SERP API Evaluation Workflow

import requests
import time
import json

# Test against Scavio - repeat pattern for other vendors
API_KEY = "your_scavio_api_key"
TEST_QUERIES = [
    {"query": "best CRM software 2026", "platform": "google"},
    {"query": "wireless earbuds under 100", "platform": "amazon"},
    {"query": "python tutorial for beginners", "platform": "youtube"},
    {"query": "paper towels bulk", "platform": "walmart"},
    {"query": "best project management tool recommendation", "platform": "reddit"},
]

def evaluate_vendor(name: str, api_key: str) -> dict:
    results = []
    for test in TEST_QUERIES:
        start = time.time()
        try:
            res = requests.post(
                "https://api.scavio.dev/api/v1/search",
                headers={"x-api-key": api_key},
                json={"platform": test["platform"], "query": test["query"]},
                timeout=15,
            )
            latency_ms = round((time.time() - start) * 1000)
            if res.status_code == 200:
                data = res.json()
                result_count = len(data.get("organic", []))
                results.append({
                    "query": test["query"],
                    "platform": test["platform"],
                    "status": "ok",
                    "latency_ms": latency_ms,
                    "result_count": result_count,
                    "has_metadata": bool(data.get("ai_overview") or data.get("related_searches")),
                })
            else:
                results.append({"query": test["query"], "platform": test["platform"], "status": f"error_{res.status_code}", "latency_ms": latency_ms})
        except Exception as e:
            results.append({"query": test["query"], "platform": test["platform"], "status": f"exception: {str(e)}"})

    ok_results = [r for r in results if r["status"] == "ok"]
    avg_latency = sum(r["latency_ms"] for r in ok_results) // max(len(ok_results), 1)
    avg_results = sum(r["result_count"] for r in ok_results) // max(len(ok_results), 1)

    return {
        "vendor": name,
        "queries_tested": len(TEST_QUERIES),
        "success_rate": f"{len(ok_results)}/{len(TEST_QUERIES)}",
        "avg_latency_ms": avg_latency,
        "avg_result_count": avg_results,
        "platforms_tested": list(set(t["platform"] for t in TEST_QUERIES)),
        "details": results,
    }

# Evaluate Scavio
# Verified pricing: $0.005/credit, 1 credit per request
# Plans: Free 250/mo, Project $30/7k, Bootstrap $100/28k, Startup $250/85k, Growth $500/200k
evaluation = evaluate_vendor("scavio", API_KEY)
print(f"Vendor: {evaluation['vendor']}")
print(f"Success: {evaluation['success_rate']}, Avg latency: {evaluation['avg_latency_ms']}ms")
print(f"Avg results: {evaluation['avg_result_count']}")
print(f"Platforms: {', '.join(evaluation['platforms_tested'])}")

const API_KEY = "your_scavio_api_key";

const TEST_QUERIES = [
  { query: "best CRM software 2026", platform: "google" },
  { query: "wireless earbuds under 100", platform: "amazon" },
  { query: "python tutorial for beginners", platform: "youtube" },
  { query: "paper towels bulk", platform: "walmart" },
  { query: "best project management tool recommendation", platform: "reddit" },
];

async function evaluateVendor(name, apiKey) {
  const results = [];
  for (const test of TEST_QUERIES) {
    const start = Date.now();
    try {
      const res = await fetch("https://api.scavio.dev/api/v1/search", {
        method: "POST",
        headers: { "x-api-key": apiKey, "content-type": "application/json" },
        body: JSON.stringify({ platform: test.platform, query: test.query }),
      });
      const latencyMs = Date.now() - start;
      if (res.ok) {
        const data = await res.json();
        results.push({ query: test.query, platform: test.platform, status: "ok", latencyMs, resultCount: (data.organic ?? []).length });
      } else {
        results.push({ query: test.query, platform: test.platform, status: `error_${res.status}`, latencyMs });
      }
    } catch (err) {
      results.push({ query: test.query, platform: test.platform, status: "exception" });
    }
  }

  const ok = results.filter((r) => r.status === "ok");
  const avgLatency = Math.round(ok.reduce((s, r) => s + r.latencyMs, 0) / Math.max(ok.length, 1));
  const avgResults = Math.round(ok.reduce((s, r) => s + r.resultCount, 0) / Math.max(ok.length, 1));
  console.log(`${name}: ${ok.length}/${results.length} success, ${avgLatency}ms avg, ${avgResults} avg results`);
  return { vendor: name, successRate: `${ok.length}/${results.length}`, avgLatency, avgResults, details: results };
}

await evaluateVendor("scavio", API_KEY);