippets, and fewer snippets means fewer emails. That is the mechanism behind the low yield, and no scraper design changes it.
What survives in the index skews premium, because fine-dining rooms were always Caviar's positioning. So the Caviar Email Scraper is best understood as a cuisine segmentation tool with a very short output.
Key features of the Caviar Email Scraper
Every row below is behaviour the Caviar Email Scraper genuinely implements. Nothing here is aspirational, and nothing compensates for a thin index.
| Feature | What it does in the Caviar Email Scraper |
|---|---|
| Google SERP parsing | Locates each result's <h3> title, then the smallest surrounding block, so it never depends on Google's CSS class names. |
site: operator queries | Every search is scoped to trycaviar.com, so nothing off-platform enters the dataset. |
| Query expansion | Base, quoted and intitle: variants plus one per query modifier, for each keyword x domain pair. Base queries run first. |
| Domain filtering | Only emails ending in your customDomains values are kept. gmail.com and @gmail.com both work. |
| Global email deduplication | An address is pushed once, across every query and every page of the run. |
| Obfuscated email decoding | Understands name [at] domain [dot] com, name (at) domain, name @ domain.com, domain .com, zero-width characters and the full-width @ sign. |
| Junk filter | Rejects placeholders such as email@, yourname@, test@, xxx@ and single-character locals. |
| Boundary-correct matching | @gmail.com never matches inside @gmail.company or @gmail.com.br. |
| Soft-wrap repair | Drops a hit that is only the tail of another email in the same block. |
| Concurrency control | An asyncio worker pool runs several queries in parallel with a shared stop signal on maxEmails. |
| Exponential backoff | Up to 3 attempts per page, with a fresh proxy session per request. |
| CAPTCHA detection | Consent, "unusual traffic" and challenge pages are detected and retried, not counted as empty. |
| Failed-query requeue | Blocked or failed queries are re-queued once at the end of the run. |
| Resumable runs | State lives in the key-value store, keyed by a hash of the input, and saves on PERSIST_STATE, MIGRATING and ABORTING. |
| Fallback parser | If Google's markup changes, the whole page is scanned, so a layout change degrades to "emails without account details" rather than "no emails". |
| Immediate dataset push | Each lead lands in the dataset as it is found, ready for CSV, JSON or Excel export. |
| Run summary | Logs pages fetched, blocked pages, retries and emails per page. |
How the Caviar Email Scraper works
Six steps, no browser, no login.
- Read input. Keywords, location, email domains and limits are loaded from the run configuration.
- Build queries. The Caviar Email Scraper composes Google searches with the
site:operator againsttrycaviar.com, for examplesite:trycaviar.com fine dining "@gmail.com" "San Francisco". - Fetch SERP pages. Result pages are pulled through the Apify GOOGLE_SERP proxy with async
aiohttprequests. - Parse structurally. Each result block is found by locating the
<h3>and taking the smallest block that surrounds it. - Extract emails. A domain-filtered regular expression pulls addresses out of the block text, with obfuscated email decoding applied.
- Deduplicate and push. Addresses are deduplicated globally, then every lead goes straight into the Apify dataset.
Snippet extraction is the whole game here. If Google never printed the address in a snippet, the Caviar Email Scraper cannot find it, because it only ever reads Google.
Input for the Caviar Email Scraper
These are the exact nine fields, with the types and defaults the Caviar Email Scraper ships with.
| Field | Type | Default | Meaning |
|---|---|---|---|
keywords | array (required) | ["restaurant", "fine dining"] | Search terms describing the Caviar listings you want. |
location | string | "" | Optional location phrase added to every query. |
customDomains | array | ["@gmail.com","@yahoo.com"] | Only emails on these domains are kept; the @ is optional. |
maxEmails | integer 1-10000 | 20 | Stop after this many unique emails. |
countryCode | string | "" | Two-letter country for the search proxy (US, GB, DE...). |
expandQueries | boolean | true | Search each keyword x domain pair in several phrasings. |
queryModifiers | array | ["email","contact","catering","reservations","owner"] | Extra words combined with each keyword when expansion is on. |
maxPagesPerQuery | integer 1-50 | 30 | Page cap per query. |
maxConcurrency | integer 1-20 | 5 | Parallel queries. |
The default modifiers are tuned for a food platform. catering, reservations and owner are the three that most often surface a real inbox on a fine-dining page.
JSON input example for the Caviar Email Scraper
{"keywords": ["restaurant", "fine dining"],"location": "San Francisco","customDomains": ["@gmail.com", "@yahoo.com"],"maxEmails": 20,"countryCode": "US","expandQueries": true,"queryModifiers": ["email", "contact", "catering", "reservations", "owner"],"maxPagesPerQuery": 30,"maxConcurrency": 5}
Keep maxEmails modest. Asking for 500 addresses does not create supply; it just makes the run spend longer proving there is none.
Getting location right
location is a plain string appended to every query, and restaurant prospecting is inherently local, so it is the most valuable lever you have.
Run one city per run and keep the datasets separate. A sales rep then gets a territory-shaped list instead of a national blob.
For dense metros, use the neighbourhood or borough rather than the city: "Brooklyn", "SoMa", "West Loop". City-level prospecting gets sharper the narrower you go.
Always pair location with countryCode so the search proxy resolves the right Google locale. For the Caviar Email Scraper that means countryCode: "US" with a major US metro.
San Francisco, New York, Los Angeles, Chicago, Seattle, Boston and Washington DC are the markets worth trying first.
Caviar's footprint covers a relatively small set of cities, so a smaller city will very likely return nothing. That is coverage, not a broken query.
Remember that location is matched as text in the indexed page. A city the platform never prints on its pages simply narrows results to zero.
Widen to a state or region only when a city returns too little, and accept that a wider net on Caviar still catches very few fish.
Two or three empty metros in a row are your signal to stop. Move the campaign to a broader food Actor rather than spending more compute on Caviar Email Scraper query expansion.
Output of the Caviar Email Scraper
Every dataset item the Caviar Email Scraper pushes carries all fourteen fields below, in this order.
| Field | Meaning |
|---|---|
network | Platform name. |
keyword | The keyword that produced the lead. |
query | The exact Google query used. |
title | Raw result title. |
accountName | Account label Google prints for the listing. |
fullName | Display name parsed from a profile-style title; empty for other result types. |
username | URL-safe handle when Caviar exposes one; otherwise null. |
profileUrl | Canonical store URL when a handle is known; otherwise empty. |
url | Direct Caviar link when exposed, else the profile URL. |
description | Snippet text, cleaned of labels and counters. |
email | Lower-cased email address. |
emailDomain | The matched domain, for example @gmail.com. |
possiblyTruncated | true when Google's snippet ellipsis touched the email. |
foundAt | ISO 8601 UTC timestamp. |
JSON output example from the Caviar Email Scraper
Profile URL coverage was only 4 in 10 in our test run, so the sample below deliberately includes a row with no handle. That is what a real dataset looks like.
[{"network": "Caviar","keyword": "fine dining","query": "site:trycaviar.com fine dining \"@gmail.com\" \"San Francisco\"","title": "Nocturne Dining Room - San Francisco | Caviar","accountName": "Nocturne Dining Room","fullName": "Nocturne Dining Room","username": "nocturne-dining-room","profileUrl": "https://www.trycaviar.com/store/nocturne-dining-room","url": "https://www.trycaviar.com/store/nocturne-dining-room","description": "Seasonal tasting menu and wine pairings. Private events and catering: nocturne.catering@gmail.com","email": "nocturne.catering@gmail.com","emailDomain": "@gmail.com","possiblyTruncated": false,"foundAt": "2026-08-30T09:12:44+00:00"},{"network": "Caviar","keyword": "restaurant","query": "site:trycaviar.com restaurant catering \"@gmail.com\"","title": "Catering and large party orders | Caviar","accountName": "Maison Verte","fullName": "","username": null,"profileUrl": "","url": "","description": "Catering enquiries for parties of 10 or more - maison.verte.events@gmail.com ...","email": "maison.verte.events@gmail.com","emailDomain": "@gmail.com","possiblyTruncated": true,"foundAt": "2026-08-30T09:14:02+00:00"},{"network": "Caviar","keyword": "fine dining","query": "site:trycaviar.com intitle:\"fine dining\" \"@yahoo.com\"","title": "Harbour and Ash - New York | Caviar","accountName": "Harbour and Ash","fullName": "Harbour and Ash","username": "harbour-and-ash","profileUrl": "https://www.trycaviar.com/store/harbour-and-ash","url": "https://www.trycaviar.com/store/harbour-and-ash","description": "Chef-led seafood room. Reservations and press enquiries: harbour.ash.nyc@yahoo.com","email": "harbour.ash.nyc@yahoo.com","emailDomain": "@yahoo.com","possiblyTruncated": false,"foundAt": "2026-08-30T09:16:35+00:00"}]
Row two shows both honest gaps at once. No handle, so username is null and profileUrl is empty, plus a snippet ellipsis that set possiblyTruncated to true.
Use cases for the Caviar Email Scraper
Every Caviar Email Scraper use case here assumes a short, upmarket list rather than a bulk one. Read the table as "who benefits from three good rows", not "who needs three thousand".
| Buyer type | How they use the Caviar Email Scraper |
|---|---|
| Food distributors and wholesale suppliers | Reach fine-dining kitchens with seasonal produce, seafood and speciality dairy offers. |
| Packaging and equipment suppliers | Build a short prospecting list of chef-led rooms that replace tableware, glassware and linens each season. |
| Restaurant-tech vendors | Add a premium slice to a pipeline sourced mainly from broader food Actors: POS, online ordering, reservations. |
| Hospitality recruiters | Identify upmarket kitchens publishing a contact address, for chef, GM and front-of-house searches. |
| Local marketing and SEO agencies | Approach independent premium restaurants that already invest in their own brand. |
| Catering and events agencies | Use the catering and reservations modifiers to surface event-facing inboxes and generate catering leads. |
| Food PR and media teams | Assemble a short outreach list of new fine-dining openings for launch coverage. |
| Market researchers | Sample how premium restaurants present contact details across delivery platforms. |
For any of these, run the Caviar Email Scraper alongside a higher-yield source. The Grubhub Email Scraper and the ChowNow Email Scraper cover far more independent restaurants.
The sensible workflow is bulk first, premium second. Build the restaurant email list elsewhere, then use a Caviar email extractor pass to tag the upmarket names inside it.
Example runs and recipes for the Caviar Email Scraper
Recipe 1 - premium catering leads in one metro. Keywords ["fine dining", "tasting menu"], location: "San Francisco", countryCode: "US", modifiers including catering. Expect single digits.
Recipe 2 - widen the domains before the keywords. Same keywords, but customDomains extended to ["@gmail.com","@yahoo.com","@outlook.com","@hotmail.com","@icloud.com"]. This is the cheapest way to lift a thin run.
Recipe 3 - borough-level sweep. Run "Manhattan", then "Brooklyn", then "Queens" as three separate Caviar Email Scraper runs, each with its own dataset, for territory-based restaurant lead generation.
Recipe 4 - cuisine segmentation. Keywords ["steakhouse", "omakase", "wine bar"] with countryCode: "US". Narrow terms match the premium catalogue better than generic food words.
Recipe 5 - supplement, not source. Run Grubhub or ChowNow for volume first, then a small Caviar Email Scraper run, and merge on the email column so Caviar rows enrich existing records rather than duplicating them.
For CRM import, export the dataset as CSV, JSON or Excel, or pull it through the Apify API. Lead enrichment works best when the Caviar row adds a premium tag to a contact you already had.
Keep maxConcurrency at the default of 5 unless you have a reason to change it. Higher concurrency against a thin catalogue mostly buys you more blocked pages.
Responsible use, GDPR and CAN-SPAM
The addresses the Caviar Email Scraper collects are business contacts that restaurants published deliberately, so diners and partners can arrange catering, large orders and reservations.
Nothing in the Caviar Email Scraper output is private. Every field is copied from a public Google result, and the small dataset size makes personalised, low-volume outreach the obvious approach anyway.
Treat them as a business channel. Identify yourself and your company, state plainly why you are contacting that business address, and make the relevance obvious in the first line.
Give a working unsubscribe link in every message, honour opt-outs immediately, and keep a record of them. Do not send to consumers, and do not resell the output as a consumer list.
Keep a lawful basis or legitimate-interest record for your processing, and verify possiblyTruncated rows before they enter a send.
This is not legal advice. Check your own obligations under GDPR, CAN-SPAM and any local rules that apply to your market before you launch cold email outreach.
Limitations of the Caviar Email Scraper
Stated in the order that will actually affect a Caviar Email Scraper run.
- Very low volume. Caviar now runs on DoorDash infrastructure and exposes few distinct indexed pages, so expect very low volume; the DoorDash Email Scraper usually returns more. In our test run of about 10 parsed results: 6 with an account identity, 4 with a profile URL, 3 unique emails.
- Public index only. The Caviar Email Scraper finds only emails that are publicly visible in Google's index. An address Google has not crawled cannot be found.
- Google's ~300 result ceiling. Google caps a single query at roughly 300 results. That is exactly why query expansion exists, and why switching it off shrinks an already small run.
- Truncated snippets.
possiblyTruncated: truemeans Google's snippet ellipsis may have cut the email. Verify those rows before sending. - Apify proxy required. The Actor requires the Apify GOOGLE_SERP proxy and cannot run without Apify proxy credentials.
- Free plan cap. Free Apify plans are capped at 100 emails per run; paid plans are uncapped. On Caviar that cap is rarely the binding constraint.
- Sparse handles.
usernameandprofileUrlare populated only when the platform exposes a handle in the Google result; otherwise you getaccountNameandfullNamewith an empty handle. This is a Google limitation, not a bug. - No guaranteed volume. Results vary with keywords, domains and location, and no volume is guaranteed. On this platform that sentence is not boilerplate.
Caviar Email Scraper FAQ
How many results should I expect from the Caviar Email Scraper?
Very few. In our own test run, a sample of about 10 parsed results gave 6 rows with an account identity, 4 with a profile URL, and 3 unique emails. Caviar now runs on DoorDash infrastructure and exposes few distinct indexed pages, so the DoorDash Email Scraper usually returns more. Treat those figures as an observation, not a promise.
Then why run the Caviar Email Scraper at all?
Because the catalogue skews toward premium and fine-dining restaurants. The rows you get are a narrow, upmarket segment, which is genuinely useful as a supplement and useless as a primary source.
What should I use instead for volume?
The Grubhub Email Scraper and the ChowNow Email Scraper cover far more independent restaurants, and the DoorDash Actor covers the same underlying network more broadly.
Where do the emails come from?
Publicly indexed Caviar pages, as they appear in Google's titles, snippets and site labels. The Caviar Email Scraper does not log in, does not use any platform API, and does not open the Caviar website.
Why is username sometimes null and profileUrl empty?
Google did not expose a store handle in that result. The row still carries accountName, the email and the snippet, so it remains usable for outreach; it simply lacks a canonical https://www.trycaviar.com/store/... link.
What does possiblyTruncated: true mean?
Google's snippet ellipsis touched the email, so the address may be cut short. Check those rows by hand before adding them to a campaign.
Do I need my own proxies to run the Caviar Email Scraper?
No, but you do need an Apify account with proxy access. The Actor uses the Apify GOOGLE_SERP proxy and cannot run without those credentials.
Which location should I set for the Caviar Email Scraper?
A major US metro, paired with countryCode: "US". Caviar's footprint covers a relatively small set of cities, so a smaller city will very likely return nothing.
My run returned two emails. Is it broken?
Almost certainly not. Add more domains, try another metro, and check the run summary for blocked pages. Two is within the normal range for this Actor.
Can I raise maxEmails to get more restaurant marketing leads?
You can, up to 10000, but it will not manufacture supply. maxEmails is a stop condition, not a target, and on Caviar the results usually run out first.
Are Caviar Email Scraper runs resumable?
Yes. Progress is saved to the key-value store keyed by a hash of your input, so a migrated or resurrected run continues instead of starting over.
How do I get the data out?
Export the dataset as CSV, JSON or Excel from the Apify console, or pull it through the Apify API for CRM import. The Caviar Email Scraper pushes each row the moment it is found, so a partial run is still exportable.
Is the Caviar Email Scraper affiliated with Caviar or DoorDash?
No. It is an independent Actor that reads public Google search results, with no connection to either company.
Can I use the Caviar Email Scraper for hospitality lead generation outside the US?
You can try, but coverage is US-centric and thin. For other markets, start with one of the sibling Actors listed below.
Related Actors
The full family of email scrapers built on the same engine.
| Actor | What it collects |
|---|---|
| Caviar Email and Phone Number Scraper | Emails and phone numbers from Caviar |
| Caviar Phone Number Scraper | Public phone numbers from Caviar |
| Airbnb Email Scraper | Public contact emails from Airbnb |
| Booking.com Email Scraper | Public contact emails from Booking.com |
| ChowNow Email Scraper | Public contact emails from ChowNow |
| Contiki Email Scraper | Public contact emails from Contiki |
| Craigslist Email Scraper | Public contact emails from Craigslist |
| Delivery.com Email Scraper | Public contact emails from Delivery.com |
| DoorDash Email Scraper | Public contact emails from DoorDash |
| EatStreet Email Scraper | Public contact emails from EatStreet |
| G Adventures Email Scraper | Public contact emails from G Adventures |
| GetYourGuide Email Scraper | Public contact emails from GetYourGuide |
| Grubhub Email Scraper | Public contact emails from Grubhub |
| Hostelbookers Email Scraper | Public contact emails from Hostelworld |
| HotelsCombined Email Scraper | Public contact emails from HotelsCombined |
| Intrepid Travel Email Scraper | Public contact emails from Intrepid Travel |
| Just Eat Email Scraper | Public contact emails from Just Eat |
| Klook Email Scraper | Public contact emails from Klook |
| OpenTable Email Scraper | Public contact emails from OpenTable |
| Resy Email Scraper | Public contact emails from Resy |
| TheFork Email Scraper | Public contact emails from TheFork |
| Toast Restaurant Email Scraper | Public contact emails from Toast |
| TourRadar Email Scraper | Public contact emails from TourRadar |
| Traveloka Email Scraper | Public contact emails from Traveloka |
| Tripadvisor Attractions Email Scraper | Public contact emails from Tripadvisor |
| Tripadvisor Hotels Email Scraper | Public contact emails from Tripadvisor |
| Tripadvisor Restaurants Email Scraper | Public contact emails from Tripadvisor |
| Uber Eats Email Scraper | Public contact emails from Uber Eats |
| Viator Email Scraper | Public contact emails from Viator |
| Wanderlog Email Scraper | Public contact emails from Wanderlog |
| Wego Email Scraper | Public contact emails from Wego |
| Zomato Email Scraper | Public contact emails from Zomato |
| Airbnb Email and Phone Number Scraper | Emails and phone numbers from Airbnb |
| Booking.com Email and Phone Number Scraper | Emails and phone numbers from Booking.com |
| ChowNow Email and Phone Number Scraper | Emails and phone numbers from ChowNow |
| Contiki Email and Phone Number Scraper | Emails and phone numbers from Contiki |
| Craigslist Email and Phone Number Scraper | Emails and phone numbers from Craigslist |
| Delivery.com Email and Phone Number Scraper | Emails and phone numbers from Delivery.com |
| DoorDash Email and Phone Number Scraper | Emails and phone numbers from DoorDash |
| EatStreet Email and Phone Number Scraper | Emails and phone numbers from EatStreet |
| G Adventures Email and Phone Number Scraper | Emails and phone numbers from G Adventures |
| Grubhub Email and Phone Number Scraper | Emails and phone numbers from Grubhub |
Leave a review
If the Caviar Email Scraper saved you time, please leave a star rating and a short review on the Actor page.
Reviews are how other buyers judge whether a tool works, and they tell us which features to build next.
If something did not work, email neurodata.apify@gmail.com instead - bugs get fixed faster than they get complained about.