El éxito de la optimización generativa del motor (GEO) depende de si las descripciones generales de IA citan su marca. Este escáner consulta las palabras clave objetivo, comprueba si aparecen descripciones generales de IA y registra si se cita su dominio. Ejecútelo diariamente para realizar un seguimiento del porcentaje de citas a lo largo del tiempo. Cada consulta cuesta $0.005.
Requisitos previos
- Python 3.8+
- solicita biblioteca
- Una clave API de Scavio de scavio.dev
- Dominio de destino y lista de palabras clave
Guia paso a paso
Paso 1: Consultar palabras clave objetivo para descripciones generales de IA
Busque cada palabra clave y verifique si hay una descripción general de IA en el SERP.
Python
import os, requests, json
from datetime import datetime
API_KEY = os.environ['SCAVIO_API_KEY']
SH = {'x-api-key': API_KEY, 'Content-Type': 'application/json'}
DOMAIN = 'yourbrand.com'
KEYWORDS = [
'best project management tool 2026',
'project management software comparison',
'how to choose project management app',
'project management for small teams',
'agile project management tools',
]
def scan_keyword(query):
data = requests.post('https://api.scavio.dev/api/v1/search',
headers=SH, json={'query': query, 'country_code': 'us'}).json()
ai_overview = data.get('ai_overview', data.get('answer_box', {}))
has_aio = bool(ai_overview)
cited = False
if has_aio:
aio_text = json.dumps(ai_overview).lower()
cited = DOMAIN.lower() in aio_text
organic = data.get('organic_results', [])
org_position = None
for i, r in enumerate(organic):
if DOMAIN in r.get('link', ''):
org_position = i + 1
break
return {'query': query, 'has_aio': has_aio, 'cited': cited, 'organic_pos': org_position}
results = []
for kw in KEYWORDS:
r = scan_keyword(kw)
results.append(r)
status = 'CITED' if r['cited'] else ('AIO present' if r['has_aio'] else 'No AIO')
print(f' [{status:12}] {kw[:50]:50} | org: #{r["organic_pos"] or "-"}')
print(f'\nCost: ${len(KEYWORDS) * 0.005:.3f}')Paso 2: Calcular métricas de porcentaje de citas
Calcule métricas GEO: tasa de citas, frecuencia AIO y correlación orgánica.
Python
def geo_metrics(results):
total = len(results)
aio_count = sum(1 for r in results if r['has_aio'])
cited_count = sum(1 for r in results if r['cited'])
aio_rate = aio_count / total * 100 if total else 0
citation_rate = cited_count / aio_count * 100 if aio_count else 0
overall_rate = cited_count / total * 100 if total else 0
organic_when_cited = [r['organic_pos'] for r in results if r['cited'] and r['organic_pos']]
avg_org = sum(organic_when_cited) / len(organic_when_cited) if organic_when_cited else 0
print(f'\n=== GEO Citation Report - {DOMAIN} ===')
print(f' Keywords scanned: {total}')
print(f' AI Overviews present: {aio_count} ({aio_rate:.0f}%)')
print(f' Your brand cited: {cited_count} ({overall_rate:.0f}% of all, {citation_rate:.0f}% of AIO)')
if avg_org:
print(f' Avg organic position when cited: #{avg_org:.1f}')
print(f'\n Not cited but AIO present:')
for r in results:
if r['has_aio'] and not r['cited']:
print(f' - {r["query"][:50]}')
return {'aio_rate': aio_rate, 'citation_rate': citation_rate}
geo_metrics(results)Paso 3: Seguimiento de los cambios de citas a lo largo del tiempo
Almacene los resultados de los análisis diarios y compárelos con análisis anteriores.
Python
def track_citations(results, history_file='geo_citations.json'):
try:
with open(history_file) as f:
history = json.load(f)
except FileNotFoundError:
history = []
today = {
'date': datetime.now().strftime('%Y-%m-%d'),
'domain': DOMAIN,
'total': len(results),
'aio_present': sum(1 for r in results if r['has_aio']),
'cited': sum(1 for r in results if r['cited']),
'details': results
}
history.append(today)
with open(history_file, 'w') as f:
json.dump(history, f, indent=2)
print(f'\nSaved scan to {history_file}')
if len(history) >= 2:
prev = history[-2]
delta_aio = today['aio_present'] - prev['aio_present']
delta_cited = today['cited'] - prev['cited']
print(f' vs {prev["date"]}: AIO {delta_aio:+d}, Citations {delta_cited:+d}')
print(f' Run daily to build citation trend data')
track_citations(results)Ejemplo en Python
Python
import os, requests, json
SH = {'x-api-key': os.environ['SCAVIO_API_KEY'], 'Content-Type': 'application/json'}
def scan_geo(query, domain):
data = requests.post('https://api.scavio.dev/api/v1/search',
headers=SH, json={'query': query, 'country_code': 'us'}).json()
aio = data.get('ai_overview', {})
cited = domain.lower() in json.dumps(aio).lower() if aio else False
print(f'{"CITED" if cited else "not cited":10} | {query[:50]}')
for q in ['best project management tool 2026', 'agile tools comparison']:
scan_geo(q, 'yourbrand.com')
print('Cost: $0.010')Ejemplo en JavaScript
JavaScript
const SH = { 'x-api-key': process.env.SCAVIO_API_KEY, 'Content-Type': 'application/json' };
async function scanGeo(query, domain) {
const data = await fetch('https://api.scavio.dev/api/v1/search', {
method: 'POST', headers: SH,
body: JSON.stringify({ query, country_code: 'us' })
}).then(r => r.json());
const aio = data.ai_overview || {};
const cited = JSON.stringify(aio).toLowerCase().includes(domain.toLowerCase());
console.log(`${cited ? 'CITED' : 'not cited'} | ${query}`);
}
await scanGeo('best project management tool 2026', 'yourbrand.com');Salida esperada
JSON
[CITED ] best project management tool 2026 | org: #3
[AIO present ] project management software comparison | org: #5
[CITED ] how to choose project management app | org: #2
[No AIO ] project management for small teams | org: #4
[AIO present ] agile project management tools | org: #7
Cost: $0.025
=== GEO Citation Report - yourbrand.com ===
Keywords scanned: 5
AI Overviews present: 4 (80%)
Your brand cited: 2 (40% of all, 50% of AIO)
Avg organic position when cited: #2.5