B2B AI DirectoryB2B AI Directory
Competitive InteladvancedPro

Build a LinkedIn ad intelligence agent that runs on a schedule

Use Claude Code, G2, TrustRadius, LinkedIn Ad Library, GitHub, Railway, and Vercel to monitor competitor ads and generate a recurring intelligence report.

What you will have

A scheduled competitive ad intelligence agent that discovers competitors, pulls LinkedIn ad data, analyzes themes and volume changes, and publishes a recurring report.

Setup time
6-10 hours for first deployed agent
Time saved
4-8 hours per competitor intelligence report
Estimated cost
$20 to $300 per month
Tools used
8 tools

Why this works

Ad libraries expose how competitors actually go to market, not just how they describe themselves on their homepage. The problem is that manual ad-library research is repetitive, scroll-heavy, and hard to compare over time. This workflow turns the research motion into a scheduled agent with versioned code, hosting, retry logic, and a report format that can be reused across companies or categories.

Step-by-step workflow

Preview the workflow

The first 2 steps are open. Pro unlocks the remaining steps, copy-paste prompts, pro tips, tool-by-tool setup guidance, and implementation details.

1

Define the report audience, decisions, and acceptance criteria

60-75 min

Decide who will use the recurring report and what decisions it should support before any code is written. Define the target category or company, competitor scope, run frequency, report owner, review deadline, historical comparison window, and approved distribution audience. Open Claude Code in Plan mode and paste the prompt below into the main terminal or chat input. Include the target URL, intended audience, required report sections, available hosting accounts, technical constraints, and examples of a useful competitive insight as source inputs. Ask Claude Code to propose the system boundary, data model, folder structure, execution sequence, failure modes, security model, and acceptance test. Copy the approved plan into `docs/ad-intel-plan.md` and save the Claude Code session reference in the project notes. Do not allow implementation to start until Marketing and the technical reviewer agree on what counts as a complete, trustworthy run.

Output

An approved report and system specification with audience, decisions, scope, architecture, failure modes, security, and acceptance criteria.

Claude Code
Pro tip

Specify decisions, not just charts. A report that cannot change a campaign, positioning, budget, or research priority becomes an automated newsletter nobody needs.

Prompt template
Open Claude Code in Plan mode and create the implementation plan for a scheduled LinkedIn ad intelligence agent. Do not write or modify implementation code yet.

Source inputs:
- Target company or category URL: {{target_url}}
- Report audience and their decisions: {{report_audience_and_decisions}}
- Competitor scope and maximum count: {{competitor_scope}}
- Required report sections: {{required_sections}}
- Run frequency and historical comparison window: {{schedule_and_history}}
- Available accounts and platforms: {{available_platforms}}
- Technical constraints and preferred stack: {{technical_constraints}}
- Security, privacy, and data-retention requirements: {{security_requirements}}
- Examples of useful and useless insights: {{quality_examples}}
- Budget and operating-owner constraints: {{operating_constraints}}

Plan a system that uses the existing workflow tools: Claude Code, G2, TrustRadius, LinkedIn Ads Library, GitHub, Railway, Vercel, and Google Sheets.

Return:
1. System objective and non-goals
2. User stories and report decisions supported
3. End-to-end architecture
4. Repository and folder structure
5. Configuration and environment-variable design
6. Competitor discovery flow
7. LinkedIn collection flow
8. Raw-data schema
9. Snapshot and history model
10. Analysis taxonomy
11. Report data contract
12. Railway scheduling approach
13. Vercel report approach
14. Manual QA and Google Sheets archive
15. Logging, retry, partial-run, and alert design
16. Security and secrets handling
17. Local development and test strategy
18. Acceptance criteria for a trustworthy production run
19. Milestones in build order
20. Known risks, source limitations, and manual fallbacks

Rules:
- Keep observed ad evidence separate from strategic interpretation.
- Preserve raw snapshots before analysis.
- Design for partial collection and visible failures.
- Do not assume LinkedIn Ads Library structure or access will remain stable.
- Do not include credentials, cookies, or private data in the repository.
- Prefer a narrow, testable first production milestone over multi-channel scope.

Save the approved plan to `docs/ad-intel-plan.md` only after human review.
2

Verify source access, permitted use, and manual fallback

60-90 min

Manually test G2, TrustRadius, and LinkedIn Ads Library for the target category and several known competitors before selecting a collection method. Record whether the required data is publicly visible, consistently searchable, available without authentication, and stable enough for repeatable collection. Document source URL patterns, visible fields, pagination behavior, JavaScript requirements, rate or access limitations, and any terms or organizational policies that constrain automated access. Define a manual or semi-manual fallback that can produce the minimum report when collection is blocked, such as exporting approved URLs and records into Google Sheets. Decide which fields may be stored, how long raw snapshots are retained, and whether creative assets are referenced by URL or downloaded. Create a Source Feasibility table with Source, Access Method, Fields Available, Reliability, Allowed Use, Risk, Fallback, and Owner. Obtain the technical and business owner's approval before building a collector around the observed source behavior.

Output

An approved source-feasibility and fallback assessment covering access, fields, reliability, permitted use, retention, risks, and manual continuity.

G2TrustRadiusLinkedIn Ads LibraryGoogle Sheets
Pro tip

The minimum viable system may use automated analysis on manually reviewed ad URLs. Reliable operations are more valuable than an opaque scraper that silently misses data.

Pro workflow preview

Previewing 2 of 16 steps

Pro membership

Unlock the full workflow

Get the remaining 14 steps, copy-paste prompts, pro tips, tool-by-tool setup guidance, and weekly new workflows.

$9/month

Build and validate the competitor discovery skill
Define the raw-ad schema, snapshot model, and repository contracts
Prototype collection on one competitor and inspect the evidence
Generalize the collector with throttling, retries, and per-source errors
Audit raw records, deduplication, and competitor attribution
Build snapshot history and evidence-based change detection
See Pro plan
3Build and validate the competitor discovery skill
Locked
4Define the raw-ad schema, snapshot model, and repository contracts
Locked
5Prototype collection on one competitor and inspect the evidence
Locked
6Generalize the collector with throttling, retries, and per-source errors
Locked
7Audit raw records, deduplication, and competitor attribution
Locked
8Build snapshot history and evidence-based change detection
Locked
9Define the analysis taxonomy and evidence rules
Locked
10Implement analysis and generate an evidence-linked report
Locked
11Run a local end-to-end acceptance test
Locked
12Commit the production repository and secure configuration
Locked
13Deploy and schedule the backend on Railway
Locked
14Publish the evidence-linked report view on Vercel
Locked
15Add observability, retries, alerts, and partial-run diagnostics
Locked
16Establish the operating cadence and expansion gate
Locked

Expected results

Report generation time

About 5-15 minutes per run after setup

Once competitor discovery, collection, analysis, and reporting are automated, the scheduled run handles the repetitive work while humans review the output.

Competitors monitored

5-20 per category

This is a realistic starting range for LinkedIn ad monitoring before collection reliability and report readability become harder to manage.

Research time saved

4-8 hours per report

The workflow replaces manual ad-library searching, screenshots, theme tagging, and report formatting with a scheduled pipeline.

Operational maturity

Versioned and scheduled agent

GitHub, Railway, and Vercel move the system from local experiment to team-accessible workflow with repeatable runs.

Related workflows

Continue with workflows that share a similar GTM motion, category, or tool stack.