← All work

Case study · 03

LinkedIn Ads AI Optimisation and Human Approval

Converts Campaign Manager exports into evidence-based recommendations, then turns marketer-approved recommendations into controlled manual execution plans.

Two-stage dry-run template · structurally verified
n8nJavaScriptOpenAIAirtableSlack

System overview — a simplified portfolio diagram showing the governed path and human-control boundary.

Project overview

A clearer route from problem to action.

I designed a two-stage system: performance evidence becomes a review queue, then only marketer-approved records become manual change plans. The workflow never edits live LinkedIn campaigns.

Business challenge

The friction behind the workflow

Paid-media teams need to spot inefficient spend without judging low-volume campaigns too early or letting generated recommendations bypass brand, budget and quality review.

How it works

From input to controlled outcome

  1. 01

    Validate and normalise a Campaign Manager export.

  2. 02

    Calculate comparable KPIs deterministically and apply minimum-volume safeguards.

  3. 03

    Generate evidence-bound observations and proposed copy, then validate the schema.

  4. 04

    Place each recommendation in Airtable with Pending approval status.

  5. 05

    Accept approved, complete, unprocessed records and prepare a manual implementation plan.

Designed value

What the system is built to improve

  • Surfaces potentially inefficient spend sooner
  • Makes review more consistent and evidence-led
  • Protects low-volume campaigns from premature judgement
  • Retains professional control over spend and messaging

Responsible AI & human control

Automation supports judgement. It does not replace it.

  • Stable recommendation IDs
  • Minimum-data safeguards
  • Structured AI validation
  • Airtable human approval checkpoint
  • Duplicate-plan prevention and no live API changes

My contribution

Architecture, logic and safeguards

I designed both workflow stages, KPI and validation logic, evidence-retaining recommendation structure, approval handoff and manual Campaign Manager implementation boundary.

Current validation status

Two-stage dry-run template · structurally verified

Both stages are structurally verified dry-run templates. OpenAI, Airtable and Slack actions remain disabled pending private configuration and live validation.

Next project

Monthly Google Ads Performance Analysis

Start a conversation

Let’s build clearer, more intelligent marketing systems.

Based in London and interested in future in-house, contract and selected freelance opportunities across Marketing Engineering, Marketing Automation and applied AI workflows.