Trend Signals

What it is

Korean retail crypto activity is concentrated on a handful of venues and, equally, on a handful of information channels — most prominently the Naver search engine, Korea's dominant search platform. A spike in Naver search volume for a token name precedes (or co-moves with) Korean exchange trading volume and KRW-denominated price action by minutes to hours. The Trend Signals strategy uses search-trend data as a leading or co-incident indicator for short-horizon directional bets, listing-arb prep, or volume-filter overlays on other strategies.

Similar logic applies to Telegram chatter (DataMaxi+ also surfaces Telegram trend data), but the Naver search dataset is the cleanest signal for the Korean retail audience specifically.

This is not arbitrage — it's a signal, and signals decay. Trend strategies are most useful as:

A filter on top of an arb strategy (only trade kimchi-premium opportunities when Naver search is rising).
A mean-reversion fade when search volume is parabolic (retail tops).
A listing-arb pre-positioner — a spiking search query without a corresponding listing announcement can foreshadow leaks or community speculation.

When it works

During Korean retail-driven regimes (alt seasons, KRW-pair-led rallies).
For mid-cap tokens with clear name recognition — search is noisy for tickers that overlap common words.
On 24h/7d/30d trend deltas — single-hour search spikes are usually too noisy to act on at retail latency.
When combined with price/volume confirmation. Search alone has too many false positives (news, scams, unrelated mentions) to be a standalone entry.

It does not work as a standalone signal for majors (BTC/ETH — search is saturated) or for tokens with non-distinct names.

Data you need

Naver search trend — /api/v1/naver-trend — 24h, 7d, 30d trend scores per asset. Private endpoint.
Naver supported symbols — /api/v1/naver/symbols — which tokens are tracked.
Telegram trend — /api/v1/telegram/channels, /api/v1/telegram/messages — channel/post engagement metrics as a secondary sentiment input.
CEX ticker — /api/v1/ticker — price confirmation overlay.
CEX candle — /api/v1/cex/candle — historical OHLC for backtesting signal-to-price correlation.
Per-exchange 24h volume — /api/v1/cex/symbol/per-exchange-24-h-volume — confirm Korean-exchange volume is moving alongside the search delta.

API recipe

Pull current Naver search-trend deltas, sorted by 24h change:

cURL
Python
Go
TypeScript

curl -G 'https://api.datamaxiplus.com/api/v1/naver-trend' \
  -H 'X-DTMX-APIKEY: '"$YOUR_API_KEY" \
  --data-urlencode 'sort=desc' \
  --data-urlencode 'key=change24h' \
  --data-urlencode 'limit=20'

import os
import requests

resp = requests.get(
    "https://api.datamaxiplus.com/api/v1/naver-trend",
    headers={"X-DTMX-APIKEY": os.environ["DTMX_API_KEY"]},
    params={"sort": "desc", "key": "change24h", "limit": 20},
    timeout=10,
)
for row in resp.json().get("data", []):
    print(f"{row['symbol']:<10}  24h={row.get('change24h'):>+8}  "
          f"7d={row.get('change7d'):>+8}  30d={row.get('change30d'):>+8}")

package main

import (
	"encoding/json"
	"fmt"
	"net/http"
	"net/url"
	"os"
)

func main() {
	q := url.Values{
		"sort":  {"desc"},
		"key":   {"change24h"},
		"limit": {"20"},
	}
	req, _ := http.NewRequest("GET",
		"https://api.datamaxiplus.com/api/v1/naver-trend?"+q.Encode(), nil)
	req.Header.Set("X-DTMX-APIKEY", os.Getenv("DTMX_API_KEY"))
	resp, _ := http.DefaultClient.Do(req)
	defer resp.Body.Close()

	var out struct{ Data []map[string]any `json:"data"` }
	json.NewDecoder(resp.Body).Decode(&out)
	for _, r := range out.Data { fmt.Println(r) }
}

import axios from "axios";

const { data } = await axios.get(
  "https://api.datamaxiplus.com/api/v1/naver-trend",
  {
    headers: { "X-DTMX-APIKEY": process.env.DTMX_API_KEY! },
    params: { sort: "desc", key: "change24h", limit: 20 },
  },
);
console.log(data.data);

For a full strategy loop you'd join this against the CEX ticker / candle endpoints and gate entries on both change24h > threshold and price/volume confirmation.

Risks & caveats

Signal decay. Once a trend is in the public dashboard, it's already in the price. Backtest realistic latency between signal publication and your fill.
Name-collision noise. Tickers that overlap with common Korean words or unrelated brands get false-positive spikes. Whitelist tokens with distinct names.
Survivorship bias in backtests. Tokens that "worked" historically were often the ones that survived; current data may not generalize. Walk-forward test.
Regime change. Search-as-leading-indicator was strong in 2017–2021 retail cycles. In institutional-led regimes the signal weakens — track its rolling correlation, don't assume it's stationary.
Confounders. Search spikes around bad news (hack, delisting) look identical to spikes around good news in the raw data. Pair with announcement/news feeds.
Privacy/ToS. Naver search trend data is provided in aggregate; no individual-user data is exposed. Still, treat redistribution restrictions as per the DataMaxi+ ToS.

What it is​

When it works​

Data you need​

API recipe​

Risks & caveats​

Further reading​

What it is

When it works

Data you need

API recipe

Risks & caveats

Further reading