Naver Blog Scraper

Search Naver Blog for any keyword and get back structured data for every matching post — title, author, publish date, like/comment counts, and the full text of the post.

Naver Blog is South Korea's largest blogging platform. This Actor is built for anyone who needs Korean blog content at scale without manually searching and copy-pasting: market researchers, brand/PR teams monitoring mentions, content marketers scouting trends, and anyone doing NLP/sentiment analysis on Korean-language content.

What it does

1. You give it one or more search keywords and a date range.
2. It searches Naver Blog for each keyword and walks through the paginated search results (up to a page limit you set).
3. For every blog post found, it opens the post and extracts the full page text, along with its like count and comment count.
4. Every result is streamed to the dataset as soon as it's scraped, so you can start using results before the run finishes.

Input

Field	Type	Required	Default	Description
keywords	array of strings	Yes	—	Keywords to search for on Naver Blog. Each keyword is searched independently.
date_start	string (YYYY-MM-DD)	Yes	—	Only include posts published on or after this date.
date_end	string (YYYY-MM-DD)	Yes	—	Only include posts published on or before this date.
max_pages	integer	No	1	Maximum number of search-result pages to scrape, per keyword (Naver shows ~7 posts per page).
max_depth	integer	No	1	Reserved for future use — currently has no effect on scraping behavior.

Example input:

{
    "keywords": ["coffee", "seongsu"],
    "date_start": "2023-01-01",
    "date_end": "2023-06-30",
    "max_pages": 3
}

Output

Each dataset item is one blog post:

Field	Type	Description
keyword	string	The keyword that matched this post.
url	string	URL of the blog post.
title	string	Post title.
author	string	Author's display name.
pub_date	date	Publish date.
short_description	string	The snippet shown on the Naver search-results page.
full_text	string	All visible text extracted from the post page (see note below).
num_of_likes	number	Number of likes ("공감") on the post.
num_of_comments	number	Number of comments ("댓글") on the post.

Example output item:

{
    "keyword": "coffee",
    "url": "https://blog.naver.com/example_user/223456789012",
    "title": "The Best Coffee Shops in Seongsu",
    "author": "example_user",
    "pub_date": "2023-04-12T00:00:00.000Z",
    "short_description": "A quick look at three cozy coffee shops in Seoul's Seongsu neighborhood...",
    "full_text": "The Best Coffee Shops in Seongsu example_user · 2023. 4. 12. 15:02 URL copy ... (full page text) ...",
    "num_of_likes": 128,
    "num_of_comments": 14
}

Note: Naver Blog posts are rendered inside an iframe, and full_text captures everything visible inside it — the post body, but also headers, menus, and any ads embedded in the post. Some post-processing/cleanup is usually needed to isolate just the article text.

How to run it

On Apify Console

1. Open the Actor's page and click Try for free / Start.
2. Fill in your keywords and date range (see Input above).
3. Click Start and watch results appear in the Dataset tab as the run progresses.
4. Export results as JSON, CSV, Excel, etc. from the Dataset tab, or pull them via the API.

Via Apify API

curl "https://api.apify.com/v2/acts/YOUR_USERNAME~naver-blog-scraper/run-sync-get-dataset-items?token=YOUR_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
        "keywords": ["coffee"],
        "date_start": "2023-01-01",
        "date_end": "2023-06-30",
        "max_pages": 1
      }'

Via Apify CLI (local development)

$apify run

This requires the Selenium Chromedriver installed locally. When run on the Apify platform, the driver is already bundled in the Actor's Docker image, so no setup is needed.

Pricing

This Actor uses Apify's pay-per-event pricing model — you only pay for what you actually get:

• $0.10 flat fee per run (covers launching the browser).
• $0.001 per blog result returned in the dataset — i.e. $1 per 1,000 results.

If you set a maximum spend limit on a run, the Actor stops scraping further pages/blogs as soon as that limit is reached instead of continuing to do work you won't be charged — and won't be delivered — for.

Limitations

• Naver's page structure can change over time, which may require updates to the scraping logic.
• full_text includes non-article text (menus, ads) since it's extracted from an iframe — see the note under Output.
• Results depend on what Naver's search index returns for a given keyword and date range; it is not an exhaustive crawl of all posts ever published.