A sales prospecting pipeline automates the process of finding potential customers, qualifying them based on signals like company size and tech stack, and enriching their profiles for outreach. Traditional prospecting tools like Apollo ($49+/user/month) or ZoomInfo provide pre-built databases, but they are expensive and often outdated. Using a search API, you can build a real-time prospecting pipeline that finds fresh leads by searching for buying signals. This tutorial builds the full pipeline at $0.005 per search via Scavio.
Prerequisites
- Python 3.9+ installed
- requests library installed
- A Scavio API key from scavio.dev
- A target customer profile (industry, size, tech stack)
Walkthrough
Step 1: Define your ideal customer profile searches
Create search queries that target companies matching your ICP. Use buying signals like job postings, technology mentions, and funding announcements.
Python
ICP_SEARCHES = [
# Companies hiring for roles that signal they need your product
'"hiring" AND "head of marketing" AND "series A" 2026',
# Companies using competitor products (potential switchers)
'"switching from HubSpot" OR "HubSpot alternative" 2026',
# Companies in target verticals with growth signals
'SaaS startup raised funding 2026 marketing automation',
# Job boards as lead source
'site:linkedin.com/jobs marketing director saas startup',
]
for i, q in enumerate(ICP_SEARCHES, 1):
print(f'{i}. {q}')Step 2: Search and extract prospect companies
Run each ICP search and extract company names and URLs from the results. Deduplicate across searches.
Python
import requests, os, re
API_KEY = os.environ['SCAVIO_API_KEY']
def search(query: str) -> list:
resp = requests.post('https://api.scavio.dev/api/v1/search',
headers={'x-api-key': API_KEY, 'Content-Type': 'application/json'},
json={'query': query, 'country_code': 'us'})
return resp.json().get('organic_results', [])
def extract_prospects(queries: list) -> list:
seen_domains = set()
prospects = []
for query in queries:
results = search(query)
for r in results:
from urllib.parse import urlparse
domain = urlparse(r.get('link', '')).netloc.replace('www.', '')
if domain and domain not in seen_domains:
seen_domains.add(domain)
prospects.append({
'domain': domain,
'title': r['title'],
'snippet': r.get('snippet', ''),
'source_query': query[:50]
})
return prospects
prospects = extract_prospects(ICP_SEARCHES)
print(f'Found {len(prospects)} unique prospects')Step 3: Qualify prospects with scoring
Score each prospect based on signals found in their search snippets. Higher scores indicate better fit with your ICP.
Python
SCORING_RULES = {
'series a': 10, 'series b': 15, 'raised': 10,
'hiring': 8, 'growing': 5, 'saas': 5,
'marketing': 5, 'automation': 5, 'startup': 3,
'2026': 3, 'remote': 2
}
def score_prospect(prospect: dict) -> int:
text = f'{prospect["title"]} {prospect["snippet"]}'.lower()
score = sum(points for keyword, points in SCORING_RULES.items() if keyword in text)
return score
def qualify_prospects(prospects: list, min_score: int = 10) -> list:
scored = []
for p in prospects:
p['score'] = score_prospect(p)
scored.append(p)
qualified = sorted([p for p in scored if p['score'] >= min_score],
key=lambda x: x['score'], reverse=True)
print(f'Qualified: {len(qualified)}/{len(prospects)} (score >= {min_score})')
return qualified
qualified = qualify_prospects(prospects)
for p in qualified[:5]:
print(f' [{p["score"]}] {p["domain"]}: {p["title"][:50]}')Step 4: Enrich qualified prospects
For qualified prospects only, run additional searches to find contact pages, team pages, and social profiles. This targeted enrichment keeps costs low.
Python
import time
def enrich_prospect(prospect: dict) -> dict:
domain = prospect['domain']
# Find contact/team page
contact_results = search(f'site:{domain} contact OR team OR about')
contact_urls = [r['link'] for r in contact_results[:3]]
# Find social profiles
social_results = search(f'{domain} linkedin OR twitter company')
social_links = [r['link'] for r in social_results
if 'linkedin.com' in r.get('link', '') or 'twitter.com' in r.get('link', '')]
prospect['contact_pages'] = contact_urls
prospect['social_profiles'] = social_links[:3]
return prospect
def enrich_batch(prospects: list) -> list:
enriched = []
for i, p in enumerate(prospects):
enriched.append(enrich_prospect(p))
if (i + 1) % 5 == 0:
print(f' Enriched {i + 1}/{len(prospects)}')
time.sleep(0.3)
return enriched
enriched = enrich_batch(qualified[:20]) # top 20 only
print(f'Enriched {len(enriched)} prospects')Step 5: Export the pipeline output
Save the qualified, enriched prospects to CSV sorted by score. Include cost tracking so you know the ROI of each prospecting run.
Python
import csv
def export_prospects(prospects: list, output: str = 'prospects.csv') -> None:
if not prospects:
return
fields = ['score', 'domain', 'title', 'snippet', 'contact_pages', 'social_profiles']
with open(output, 'w', newline='') as f:
writer = csv.DictWriter(f, fieldnames=fields, extrasaction='ignore')
writer.writeheader()
for p in prospects:
row = {**p}
row['contact_pages'] = ' | '.join(p.get('contact_pages', []))
row['social_profiles'] = ' | '.join(p.get('social_profiles', []))
writer.writerow(row)
# Cost calculation:
# ICP searches: 4 queries = $0.02
# Enrichment: 20 prospects x 2 searches = $0.20
total_cost = (len(ICP_SEARCHES) + len(prospects) * 2) * 0.005
print(f'Saved {len(prospects)} prospects to {output}')
print(f'Pipeline cost: ${total_cost:.2f}')
print(f'Cost per qualified lead: ${total_cost / max(len(prospects), 1):.3f}')
export_prospects(enriched)Python Example
Python
import os, requests, time
API_KEY = os.environ['SCAVIO_API_KEY']
def search(query):
resp = requests.post('https://api.scavio.dev/api/v1/search',
headers={'x-api-key': API_KEY, 'Content-Type': 'application/json'},
json={'query': query, 'country_code': 'us'})
return resp.json().get('organic_results', [])
def prospect(queries):
prospects = []
for q in queries:
for r in search(q):
prospects.append({'title': r['title'], 'url': r['link'], 'snippet': r.get('snippet', '')})
time.sleep(0.3)
return prospects
queries = ['SaaS startup hiring marketing 2026', 'series A marketing automation']
results = prospect(queries)
print(f'Found {len(results)} prospects at ${len(queries) * 0.005:.3f}')
for r in results[:5]:
print(f' {r["title"][:60]}')JavaScript Example
JavaScript
const API_KEY = process.env.SCAVIO_API_KEY;
async function search(query) {
const resp = await fetch('https://api.scavio.dev/api/v1/search', {
method: 'POST',
headers: { 'x-api-key': API_KEY, 'Content-Type': 'application/json' },
body: JSON.stringify({ query, country_code: 'us' })
});
return (await resp.json()).organic_results || [];
}
async function main() {
const queries = ['SaaS startup hiring marketing 2026', 'series A marketing automation'];
const prospects = [];
for (const q of queries) {
const results = await search(q);
results.forEach(r => prospects.push({ title: r.title, url: r.link }));
}
console.log(`Found ${prospects.length} prospects`);
prospects.slice(0, 5).forEach(p => console.log(` ${p.title.slice(0, 60)}`));
}
main().catch(console.error);Expected Output
JSON
Found 38 unique prospects
Qualified: 15/38 (score >= 10)
[25] acmesaas.com: Acme SaaS Raises $12M Series A for Marketing...
[20] growthco.io: GrowthCo Hiring Head of Marketing, Series B...
[18] betastart.com: BetaStart Switches from HubSpot to...
Enriched 5/15
Enriched 10/15
Enriched 15/15
Saved 15 prospects to prospects.csv
Pipeline cost: $0.17
Cost per qualified lead: $0.011