Trend detection involves identifying topics that are gaining rapid search and engagement momentum before they reach mainstream awareness. Combining Google search result velocity with YouTube video upload rates and view counts gives a multi-signal view of emerging trends. This tutorial builds a trend detection agent that queries both platforms via the Scavio API, scores topics by signal strength, and outputs a ranked list of trending subjects for a given niche.
Prerequisites
- Python 3.10 or higher
- requests library installed
- A Scavio API key
- Basic understanding of data normalization and scoring
Walkthrough
Step 1: Define topic seeds
Start with a set of broad seed terms in your niche. The agent will expand these into related trend queries.
Python
SEEDS = ["generative ai", "open source llm", "ai agents 2026"]
NICHE = "artificial intelligence"Step 2: Collect Google search result counts
For each seed, fetch Google SERP and note the number of organic results and their recency signals.
Python
def google_signal(topic: str) -> int:
data = search_google(f"{topic} 2026")
results = data.get("organic_results", [])
# Count results mentioning the current year as recency signal
recent = sum(1 for r in results if "2026" in r.get("title", "") + r.get("snippet", ""))
return recentStep 3: Collect YouTube video metrics
Search YouTube for each seed topic using the Scavio search endpoint and count recent video uploads.
Python
def youtube_signal(topic: str) -> int:
response = requests.post(
"https://api.scavio.dev/api/v1/search",
headers={"x-api-key": API_KEY},
json={"platform": "youtube", "query": topic}
)
videos = response.json().get("videos", [])
return len([v for v in videos if "2026" in v.get("published_at", "")])Step 4: Score and rank trends
Combine Google and YouTube signals into a composite score and rank topics.
Python
def rank_trends(seeds: list[str]) -> list[dict]:
scored = []
for seed in seeds:
g = google_signal(seed)
y = youtube_signal(seed)
scored.append({"topic": seed, "google": g, "youtube": y, "score": g + y * 2})
return sorted(scored, key=lambda x: x["score"], reverse=True)Python Example
Python
import os
import requests
API_KEY = os.environ.get("SCAVIO_API_KEY", "your_scavio_api_key")
ENDPOINT = "https://api.scavio.dev/api/v1/search"
SEEDS = ["generative ai", "open source llm", "ai agents", "model distillation"]
def search_google(q: str) -> dict:
r = requests.post(ENDPOINT, headers={"x-api-key": API_KEY},
json={"query": q, "country_code": "us"})
r.raise_for_status()
return r.json()
def search_youtube(q: str) -> list:
r = requests.post(ENDPOINT, headers={"x-api-key": API_KEY},
json={"platform": "youtube", "query": q})
r.raise_for_status()
return r.json().get("videos", [])
def score(seed: str) -> dict:
g_results = search_google(f"{seed} 2026").get("organic_results", [])
y_results = search_youtube(seed)
g_score = sum(1 for r in g_results if "2026" in r.get("title", ""))
y_score = len(y_results)
return {"topic": seed, "score": g_score + y_score * 2}
if __name__ == "__main__":
scores = sorted([score(s) for s in SEEDS], key=lambda x: x["score"], reverse=True)
for s in scores:
print(f"{s['topic']}: {s['score']}")JavaScript Example
JavaScript
const API_KEY = process.env.SCAVIO_API_KEY || "your_scavio_api_key";
const ENDPOINT = "https://api.scavio.dev/api/v1/search";
const SEEDS = ["generative ai", "open source llm", "ai agents", "model distillation"];
async function searchGoogle(q) {
const res = await fetch(ENDPOINT, {
method: "POST",
headers: { "x-api-key": API_KEY, "Content-Type": "application/json" },
body: JSON.stringify({ query: q, country_code: "us" })
});
return res.json();
}
async function scoreTopic(seed) {
const data = await searchGoogle(`${seed} 2026`);
const organic = data.organic_results || [];
const gScore = organic.filter(r => (r.title + " " + (r.snippet || "")).includes("2026")).length;
return { topic: seed, score: gScore };
}
async function main() {
const scores = await Promise.all(SEEDS.map(scoreTopic));
scores.sort((a, b) => b.score - a.score).forEach(s => console.log(`${s.topic}: ${s.score}`));
}
main().catch(console.error);Expected Output
JSON
ai agents: 8
generative ai: 7
open source llm: 5
model distillation: 3
Top trend: ai agents (score: 8)