Naver AI Briefing Monitor — brands in Korean AI answers

Part of LeoWorks Korea market tools: shop rank → Naver Shopping Rank Tracker · blog posts and sponsorship → Naver Blog Brand Monitor · AI answers → Naver AI Briefing Monitor · beauty rankings → Olive Young & Amore Mall Rankings · reviews → Korean Review Classifier.

For SEO / GEO teams and brands selling in Korea who need to know what Naver's AI briefing (AI 브리핑) — the generated answer at the top of Korea's #1 search engine, like Google AI Overviews — says for their keywords: the answer, the sources it cites, the ads inside it, and whether their brand and competitors are mentioned or recommended. $3 per 1,000 queries checked, plus $12 per 1,000 briefings captured. It shows what 네이버 AI 브리핑 cites (AI 브리핑 인용) and which ads appear inside the answer.

Use it to: track brand mentions in Naver AI answers · see which blogs and sites Naver's AI cites for your category · monitor competitors' AI search visibility · find AI-briefing ads in your keywords · Korean generative engine optimization (GEO / AEO) reporting.

Output sample

Real rows (one per query) from runs uaTe9x2Kxh71xQeQh (brand 다이슨, competitors 삼성 / LG, 2026-10-06) and PZd7R1emu8cZZaIt2 (2026-10-04). English in parentheses is added here for readers.

query	hasBriefing	brandMentions (mentioned / recommended)	citations	media	adsInBriefing	answerText (first words)
무선청소기 추천 2026 (cordless vacuum recommendations 2026)	true	다이슨 true / true · 삼성 true / true · LG true / true	5	3	0	2026년 무선청소기는 …
콜라겐 효과 (collagen benefits)	true	다이슨 false · 삼성 false · LG false	4	3	0	콜라겐은 피부, 뼈, 근육, 힘줄, 인대 등 인체의 다양한 조직을 구성하는 중요한 단백질입니다…
레티놀 부작용 (retinol side effects)	false	—	0	0	0	— (charged only as a query check)

Every row also has quickAnswer, relatedQuestions[], the full citations[] (title, URL, platform), each brand's sentiment and citedSources, and checkedAt / checkedAtKst — see Output fields below.

Input example

The form default — two queries, no brands (about $0.03, 10 seconds):

{
  "queries": ["콜라겐 효과", "전기포트 세척 방법"]
}

Add your brand and competitors to get the mention / recommendation judgments; use brand names (e.g. 삼성, LG) rather than product-line names, because the AI answer usually names the brand:

{
  "queries": ["무선청소기 추천 2026", "콜라겐 효과"],
  "brandNames": ["다이슨"],
  "competitorNames": ["삼성", "LG"]
}

Schedule the Actor to watch how answers change over time.

Pricing

Pay only for results — no subscription. You pay the full price only when an AI briefing is actually shown.

Event	Price	When
query-checked	$0.003	One query checked on Naver search (whether an AI briefing appears).
briefing-captured	$0.012	An AI briefing was shown: answer, citations, related questions, ads and brand judgments captured (added on top of the query check).

That is $3 per 1,000 queries checked and $12 per 1,000 briefings captured on top ($15 per 1,000 queries if every query had a briefing; in our tests about 40% do). First run with the form defaults: about $0.03 (2 queries, both with a briefing, 10 seconds).

Cost examples

Run	Cost
100 queries, no AI briefing shown	$0.30
100 queries, all with an AI briefing	$1.50
100 queries, 40% with an AI briefing (typical in our test)	$0.78
20 queries checked daily for 30 days, 40% with briefing	$4.68

With the free $5 monthly Apify credit you can check about 641 queries (at a 40% briefing rate).

Queries that fail to load are reported with an error field and not charged. If you set a maximum cost per run, the Actor stops cleanly when it is reached.

Works with

• Apify API, JavaScript and Python clients — start a run and read the dataset like any Actor (leoworks/naver-ai-briefing-monitor).
• Apify Schedules — a daily schedule shows how answers, citations and brand mentions change; our own monitoring runs this Actor every morning.
• Claude, Cursor and Claude Code through the Apify MCP server — see the next section (verified 2026-10-06).
• Naver Blog Brand Monitor — when the briefing cites Naver blogs, leoworks/naver-blog-brand-monitor tells you which of those posts are sponsored.

Use with Claude, Cursor or Claude Code (MCP)

Ask your AI agent. Add the Apify MCP server with this Actor as a tool, then say, for example:

• "Check whether 다이슨 is mentioned in Naver's AI briefing for 무선청소기 추천 2026 and list the cited sources."
• "Check Naver's AI briefing for 콜라겐 효과 and 전기포트 세척 방법 and tell me which blogs and sites it cites."
• "For the query 에어프라이어 추천, does Naver's AI briefing mention 쿠쿠 or 필립스, and is either of them recommended?"

The agent calls the tool leoworks--naver-ai-briefing-monitor and reads the rows with get-dataset-items.

Claude Desktop or Cursor (mcp.json):

{
  "mcpServers": {
    "apify": {
      "url": "https://mcp.apify.com?tools=leoworks/naver-ai-briefing-monitor",
      "headers": { "Authorization": "Bearer YOUR_APIFY_TOKEN" }
    }
  }
}

Claude Code: claude mcp add --transport http apify "https://mcp.apify.com?tools=leoworks/naver-ai-briefing-monitor" --header "Authorization: Bearer YOUR_APIFY_TOKEN". Leave out the header to sign in with OAuth in the browser instead. Your Apify token is in Console → Settings → API & Integrations. We verified this setup with the Apify MCP server (v0.17.2) on 2026-10-06.

Output fields

Field	Meaning
hasBriefing	Whether Naver showed an AI briefing for the query
answerText	The AI answer (clean text, footnotes removed)
quickAnswer	The short highlighted answer, when shown
citations[]	Cited sources: title, url, source, platform (Naver Blog, Naver Clip, Naver Influencer, external sites…), publishedAt
brandMentions[]	For each brand/competitor: mentioned, recommended, sentiment, probabilities, and citedSources (how many cited sources name it)
adsInBriefing[]	Ads Naver inserted into the AI briefing: advertiser, text, headline
relatedQuestions[]	Follow-up questions, when shown
media[]	Images and video cards shown inside the briefing: type (image / video), title, source (blog, clinic, news site, Naver Clip channel…), url, thumbnailUrl
checkedAt, checkedAtKst	Check time (UTC and Korea time)

Query tips (observed 2026-10-02, sample of 20 queries)

AI briefings appeared for 8 of 20 test queries. They showed up mostly for informational, how-to, review and comparison queries ("콜라겐 효과", "전기포트 세척 방법", "갤럭시 버즈3 후기", "헬리녹스 체어원 vs 체어제로") and rarely for bare "OO 추천" or brand-name-only queries. Queries without a briefing cost only the basic check fee.

Accuracy

• AI briefing presence, answer and cited sources matched what a browser shows for 20 test queries.
• Brand mention / recommendation judgments: 97% agreement with hand labels (38 judgments).

Limits

Item	Limit
Brands + competitors	Up to 10 per run
Surface	PC search (default) or mobile search; in our tests mobile showed briefings slightly less often
Personalization	Checks run logged out; answers can change over time and between sessions
Ads	Reported when Naver attaches them to the briefing; impression tracking URLs are never called

Run it on a schedule

Daily is the usual rhythm: AI answers, citations and brand mentions change from day to day, and a daily run shows how. Create an Apify Schedule for a saved task with your keywords and brands. To start it from other tools, use the Apify n8n or Make integration, or the Apify API.

FAQ

Why not use the official Naver API?

Naver's search API has no endpoint for the AI briefing (AI 브리핑) — the generated answer only exists on the search result page. This Actor reads that page without a Naver login and returns the answer, the sources it cites, the ads inside it and your brand and competitor mentions.

What are the alternatives to Naver AI Briefing Monitor?

Searching each keyword by hand and taking screenshots, or a GEO / AI-visibility tool — most of them track Google, ChatGPT and Perplexity, not Naver. This Actor gives one row per query for $3 per 1,000 queries checked plus $12 per 1,000 briefings captured.

How do I monitor my brand in Naver AI briefing?

Enter your keywords, your brand and your competitors, and schedule a daily run. Each row shows whether a briefing appeared and whether each brand is mentioned or recommended, plus the cited sources, so you can see how the answers change.

Which AI makes the judgments?

Jev, TypeSafe's decision model (version jev-1.13.0, pinned). Jev answers each label with a calibrated probability instead of generated text, so a 0.9 means the same thing for every item. Answers are stable but not fixed: re-running the same input can change a few borderline answers (under 1% of answers in our repeat tests) — filter on probability when an item must never flip. We re-measure accuracy on our labelled test set before we change the model version. Only the AI answer text (up to 6,000 characters), the names and titles of cited sources and your brand names are sent to Jev.

Does it need a Naver account?

No. Checks run without a login.

Disclaimer

Independent tool — not affiliated with, endorsed by or sponsored by NAVER Corp. "Naver" is a trademark of its owner and is used only to describe the data source. AI briefings are generated by Naver and may contain inaccuracies.

Reviews and support

If this Actor saved you time, a short review on Apify Store helps others find it. Questions or a query that does not work? Open an issue in the Issues tab — we answer within a day.

Changelog

See the Changelog tab.