Google Search Scraping: Code Samples (PHP, Node.js, Rust)

Scraping Google Search results powers rank tracking, SERP monitoring, keyword research, and competitive intelligence. The catch: Google is one of the hardest targets on the web, with aggressive bot detection that blocks raw requests almost immediately. The reliable way to do this in production is through a SERP API that handles unblocking and returns structured results.

The samples below use Bright Data's SERP API, which accepts a normal Google search URL with brd_json=1 and returns parsed JSON (organic results, ranks, links, and snippets) instead of raw HTML.

Prerequisites

Set a PROXY_URL environment variable pointing at your SERP API zone:

export PROXY_URL="http://brd-customer-<id>-zone-<serp_zone>:<password>@brd.superproxy.io:33335"

No account yet? Bright Data offers a 7-day free trial, no credit card required. Get started →

PHP

<?php
// Run: php google.php "web scraping tools"
// Requires the bundled cURL + JSON extensions.

$proxy = getenv('PROXY_URL');
$query = $argv[1] ?? 'web scraping tools';

$url = 'https://www.google.com/search?' . http_build_query([
    'q'        => $query,
    'gl'       => 'us',   // country
    'hl'       => 'en',   // language
    'brd_json' => 1,      // ask the SERP API for parsed JSON instead of HTML
]);

$ch = curl_init($url);
curl_setopt_array($ch, [
    CURLOPT_RETURNTRANSFER => true,
    CURLOPT_PROXY          => $proxy,
    CURLOPT_SSL_VERIFYPEER => false, // or install Bright Data's CA certificate
    CURLOPT_TIMEOUT        => 60,
]);

$response = curl_exec($ch);
if ($response === false) {
    fwrite(STDERR, 'Request failed: ' . curl_error($ch) . PHP_EOL);
    exit(1);
}
curl_close($ch);

$data = json_decode($response, true);
foreach ($data['organic'] ?? [] as $r) {
    printf("%d. %s\n   %s\n   %s\n\n",
        $r['rank']        ?? 0,
        $r['title']       ?? '',
        $r['link']        ?? '',
        $r['description'] ?? ''
    );
}

Node.js

// google.mjs — run: node google.mjs "web scraping tools"
// Install: npm i axios https-proxy-agent
import axios from 'axios';
import { HttpsProxyAgent } from 'https-proxy-agent';

const agent = new HttpsProxyAgent(process.env.PROXY_URL);
const query = process.argv[2] ?? 'web scraping tools';

const url = 'https://www.google.com/search?' + new URLSearchParams({
  q: query, gl: 'us', hl: 'en', brd_json: '1',
});

const { data } = await axios.get(url, {
  httpsAgent: agent,
  proxy: false,        // route everything through the agent
  timeout: 60_000,
});

for (const r of data.organic ?? []) {
  console.log(`${r.rank}. ${r.title}\n   ${r.link}\n   ${r.description ?? ''}\n`);
}

Rust

// Cargo.toml:
//   reqwest = { version = "0.12", features = ["blocking", "json"] }
//   serde_json = "1"
use serde_json::Value;

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let query = std::env::args().nth(1).unwrap_or_else(|| "web scraping tools".into());

    let client = reqwest::blocking::Client::builder()
        .proxy(reqwest::Proxy::all(std::env::var("PROXY_URL")?)?)
        .danger_accept_invalid_certs(true) // or install Bright Data's CA
        .build()?;

    let data: Value = client
        .get("https://www.google.com/search")
        .query(&[("q", query.as_str()), ("gl", "us"), ("hl", "en"), ("brd_json", "1")])
        .send()?
        .json()?;

    if let Some(results) = data["organic"].as_array() {
        for r in results {
            println!(
                "{}. {}\n   {}\n   {}\n",
                r["rank"],
                r["title"].as_str().unwrap_or(""),
                r["link"].as_str().unwrap_or(""),
                r["description"].as_str().unwrap_or("")
            );
        }
    }
    Ok(())
}

Notes

• The SERP API returns far more than organic results — data also includes ads, related_searches, people_also_ask, and pagination. Inspect the full JSON to see every field.
• For rank tracking, run this on a schedule and store each result's rank per keyword over time.
• Want raw HTML instead of JSON? Drop the brd_json parameter and parse the markup yourself (selectors change often, which is why JSON is recommended).

Struggling with blocks on your own setup? Read How to Avoid Getting Blocked While Web Scraping and our Bright Data SERP API review.

Start your free 7-day Bright Data trial →