Authority & Citation Building Agent

5 min read

🔗 Authority & Citation Building Agent

BaC Principle

“A backlink is a vote from the web. An AI citation is a vote from the model. Both require the same thing: content worth referencing. Automate the identification; systematize the outreach.” — Business as Code Manifesto

Stage: S8 — Authority & Citation Building Type: AUTO (~70% automated — outreach execution requires human or semi-automated email) Trigger: After S7 produces approved_publish_batch (content live on site) Output feeds: 10-Metrics-and-Self-Improvement

See Map of Content for full vault navigation.

🎯 Objective

Build domain authority through targeted link acquisition and establish AI citation presence by ensuring the site’s content is discoverable and structured in ways that LLMs cite. This agent operates on two parallel tracks: traditional backlink building and AEO citation building.

🛠️ Tools & Integrations

tools:
  - name: Ahrefs / SEMrush API
    used_for: >
      Competitor backlink analysis, broken link building prospecting,
      unlinked brand mention discovery, link gap analysis
    api: REST API
 
  - name: Hunter.io / Apollo.io
    used_for: Contact enrichment for outreach targets (editor/webmaster emails)
    api: REST API
 
  - name: Claude API
    used_for: >
      Personalized outreach email drafts per prospect,
      AI citation probe queries (simulate LLM search),
      content repurposing suggestions for citation building
    api: Messages API
 
  - name: HARO / Connectively
    used_for: Journalist query monitoring — respond to relevant media requests
    api: Email digest monitoring (automated filtering by topic)
 
  - name: Reddit / Quora / LinkedIn
    used_for: Community citation seeding — answer questions with site content
    api: Manual + SerpAPI monitoring for relevant threads
 
  - name: OpenAI API / Perplexity API
    used_for: >
      Weekly AI citation probes — query AI platforms with target questions
      to measure if site content is being cited
    api: REST API

🔀 Dual-Track Architecture

flowchart TD
    A["Receive approved_publish_batch<br/>Content live on site"]

    subgraph BACKLINKS["🔗 Track 1: Backlink Building"]
        B1["Competitor backlink gap analysis"]
        B2["Broken link building prospecting"]
        B3["Unlinked brand mention outreach"]
        B4["HARO response drafting"]
        B5["Outreach queue: prioritized prospects"]
    end

    subgraph CITATIONS["🤖 Track 2: AI Citation Building"]
        C1["Probe AI platforms with target queries"]
        C2["Identify citation gaps"]
        C3["Content submission to AI training sources"]
        C4["Community seeding (Reddit, Quora)"]
        C5["Citation log update"]
    end

    A --> B1 & C1
    B1 --> B2 --> B3 --> B4 --> B5
    C1 --> C2 --> C3 --> C4 --> C5

⚡ Execution Steps

  1. Competitor backlink gap — Pull top 50 backlink sources for each SERP competitor. Identify domains linking to competitors but not to the site. Score by: Domain Rating (≥40), topic relevance, link type (editorial vs. directory).

  2. Broken link building — Find broken external links on relevant domains that used to point to similar content. Propose the site’s equivalent page as a replacement.

  3. Unlinked brand mentions — Search for mentions of “Business as Code” or “businessascode.co” without a hyperlink. Draft outreach requesting a link addition.

  4. HARO monitoring — Filter incoming HARO digests for queries matching target topics (automation, AI, business operations). Draft expert response with embedded attribution.

  5. Outreach queue — Compile prioritized prospect list with personalized email draft per contact.

Track 2: AEO Citation Building

  1. AI platform probes — Send 10 target queries per week to ChatGPT, Perplexity, Claude, Gemini. Record: was the site cited? What was cited instead? What is the answer gap?

  2. Citation gap analysis — Identify queries where competitors are cited but the site is not. Cross-reference with content inventory — is there a page covering this topic?

  3. Content submission — Submit key pages to: Google Search index (via GSC), Bing Webmaster Tools, Common Crawl ping endpoints. Ensure all pages have full schema markup (feeds AI training data pipelines).

  4. Community seeding — Identify Reddit threads and Quora questions where target queries are asked. Post substantive answers with links to relevant site content (non-spammy, value-first).

  5. Citation log — Update ai_citation_tracker with weekly probe results.

📤 Output Schema

{
  "authority_run_id": "auth_2026-04-02",
  "generated_at": "2026-04-02T23:00:00Z",
  "backlink_track": {
    "prospects_identified": 34,
    "prospects_qualified": 18,
    "outreach_queue": [
      {
        "prospect_id": "p001",
        "domain": "example-blog.com",
        "domain_rating": 52,
        "contact_email": "editor@example-blog.com",
        "link_opportunity_type": "broken_link_replacement",
        "target_page": "/what-is-business-as-code",
        "email_draft": "Hi [Name], I noticed your article on automation tools links to a page that's no longer live. We have an updated resource on Business as Code that covers the same topic in depth — would you consider linking to it instead? [link]",
        "priority": "high"
      }
    ],
    "haro_responses_drafted": 3
  },
  "citation_track": {
    "queries_probed": 30,
    "site_cited": 4,
    "citation_rate": 0.13,
    "top_citation_gaps": [
      "what is business as code definition",
      "how to automate SOPs with AI agents"
    ],
    "competitor_citations": {
      "zapier.com": 12,
      "processst.com": 7
    },
    "community_posts_queued": 5,
    "ai_citation_tracker_updated": true
  }
}

📊 KPIs

MetricTargetAlert Condition
Backlink prospects qualified/month≥ 15< 5 → expand competitor list
Outreach emails sent/month≥ 100 → escalate to human
New referring domains/month≥ 50 for 2 months → review strategy
AI citation rate (queries probed)≥ 20%< 10% → prioritize AEO pass on gap topics
Citation gap queries addressed100% of top 10Any unaddressed → add to content strategy
HARO responses submitted/month≥ 30 → check HARO topic filters

⚠️ Error Handling

error_rules:
  - condition: Ahrefs API returns empty backlink data for competitor
    action: Try SEMrush as fallback; use Moz as tertiary
    fallback: Manual backlink check via Ahrefs web UI; log result manually
 
  - condition: AI citation probe returns no usable response (API error)
    action: Retry after 1 hour; use alternative phrasing of query
    fallback: Mark probe as "failed" in citation log; retry next week
 
  - condition: HARO response rejected by journalist
    action: Log as rejected; do not contact same journalist again for 90 days
    fallback: None — move to next opportunity
 
  - condition: Community post flagged/removed (Reddit, Quora)
    action: Review post for spam signals; adjust content-to-link ratio
    fallback: Pause community seeding for 7 days; review guidelines

📎 Related Files

Graph View

Backlinks