Depending on the study, Naver's share of search in Korea ranges from 28% to 65%. None of these estimates is wrong: each one measures something different, so the useful question for a launch plan is not who is number one but where your own customers look for you.
Same market, different numbers
Here are the studies published in 2026, side by side.
| Source | What it measured | Naver | |
|---|---|---|---|
| Chosun Ilbo (IGAWorks Mobile Index) | App MAU, July 2026 | 46.84 million | 47.02 million |
| seonews (citing StatCounter) | Web traffic, July 2026 | 40.87% | 49.52% |
| Korea Data Economy News (citing StatCounter) | Web traffic, March 2026 | 43.96% | 46.81% |
| Inblog | 2026 compilation | 48.6% | 42.9% |
| A Korean blog (citing Internet Trend) | Korean panel, first half of 2026 | 64.28% | 28.37% |
We compiled this table from reports and materials published as of August 21, 2026. The original source of each figure is the research body named in the table; some values are cited through news outlets or blogs.
Why the numbers differ: three different rulers
1. App MAU: did someone open the app at least once a month?
The Chosun Ilbo report that the Google app passed Naver in Korean monthly active users for the first time uses this measure. Based on IGAWorks Mobile Index data, Google had 47.02 million users against Naver's 46.84 million.
But MAU counts anyone who opens the app just once in a month, and the Google app comes preinstalled on Android devices. Other figures in the same report make the difference in behavior clear:
- Daily active users (DAU). Naver 26 to 27 million vs Google 16 to 17 million.
- Daily time spent per user. Naver 21 to 24 minutes vs Google 6 minutes.
- Total daily time spent. Naver 9 to 10 million hours vs Google 1.6 to 1.7 million hours.
On a monthly basis Google is ahead. On a daily basis Naver leads by three to four times. Both are true; they simply answer different questions.
2. Web traffic: which search engine sent the visit?
StatCounter-type tools count where visitors to websites carrying their tracking tag came from. The result therefore depends on what kind of sites have installed the tag. This is where the tendency for Google to show higher figures in Korea comes from.
3. Korean panel studies: where did a sample group actually search?
Korean studies such as Internet Trend look at the real search behavior of a sample. On this basis Naver stays in the 60% range. Even for the same month, the gap with web traffic figures can exceed 30 percentage points.
We apply the same principle to measuring AI visibility; see how we measure AI search and traffic.
What this means for a brand entering Korea
Whatever the national split, the path a customer takes to find you is shaped by your category. The national share debate matters far less than that.
- Local services customers find on the spot. Think restaurants, beauty, clinics and accommodation. Naver Place and Naver AI Briefing still carry a large share.
- Categories people compare and research before choosing. Think procedures, professional services and B2B. More buyers now gather information on Google and ChatGPT first.
The right move is to measure your own acquisition channels directly, not national averages. For the wider context, see our overview of marketing in Korea in 2026.
Why you cannot pick just one
Separate from the share debate, the two paths do not substitute for each other. This is something we have verified ourselves.
- Naver blocks global AI crawlers. It is stated in Naver's robots.txt, so content built up on Naver Blog cannot be read by ChatGPT or Perplexity. Details: how Naver blocks AI crawlers.
- Naver AI Briefing reads only Naver's index. However well your website is built, if it is not indexed by Naver it will not be cited in the briefing.
- Our own page split the results. It ranked first on Naver AI Briefing but did not appear on ChatGPT. Same content, same day, different outcome.
"Google won, so drop Naver" and "Naver is still stronger, so Google can wait" are both risky. Each path is currently closed to the other, so giving one up means losing that side's customers entirely.
What to do now
- Read the measurement basis, not the headline. Whether a figure is MAU, web traffic or a panel study can reverse the conclusion.
- Measure your own acquisition instead of national share. In GA4 traffic acquisition, check the share of visits coming from Naver, Google and AI.
- Manage the two paths separately. On Naver, Place and Blog are the assets; for global AI, your own website is the asset.
WizThePlanning is a Seoul-based marketing agency operating since 2016 that has worked with 7,000+ advertisers. If you are planning a launch, see our Korea market entry support.
Frequently asked questions
Q1.Has Google really overtaken Naver in Korea?
It depends on what is counted. On monthly active app users (MAU), Google moved ahead of Naver for the first time in July 2026. In the same data, Naver was far ahead on daily active users and time spent, and Korean panel studies of search share still put Naver in the 60% range. The situation does not reduce to one sentence.
Q2.Why do search share figures differ so much between studies?
They measure different things. App MAU counts anyone who opens the app once a month, and the Google app comes preinstalled on Android devices. StatCounter-type tools count where visitors to websites carrying their tag came from. Korean panel studies look at the actual search behavior of a sample group. These are answers to three different questions, so the numbers cannot match.
Q3.Which number should a brand planning a Korea launch look at?
Treat national share as background only. What matters is where customers in your category look for you. Local services still lean heavily on Naver Place and Naver AI Briefing, while categories that involve comparison and research lean more on Google and ChatGPT. Measure your own acquisition by source rather than relying on national averages.
Q4.Can we focus on only Naver or only Google?
Not right now. Naver blocks global AI crawlers in its robots.txt, so content on Naver Blog cannot be read by ChatGPT or Perplexity. In the other direction, Naver AI Briefing only draws on Naver's own index. One of our own pages ranked first on Naver AI Briefing but did not appear on ChatGPT. The two paths do not substitute for each other.
Korean version: 구글 네이버 검색 점유율
