Full Merlino Audit SOP Crown Jewel

This is the master SOP for running a complete Merlino-style SEO audit. No stone unturned. Every endpoint listed. Every data point pulled. Agents reference this page when executing an audit.

Audit cost estimate: $12–$22 per client domain (depends on backlink index size and crawl depth)

Execution time: 8–15 minutes (with parallel execution)

Output format: Static TypeScript files at clients/{slug}/ consumed by the audit dashboard

---

Pre-Audit Checklist

• [ ] Confirm DATAFORSEO_LOGIN and DATAFORSEO_PASSWORD env vars are set
• [ ] Confirm GOOGLE_PLACES_API_KEY is set (for GMB verification)
• [ ] Collect: target domain, 4–6 target keywords, all GMB business names and CIDs
• [ ] Confirm client location (city, state, zip) for local SERP accuracy
• [ ] Set location_code: 2840 (US) or override to specific metro code
• [ ] Confirm output directory: clients/{slug}/

---

Execution Groups

Run all calls within a group in parallel. Wait for each group to complete before starting the next.

Group 1 (parallel): Backlinks Summary + Labs Domain Overview + Domain Technologies + WHOIS + Content Analysis Summary
Group 2 (parallel): All SERP calls (4 target keywords × organic + maps + local finder + news + images + AI overview)
Group 3 (parallel): Keywords Data (volume + keywords for site + keywords for keywords + trends)
Group 4 (parallel): Labs (ranked keywords + competitors + historical rank overview + bulk traffic + keyword difficulty + search intent)
Group 5 (parallel): Backlinks Detail (backlinks list + anchors + referring domains + domain pages + timeseries)
Group 6 (parallel): Business Data (GMB info + reviews for each location)
Group 7 (sequential): On-Page crawl task post → poll for completion → retrieve pages + errors + links + duplicates
Group 8 (parallel): AI Optimization (LLM mentions brand + service keywords, ChatGPT/Claude/Gemini/Perplexity responses)

---

DataForSEO Pulls

Backlinks API — 12 Endpoints

All backlinks endpoints are Live-only. Run as early as possible — largest dataset.

#	Endpoint	Payload Key Fields	Cost Est.	Group
1	POST /v3/backlinks/summary/live	target, include_subdomains: true	~$0.01	1
2	POST /v3/backlinks/history/live	target, date_from, date_to	~$0.01	5
3	POST /v3/backlinks/backlinks/live	target, limit: 100, order_by: ["rank,desc"]	~$0.02–$0.05	5
4	POST /v3/backlinks/anchors/live	target, limit: 50	~$0.005	5
5	POST /v3/backlinks/referring_domains/live	target, limit: 100, order_by: ["rank,desc"]	~$0.01–$0.05	5
6	POST /v3/backlinks/domain_pages/live	target, limit: 50	~$0.01	5
7	POST /v3/backlinks/domain_pages_summary/live	target, limit: 50	~$0.01	1
8	POST /v3/backlinks/referring_networks/live	target, limit: 50	~$0.01	5
9	POST /v3/backlinks/competitors/live	target, limit: 10	~$0.01	5
10	POST /v3/backlinks/bulk_spam_score/live	targets: [competitor_domains]	~$0.002/domain	5
11	POST /v3/backlinks/bulk_ranks/live	targets: [domain, competitors...]	~$0.002/domain	1
12	POST /v3/backlinks/timeseries_referring_domains/live	target, date_from, date_to, group_by: "month"	~$0.01	5

Python — Backlinks Summary:

import os, requests
from requests.auth import HTTPBasicAuth

auth = HTTPBasicAuth(os.environ["DATAFORSEO_LOGIN"], os.environ["DATAFORSEO_PASSWORD"])
BASE = "https://api.dataforseo.com"

def backlinks_summary(target):
    r = requests.post(
        f"{BASE}/v3/backlinks/summary/live",
        auth=auth,
        headers={"Content-Type": "application/json"},
        json=[{"target": target, "include_subdomains": True}]
    )
    return r.json()["tasks"][0]["result"][0]

---

SERP API — All Tabs, 4+ Keywords

Run for each target keyword. Cover every SERP tab.

#	Endpoint	When to Use	Cost Est.	Group
1	POST /v3/serp/google/organic/live/advanced	All 4+ target keywords	~$0.005/kw	2
2	POST /v3/serp/google/maps/live/advanced	Primary local query (city + service)	~$0.003	2
3	POST /v3/serp/google/local_finder/live/advanced	Same local query	~$0.003	2
4	POST /v3/serp/google/news/live/advanced	Brand name query	~$0.003	2
5	POST /v3/serp/google/images/live/advanced	Brand name + primary service query	~$0.003	2
6	POST /v3/serp/google/ai_overview/live	Primary service keyword	~$0.005	2
7	POST /v3/serp/google/ai_mode/tasks_post	Primary service keyword (async)	~$0.01	2
8	GET /v3/serp/google/autocomplete/live/advanced	Brand name, primary service	~$0.001	2

Standard organic SERP payload:

def serp_organic(keyword, location_code=2840, language_code="en", depth=20):
    r = requests.post(
        f"{BASE}/v3/serp/google/organic/live/advanced",
        auth=auth,
        headers={"Content-Type": "application/json"},
        json=[{
            "keyword": keyword,
            "location_code": location_code,
            "language_code": language_code,
            "depth": depth,
            "calculate_rectangles": False
        }]
    )
    return r.json()["tasks"][0]["result"][0]

---

Keywords Data API — 5 Endpoints

#	Endpoint	Payload	Cost Est.	Group
1	POST /v3/keywords_data/google_ads/search_volume/live	All target keywords (batch)	~$0.0005/kw	3
2	POST /v3/keywords_data/google_ads/keywords_for_site/live	Client domain	~$0.01	3
3	POST /v3/keywords_data/google_ads/keywords_for_keywords/live	Primary service keywords as seeds	~$0.005	3
4	POST /v3/keywords_data/google_trends/explore/live	Primary 2–3 keywords	~$0.001/kw	3
5	POST /v3/keywords_data/clickstream/search_volume/live	All target keywords	~$0.001/kw	3

---

DataForSEO Labs API — 9 Endpoints

#	Endpoint	Payload	Cost Est.	Group
1	POST /v3/dataforseo_labs/google/ranked_keywords/live	Client domain, limit: 100	~$0.01	4
2	POST /v3/dataforseo_labs/google/competitors_domain/live	Client domain, limit: 10	~$0.01	4
3	POST /v3/dataforseo_labs/google/domain_rank_overview/live	Client domain	~$0.01	1
4	POST /v3/dataforseo_labs/google/historical_rank_overview/live	Client domain	~$0.01	4
5	POST /v3/dataforseo_labs/google/keyword_suggestions/live	Primary service keyword	~$0.005	4
6	POST /v3/dataforseo_labs/google/related_keywords/live	Primary service keyword	~$0.005	4
7	POST /v3/dataforseo_labs/google/keyword_difficulty/live	All target keywords (batch)	~$0.002/kw	4
8	POST /v3/dataforseo_labs/google/search_intent/live	All target keywords (batch)	~$0.001/kw	4
9	POST /v3/dataforseo_labs/google/bulk_traffic_estimation/live	Client domain + top 5 competitors	~$0.005	4

---

AI Optimization API — 8+ Endpoint Calls

Brand and service keyword visibility across all major LLMs.

#	What to Call	Endpoint	Cost Est.	Group
1	LLM Mentions — brand name	POST /v3/ai_optimization/llm_mentions/search/live	~$0.005	8
2	LLM Mentions — primary service keyword	POST /v3/ai_optimization/llm_mentions/search/live	~$0.005	8
3	LLM Mentions — secondary service keyword	POST /v3/ai_optimization/llm_mentions/search/live	~$0.005	8
4	ChatGPT response — "best [service] in [city]"	POST /v3/ai_optimization/chatgpt/live/tasks_post	~$0.01	8
5	Claude response — "best [service] in [city]"	POST /v3/ai_optimization/claude/live/tasks_post	~$0.01	8
6	Gemini response — "best [service] in [city]"	POST /v3/ai_optimization/gemini/live/tasks_post	~$0.01	8
7	Perplexity response — "best [service] in [city]"	POST /v3/ai_optimization/perplexity/live	~$0.01	8
8	AI Keyword Data — primary keywords	POST /v3/ai_optimization/ai_keyword_data/live	~$0.005	8

LLM Mentions payload:

def llm_mentions(keyword, location_code=2840):
    r = requests.post(
        f"{BASE}/v3/ai_optimization/llm_mentions/search/live",
        auth=auth,
        headers={"Content-Type": "application/json"},
        json=[{
            "keyword": keyword,
            "location_code": location_code,
            "language_code": "en"
        }]
    )
    return r.json()["tasks"][0]["result"][0]

---

On-Page API — Full Site Crawl

#	Step	Endpoint	Cost Est.	Group
1	Start crawl	POST /v3/on_page/task_post	~$0.004/page × pages	7a
2	Poll status	GET /v3/on_page/summary/{task_id}	Free	7b
3	Get pages	POST /v3/on_page/pages/{task_id}	Free (included)	7c
4	Get errors	POST /v3/on_page/errors/{task_id}	Free	7c
5	Get links	POST /v3/on_page/links/{task_id}	Free	7c
6	Duplicate tags	POST /v3/on_page/duplicate_tags/{task_id}	Free	7c
7	Duplicate content	POST /v3/on_page/duplicate_content/{task_id}	Free	7c
8	Resources	POST /v3/on_page/resources/{task_id}	Requires load_resources: true	7c
9	Instant page scan	POST /v3/on_page/instant_pages	~$0.004	7c
10	Lighthouse audit	POST /v3/on_page/lighthouse/live/json	~$0.02	7c

Crawl task payload:

def start_onpage_crawl(domain, max_pages=100):
    r = requests.post(
        f"{BASE}/v3/on_page/task_post",
        auth=auth,
        headers={"Content-Type": "application/json"},
        json=[{
            "target": domain,
            "max_crawl_pages": max_pages,
            "load_resources": False,
            "enable_javascript": False,
            "enable_browser_rendering": False,
            "store_raw_html": False,
            "check_spell": False
        }]
    )
    return r.json()["tasks"][0]["id"]

---

Business Data API — GMB + Reviews

Run for every client GMB location.

#	Endpoint	Payload	Cost Est.	Group
1	POST /v3/business_data/google/my_business_info/live	Business name + city	~$0.01/location	6
2	POST /v3/business_data/google/reviews/task_post	Business CID or name	~$0.01–$0.02	6
3	POST /v3/business_data/google/my_business_updates/task_post	Business CID	~$0.01	6

If client is in hospitality, add:

• POST /v3/business_data/tripadvisor/reviews/task_post
• POST /v3/business_data/trustpilot/reviews/task_post

---

Domain Analytics API — 2 Endpoints

#	Endpoint	Payload	Cost Est.	Group
1	POST /v3/domain_analytics/technologies/domain_technologies/live	Client domain + top 3 competitors	~$0.005/domain	1
2	POST /v3/domain_analytics/whois/overview/live	Client domain	~$0.005	1

---

Content Analysis API — 3 Endpoints

#	Endpoint	Payload	Cost Est.	Group
1	POST /v3/content_analysis/search/live	Brand name, limit: 50	~$0.005	1
2	POST /v3/content_analysis/summary/live	Brand name	~$0.002	1
3	POST /v3/content_analysis/sentiment_analysis/live	Brand name	~$0.002	1

---

Non-DataForSEO Pulls

These run in parallel with Group 1 or Group 2.

#	Source	What to Pull	How
1	Direct HTTP	Homepage HTML — verify schema markup, meta tags, GHL/LeadConnector detection	curl -s https://domain.com
2	Google Places API	Verified GBP data — CID, categories, hours, photos count	GET https://maps.googleapis.com/maps/api/place/details/json
3	ReddAPI	Reddit brand mentions and sentiment	GET https://api.reddapi.com/...
4	DomDetailer API	DA, TF, CF, Trust Flow, Spam Score	POST https://www.domdetailer.com/api/...
5	Queen browser agent	Social media follower counts (Facebook, Instagram, YouTube)	Browser visit — NEVER use scraped estimates
6	Queen browser agent	YouTube channel presence — subscribers, videos, last post date	Browser visit
7	Queen browser agent	LSA (Local Services Ads) presence check — "Google Guaranteed" badge	Browser visit + screenshot
8	Queen browser agent	Google Ads presence check — paid listings in SERP screenshots	Browser visit
9	Direct HTTP	robots.txt, sitemap.xml — parse and validate	curl -s https://domain.com/robots.txt

---

Output Files (Static TypeScript)

Write these files to clients/{slug}/:

File	Contents
audit.ts	Top-level audit summary, score, action items, technologies
backlinks-data.ts	Backlinks summary, referring domains, anchors, history, timeseries
serp-data.ts	Organic rankings, map pack, local finder, AI overview results
keywords.ts	Search volume, difficulty, intent, trends for all target keywords
onpage-data.ts	Crawl summary, page issues, Lighthouse scores, errors
gmb-data.ts	GMB info, reviews, ratings, Q&A for each location
llm-data.ts	LLM mentions, AI engine responses, AI keyword data
competitors-data.ts	Competitor domains, traffic estimates, keyword overlap
geogrid-data.ts	Geogrid rank positions (if geogrid service used)

All exports must be typed TypeScript — no any, no untyped objects. Dashboard components import directly.

---

Cost Estimate by Category

Category	Endpoints	Cost Range
Backlinks (12 endpoints)	summary, history, backlinks, anchors, domains, pages, networks, competitors, bulk spam, bulk ranks, timeseries	$0.15–$0.50
SERP (8 calls × 4 keywords)	organic, maps, local finder, news, images, AI overview, AI mode, autocomplete	$0.20–$0.40
Keywords Data (5 endpoints)	volume, for site, for keywords, trends, clickstream	$0.05–$0.10
Labs (9 endpoints)	ranked keywords, competitors, rank overview, historical, suggestions, related, difficulty, intent, traffic	$0.10–$0.15
On-Page (crawl 50 pages)	task post, summary, pages, errors, links, dupes, Lighthouse	$0.20–$0.40
Business Data (2 locations)	GMB info, reviews per location	$0.04–$0.08
Domain Analytics	technologies, WHOIS	$0.01–$0.02
Content Analysis	search, summary, sentiment	$0.01
AI Optimization (8 calls)	LLM mentions × 3, LLM responses × 4, AI keyword data	$0.07–$0.10
Total	~48 DataForSEO calls	$0.83–$1.75

Plus non-DataForSEO: Google Places (~$0.01), DomDetailer (~$0.01), Queen browser time (~$0).

Total per full audit: approximately $1–$2 in API costs.

---

QA Checklist — 3 Rounds

Round 1: Data Completeness

• [ ] All 48+ DataForSEO calls returned status_code: 20000
• [ ] No None or empty result arrays in critical fields
• [ ] All TypeScript files compile without errors (tsc --noEmit)
• [ ] All export names match what dashboard components import
• [ ] backlinks-data.ts has real integer values for backlinks, referring_domains, rank
• [ ] keywords.ts has real integer search_volume (not strings, not 0 for all)
• [ ] serp-data.ts has at least 1 organic result for each keyword
• [ ] gmb-data.ts has correct business name, address, rating, review count
• [ ] llm-data.ts has llm_mentions entries with se_type field

Round 2: Data Accuracy

• [ ] Domain rank (DataForSEO) matches DomDetailer DA within 20%
• [ ] Review count in GMB data matches visible Google count (Queen verifies)
• [ ] Top-ranking keywords in ranked_keywords match known real rankings
• [ ] Social follower counts are browser-verified — NOT from any API estimate
• [ ] Schema markup detected on homepage matches what's actually in the HTML

Round 3: Dashboard Render

• [ ] npm run build passes 30/30 pages with 0 errors
• [ ] All 7 dashboard tabs render without blank panels
• [ ] Recharts / chart components have data (not empty arrays)
• [ ] No console errors in browser (Queen screenshot)
• [ ] All numbers visible in UI match source TypeScript data

---

Parallel Execution Script (Python Skeleton)

import os
import asyncio
import aiohttp
from aiohttp import BasicAuth

LOGIN = os.environ["DATAFORSEO_LOGIN"]
PASSWORD = os.environ["DATAFORSEO_PASSWORD"]
BASE = "https://api.dataforseo.com"

auth = BasicAuth(LOGIN, PASSWORD)

async def post(session, path, payload):
    async with session.post(f"{BASE}{path}", json=payload) as r:
        return await r.json()

async def run_group(session, calls):
    """Run a list of (path, payload) tuples in parallel."""
    tasks = [post(session, path, payload) for path, payload in calls]
    return await asyncio.gather(*tasks)

async def full_audit(domain, keywords, location_code=2840):
    async with aiohttp.ClientSession(auth=auth, headers={"Content-Type": "application/json"}) as session:

        # Group 1 — Domain-level data
        group1 = [
            ("/v3/backlinks/summary/live", [{"target": domain}]),
            ("/v3/backlinks/domain_pages_summary/live", [{"target": domain, "limit": 50}]),
            ("/v3/backlinks/bulk_ranks/live", [{"targets": [domain]}]),
            ("/v3/dataforseo_labs/google/domain_rank_overview/live", [{"target": domain, "location_code": location_code, "language_code": "en"}]),
            ("/v3/domain_analytics/technologies/domain_technologies/live", [{"target": domain}]),
            ("/v3/domain_analytics/whois/overview/live", [{"target": domain}]),
            ("/v3/content_analysis/summary/live", [{"keyword": domain.replace(".com", "")}]),
            ("/v3/content_analysis/sentiment_analysis/live", [{"keyword": domain.replace(".com", "")}]),
        ]
        results_g1 = await run_group(session, group1)

        # Group 2 — SERP for all keywords
        group2 = [
            ("/v3/serp/google/organic/live/advanced", [{"keyword": kw, "location_code": location_code, "language_code": "en", "depth": 20}])
            for kw in keywords
        ] + [
            ("/v3/serp/google/maps/live/advanced", [{"keyword": keywords[0] + " near me", "location_code": location_code, "language_code": "en"}]),
            ("/v3/serp/google/local_finder/live/advanced", [{"keyword": keywords[0], "location_code": location_code, "language_code": "en"}]),
            ("/v3/serp/google/ai_overview/live", [{"keyword": keywords[0], "location_code": location_code, "language_code": "en"}]),
        ]
        results_g2 = await run_group(session, group2)

        # ... continue for groups 3-8
        return {"group1": results_g1, "group2": results_g2}

if __name__ == "__main__":
    domain = "example.com"
    keywords = ["roofing contractor dallas", "roof replacement dallas", "emergency roof repair", "metal roof installation"]
    asyncio.run(full_audit(domain, keywords))

---

Key Rules

1. Never use search snippets or scraped estimates for data — DataForSEO API is the mandatory primary source for all SEO metrics.
2. Social follower counts must be browser-verified by Queen — no API ever returns accurate follower counts.
3. All TypeScript output must be static — no API calls from the dashboard. Data is pre-fetched and committed to the repo.
4. Mark clearly when data is unavailable — if an endpoint returns no result, set the field to null and add a comment. Never fabricate values.
5. Retry each failed call once — if still failing after retry, log the error and continue. Don't let one bad endpoint block the entire audit.
6. Cost logging is mandatory — log data.cost from every response. Alert if any single call exceeds $0.50.