How long does this run a bulk serp ranking study with a search api tutorial take?

Most developers complete this tutorial in 15 to 30 minutes. You will need a Scavio API key (free tier works) and a working Python or JavaScript environment.

What do I need before starting?

A Scavio API key (50 free credits on signup; each Google call with light_request:true costs 1 credit). Python 3.9+ or Node 18+ with the requests/fetch ability to POST JSON. A list of 50 to a few thousand keywords in your niche (a CSV or plain text file). Basic comfort reading a correlation coefficient (a number between -1 and 1). A Scavio API key gives you 50 free credits on signup.

Can I run this tutorial with the free tier?

Yes. The free tier includes 50 credits on signup, which is more than enough to complete this tutorial and prototype a working solution.

What frameworks does this work with?

Scavio has a native LangChain package (langchain-scavio), an MCP server, and a plain REST API that works with any HTTP client. This tutorial uses the raw REST API, but you can adapt to your framework of choice.

Bulk SERP Ranking Study with a Search API

To run a bulk SERP ranking study, collect a list of keywords, fetch the organic results for each one with POST /api/v1/google, pull the feature you care about (say URL character length) at every ranking position, then aggregate by position bucket and compute a correlation. That's it. A German SEO once tested whether short URLs rank better across 27,000 SERPs and the effect was tiny. You can settle questions like that with real data instead of a hot take. This tutorial walks the whole loop: keyword list, paced API calls, feature extraction, aggregation, and a Pearson correlation. Correlation isn't causation, and SERPs have confounders, so read the number as a hint, not proof.

Prerequisites

A Scavio API key (50 free credits on signup; each Google call with light_request:true costs 1 credit)
Python 3.9+ or Node 18+ with the requests/fetch ability to POST JSON
A list of 50 to a few thousand keywords in your niche (a CSV or plain text file)
Basic comfort reading a correlation coefficient (a number between -1 and 1)

Walkthrough

Step 1: Build your keyword list

A study is only as honest as its sample. Pull real keywords from your niche rather than cherry-picking. Load them from a file so the run is reproducible. Aim for at least a few hundred if you want a stable correlation.

Python

# keywords.txt has one query per line
with open('keywords.txt') as f:
    keywords = [line.strip() for line in f if line.strip()]
print(f'Loaded {len(keywords)} keywords')

Step 2: Fetch one SERP and inspect the shape

Before looping over thousands, call the endpoint once and look at the response. Use light_request:true to keep each call at 1 credit. You get organic results, each with position, title, link, and snippet.

Python

import requests

API_KEY = 'YOUR_API_KEY'
resp = requests.post(
    'https://api.scavio.dev/api/v1/google',
    headers={'Authorization': f'Bearer {API_KEY}'},
    json={'query': 'best running shoes', 'light_request': True},
)
data = resp.json()
for r in data['organic_results'][:3]:
    print(r['position'], r['link'])

Step 3: Extract the feature per position

Here the feature is URL character length, but swap in title length, domain depth, or HTTPS presence as your hypothesis demands. Record one (position, value) pair per result so you can correlate later.

Python

def url_length(result):
    return len(result['link'])

def extract_pairs(serp):
    return [(r['position'], url_length(r)) for r in serp['organic_results']]

Step 4: Loop over keywords and respect the rate limit

Free keys allow 1 request per second; the $30 plan allows 2. Sleep between calls so you don't get throttled. Wrap each call in try/except so one bad query doesn't kill the whole run.

Python

import time

all_pairs = []
for kw in keywords:
    try:
        resp = requests.post(
            'https://api.scavio.dev/api/v1/google',
            headers={'Authorization': f'Bearer {API_KEY}'},
            json={'query': kw, 'light_request': True},
        )
        all_pairs += extract_pairs(resp.json())
    except Exception as e:
        print(f'skip {kw}: {e}')
    time.sleep(1.0)  # 1 req/sec on the free plan

Step 5: Aggregate by position bucket

Group results into buckets (1-3, 4-6, 7-10) and average the feature in each. This shows the trend at a glance before you trust any single number.

Python

from collections import defaultdict

buckets = defaultdict(list)
for pos, val in all_pairs:
    if pos <= 3: buckets['1-3'].append(val)
    elif pos <= 6: buckets['4-6'].append(val)
    else: buckets['7-10'].append(val)

for name in ['1-3', '4-6', '7-10']:
    vals = buckets[name]
    print(name, round(sum(vals)/len(vals), 1))

Step 6: Compute the correlation

A Pearson coefficient between position and the feature tells you the direction and strength. Near 0 means no relationship; the German URL-length study landed close to 0. Remember: confounders abound and correlation isn't causation.

Python

import statistics

def pearson(xs, ys):
    mx, my = statistics.mean(xs), statistics.mean(ys)
    num = sum((x-mx)*(y-my) for x, y in zip(xs, ys))
    den = (sum((x-mx)**2 for x in xs) * sum((y-my)**2 for y in ys)) ** 0.5
    return num/den if den else 0.0

positions = [p for p, _ in all_pairs]
values = [v for _, v in all_pairs]
print('correlation:', round(pearson(positions, values), 3))

Python Example

Python

import time
import statistics
from collections import defaultdict

import requests

API_KEY = 'YOUR_API_KEY'
ENDPOINT = 'https://api.scavio.dev/api/v1/google'


def fetch_serp(query):
    resp = requests.post(
        ENDPOINT,
        headers={'Authorization': f'Bearer {API_KEY}'},
        json={'query': query, 'light_request': True},
    )
    resp.raise_for_status()
    return resp.json()


def url_length(result):
    return len(result['link'])


def pearson(xs, ys):
    mx, my = statistics.mean(xs), statistics.mean(ys)
    num = sum((x - mx) * (y - my) for x, y in zip(xs, ys))
    den = (sum((x - mx) ** 2 for x in xs) * sum((y - my) ** 2 for y in ys)) ** 0.5
    return num / den if den else 0.0


def main():
    with open('keywords.txt') as f:
        keywords = [line.strip() for line in f if line.strip()]
    print(f'Loaded {len(keywords)} keywords')

    pairs = []
    for kw in keywords:
        try:
            serp = fetch_serp(kw)
            pairs += [(r['position'], url_length(r)) for r in serp['organic_results']]
        except Exception as e:
            print(f'skip {kw}: {e}')
        time.sleep(1.0)  # 1 req/sec on the free plan; 0.5 on the $30 plan

    buckets = defaultdict(list)
    for pos, val in pairs:
        if pos <= 3:
            buckets['1-3'].append(val)
        elif pos <= 6:
            buckets['4-6'].append(val)
        else:
            buckets['7-10'].append(val)

    print('avg URL length by position bucket:')
    for name in ['1-3', '4-6', '7-10']:
        vals = buckets[name]
        if vals:
            print(f'  {name}: {round(sum(vals) / len(vals), 1)}')

    positions = [p for p, _ in pairs]
    values = [v for _, v in pairs]
    print('correlation (position vs url length):', round(pearson(positions, values), 3))


if __name__ == '__main__':
    main()

JavaScript Example

JavaScript

import fs from 'node:fs';

const API_KEY = 'YOUR_API_KEY';
const ENDPOINT = 'https://api.scavio.dev/api/v1/google';
const sleep = (ms) => new Promise((r) => setTimeout(r, ms));

async function fetchSerp(query) {
  const resp = await fetch(ENDPOINT, {
    method: 'POST',
    headers: { Authorization: `Bearer ${API_KEY}`, 'Content-Type': 'application/json' },
    body: JSON.stringify({ query, light_request: true }),
  });
  return resp.json();
}

function pearson(xs, ys) {
  const mx = xs.reduce((a, b) => a + b, 0) / xs.length;
  const my = ys.reduce((a, b) => a + b, 0) / ys.length;
  let num = 0, dx = 0, dy = 0;
  for (let i = 0; i < xs.length; i++) {
    num += (xs[i] - mx) * (ys[i] - my);
    dx += (xs[i] - mx) ** 2;
    dy += (ys[i] - my) ** 2;
  }
  const den = Math.sqrt(dx * dy);
  return den ? num / den : 0;
}

const keywords = fs.readFileSync('keywords.txt', 'utf8').split('\n').map((s) => s.trim()).filter(Boolean);
const pairs = [];
for (const kw of keywords) {
  try {
    const serp = await fetchSerp(kw);
    for (const r of serp.organic_results) pairs.push([r.position, r.link.length]);
  } catch (e) {
    console.log(`skip ${kw}: ${e}`);
  }
  await sleep(1000); // 1 req/sec on the free plan
}

const positions = pairs.map((p) => p[0]);
const values = pairs.map((p) => p[1]);
console.log('correlation:', pearson(positions, values).toFixed(3));

Expected Output

JSON

Loaded 500 keywords
avg URL length by position bucket:
  1-3: 48.2
  4-6: 51.7
  7-10: 53.9
correlation (position vs url length): 0.072

Prerequisites

A Scavio API key (50 free credits on signup; each Google call with light_request:true costs 1 credit)
Python 3.9+ or Node 18+ with the requests/fetch ability to POST JSON
A list of 50 to a few thousand keywords in your niche (a CSV or plain text file)
Basic comfort reading a correlation coefficient (a number between -1 and 1)

Walkthrough

Step 1: Build your keyword list

Python

# keywords.txt has one query per line
with open('keywords.txt') as f:
    keywords = [line.strip() for line in f if line.strip()]
print(f'Loaded {len(keywords)} keywords')

Step 2: Fetch one SERP and inspect the shape

Python

import requests

API_KEY = 'YOUR_API_KEY'
resp = requests.post(
    'https://api.scavio.dev/api/v1/google',
    headers={'Authorization': f'Bearer {API_KEY}'},
    json={'query': 'best running shoes', 'light_request': True},
)
data = resp.json()
for r in data['organic_results'][:3]:
    print(r['position'], r['link'])

Step 3: Extract the feature per position

Python

def url_length(result):
    return len(result['link'])

def extract_pairs(serp):
    return [(r['position'], url_length(r)) for r in serp['organic_results']]

Step 4: Loop over keywords and respect the rate limit

Free keys allow 1 request per second; the $30 plan allows 2. Sleep between calls so you don't get throttled. Wrap each call in try/except so one bad query doesn't kill the whole run.

Python

import time

all_pairs = []
for kw in keywords:
    try:
        resp = requests.post(
            'https://api.scavio.dev/api/v1/google',
            headers={'Authorization': f'Bearer {API_KEY}'},
            json={'query': kw, 'light_request': True},
        )
        all_pairs += extract_pairs(resp.json())
    except Exception as e:
        print(f'skip {kw}: {e}')
    time.sleep(1.0)  # 1 req/sec on the free plan

Step 5: Aggregate by position bucket

Group results into buckets (1-3, 4-6, 7-10) and average the feature in each. This shows the trend at a glance before you trust any single number.

Python

from collections import defaultdict

buckets = defaultdict(list)
for pos, val in all_pairs:
    if pos <= 3: buckets['1-3'].append(val)
    elif pos <= 6: buckets['4-6'].append(val)
    else: buckets['7-10'].append(val)

for name in ['1-3', '4-6', '7-10']:
    vals = buckets[name]
    print(name, round(sum(vals)/len(vals), 1))

Step 6: Compute the correlation

Python

import statistics

def pearson(xs, ys):
    mx, my = statistics.mean(xs), statistics.mean(ys)
    num = sum((x-mx)*(y-my) for x, y in zip(xs, ys))
    den = (sum((x-mx)**2 for x in xs) * sum((y-my)**2 for y in ys)) ** 0.5
    return num/den if den else 0.0

positions = [p for p, _ in all_pairs]
values = [v for _, v in all_pairs]
print('correlation:', round(pearson(positions, values), 3))

Python Example

Python

import time
import statistics
from collections import defaultdict

import requests

API_KEY = 'YOUR_API_KEY'
ENDPOINT = 'https://api.scavio.dev/api/v1/google'


def fetch_serp(query):
    resp = requests.post(
        ENDPOINT,
        headers={'Authorization': f'Bearer {API_KEY}'},
        json={'query': query, 'light_request': True},
    )
    resp.raise_for_status()
    return resp.json()


def url_length(result):
    return len(result['link'])


def pearson(xs, ys):
    mx, my = statistics.mean(xs), statistics.mean(ys)
    num = sum((x - mx) * (y - my) for x, y in zip(xs, ys))
    den = (sum((x - mx) ** 2 for x in xs) * sum((y - my) ** 2 for y in ys)) ** 0.5
    return num / den if den else 0.0


def main():
    with open('keywords.txt') as f:
        keywords = [line.strip() for line in f if line.strip()]
    print(f'Loaded {len(keywords)} keywords')

    pairs = []
    for kw in keywords:
        try:
            serp = fetch_serp(kw)
            pairs += [(r['position'], url_length(r)) for r in serp['organic_results']]
        except Exception as e:
            print(f'skip {kw}: {e}')
        time.sleep(1.0)  # 1 req/sec on the free plan; 0.5 on the $30 plan

    buckets = defaultdict(list)
    for pos, val in pairs:
        if pos <= 3:
            buckets['1-3'].append(val)
        elif pos <= 6:
            buckets['4-6'].append(val)
        else:
            buckets['7-10'].append(val)

    print('avg URL length by position bucket:')
    for name in ['1-3', '4-6', '7-10']:
        vals = buckets[name]
        if vals:
            print(f'  {name}: {round(sum(vals) / len(vals), 1)}')

    positions = [p for p, _ in pairs]
    values = [v for _, v in pairs]
    print('correlation (position vs url length):', round(pearson(positions, values), 3))


if __name__ == '__main__':
    main()

JavaScript Example

JavaScript

import fs from 'node:fs';

const API_KEY = 'YOUR_API_KEY';
const ENDPOINT = 'https://api.scavio.dev/api/v1/google';
const sleep = (ms) => new Promise((r) => setTimeout(r, ms));

async function fetchSerp(query) {
  const resp = await fetch(ENDPOINT, {
    method: 'POST',
    headers: { Authorization: `Bearer ${API_KEY}`, 'Content-Type': 'application/json' },
    body: JSON.stringify({ query, light_request: true }),
  });
  return resp.json();
}

function pearson(xs, ys) {
  const mx = xs.reduce((a, b) => a + b, 0) / xs.length;
  const my = ys.reduce((a, b) => a + b, 0) / ys.length;
  let num = 0, dx = 0, dy = 0;
  for (let i = 0; i < xs.length; i++) {
    num += (xs[i] - mx) * (ys[i] - my);
    dx += (xs[i] - mx) ** 2;
    dy += (ys[i] - my) ** 2;
  }
  const den = Math.sqrt(dx * dy);
  return den ? num / den : 0;
}

const keywords = fs.readFileSync('keywords.txt', 'utf8').split('\n').map((s) => s.trim()).filter(Boolean);
const pairs = [];
for (const kw of keywords) {
  try {
    const serp = await fetchSerp(kw);
    for (const r of serp.organic_results) pairs.push([r.position, r.link.length]);
  } catch (e) {
    console.log(`skip ${kw}: ${e}`);
  }
  await sleep(1000); // 1 req/sec on the free plan
}

const positions = pairs.map((p) => p[0]);
const values = pairs.map((p) => p[1]);
console.log('correlation:', pearson(positions, values).toFixed(3));

Expected Output

JSON

Loaded 500 keywords
avg URL length by position bucket:
  1-3: 48.2
  4-6: 51.7
  7-10: 53.9
correlation (position vs url length): 0.072

How to Run a Bulk SERP Ranking Study with a Search API

Prerequisites

Walkthrough

Step 1: Build your keyword list

Step 2: Fetch one SERP and inspect the shape

Step 3: Extract the feature per position

Step 4: Loop over keywords and respect the rate limit

Step 5: Aggregate by position bucket

Step 6: Compute the correlation

Python Example

JavaScript Example

Expected Output

Related Tutorials

Frequently Asked Questions

How long does this run a bulk serp ranking study with a search api tutorial take?

What do I need before starting?

Can I run this tutorial with the free tier?

What frameworks does this work with?

Related Resources

Best Google Maps Business Data APIs (May 2026)

Best Queue-Based SERP API in 2026

Google Ads Data from SERP APIs

SERP API Procurement Comparison

Google Places API vs SERP Local Pack API

Semrush API vs Raw SERP API

Start Building

How to Run a Bulk SERP Ranking Study with a Search API

Prerequisites

Walkthrough

Step 1: Build your keyword list

Step 2: Fetch one SERP and inspect the shape

Step 3: Extract the feature per position

Step 4: Loop over keywords and respect the rate limit

Step 5: Aggregate by position bucket

Step 6: Compute the correlation

Python Example

JavaScript Example

Expected Output

Related Tutorials

Frequently Asked Questions

How long does this run a bulk serp ranking study with a search api tutorial take?

What do I need before starting?

Can I run this tutorial with the free tier?

What frameworks does this work with?

Related Resources

Best Google Maps Business Data APIs (May 2026)

Best Queue-Based SERP API in 2026

Google Ads Data from SERP APIs

SERP API Procurement Comparison

Google Places API vs SERP Local Pack API

Semrush API vs Raw SERP API

Start Building