Skip to content

Article

Ops Lab Playbook: Campaign Ads

10X product documentation.

Article Docs 10xDotIn Example Docs
10X Docs
10X product documentation.
Review intent

Browse 10X documentation.

Make the next growth move easier to approve.

10X product documentation.

Ops Lab Playbook: Campaign Ads

Last updated: 2026-03-08

1. Purpose and when to use this

Use Campaign Ads to monitor ad campaign budget pacing, request AI-powered budget reallocation proposals, analyze keyword performance, and review or approve optimization proposals from the Ops Lab creator-operations track.

Campaign Ads reuses existing Ad Intelligence API endpoints but surfaces them in a creator-friendly workflow focused on operational control rather than deep analysis.

2. Included scenarios

  1. CA-01 Check campaign budget pacing: view per-campaign pacing status with overpacing alerts.
  2. CA-02 Get budget reallocation proposals: AI-generated budget shift recommendations.
  3. CA-03 Analyze keyword performance: negative keyword, cannibalization, and expansion opportunities.
  4. CA-04 Review agent proposals for ad changes: list pending optimization proposals.
  5. CA-05 Approve ad optimization proposal: approve a specific budget or bid proposal.

3. Prerequisites

  • JWT operator or creator token.
  • At least one ad account registered via Ad Intelligence.
  • The canonical Gharart.com handle gharart is loaded with the fixed window 2026-02-07 through 2026-03-08.
  • amazon-festive-kitchen, meta-retargeting-cart, and linkedin-wholesale-leads are the reference campaigns for pacing, budget, and approval examples.

4. UI workflow

  1. Open https://app.10x.in/creator/campaign-ads.
  2. Select scenario from dropdown.
  3. Fill in handle and run request.
  4. Review results:
  • Pacing (CA-01): amazon-festive-kitchen should appear as the overpacing priority.
  • Budget proposals (CA-02): expect shifts away from linkedin-wholesale-leads and toward brand-search-q1 or meta-retargeting-cart.
  • Keywords (CA-03): expect cannibalization and negative-keyword opportunities.
  • Proposals (CA-04): list of pending Gharart optimization proposals with timestamps and details.
  • Approval (CA-05): confirm proposal approval with the Gharart approval note.

5. Pacing classifications

ClassificationSpend ratio vs. yesterdayAction
on_track0.5–1.15xNo action needed
underpacing<0.5xConsider increasing bids or expanding targeting
overpacing1.15–1.5xMonitor closely, consider budget caps
critical_overpace>1.5xEmail alert sent automatically, reduce spend immediately

6. Approval workflow

Budget and bid optimization proposals follow the agent proposal framework:

  1. Creator requests optimization with createProposal=true.
  2. Platform creates proposal with PENDING_APPROVAL status.
  3. Proposal appears in CA-04 list.
  4. Operator reviews and approves (CA-05) or rejects.
  5. Approved proposals are logged for audit trail.

7. API workflow

Check pacing

curl -X POST -H "Authorization: Bearer $JWT" \
  "https://ai.10x.in/v2/handles/gharart/ad-intelligence/pacing?notify=true"

Get budget proposals

curl -X POST -H "Authorization: Bearer $JWT" \
  "https://ai.10x.in/v2/handles/gharart/ad-intelligence/budget-optimize?period=30&createProposal=true"

Review proposals

curl -H "Authorization: Bearer $JWT" \
  "https://ai.10x.in/v2/handles/gharart/agent/proposals?status=PENDING_APPROVAL"

Approve proposal

curl -X POST -H "Authorization: Bearer $JWT" \
  -H "Content-Type: application/json" \
  -d '{"note":"Approve Gharart optimization proposal from Campaign Ads"}' \
  "https://ai.10x.in/v2/handles/gharart/agent/proposals/{proposalId}/approve"

8. Failure modes and remediation

  • Empty pacing results: ensure performance data is ingested for both today and yesterday.
  • Budget proposals show no recommendations: all campaigns may be performing similarly; try a longer period.
  • Proposal approval fails with 404: proposal may have been already approved or rejected.
  • 401 unauthorized: verify JWT token.
  • 403 forbidden: verify handle access and operator role.

Canonical walkthrough reminders:

  • amazon-festive-kitchen is the pacing escalation case.
  • linkedin-wholesale-leads is the budget decrease candidate.
  • meta-retargeting-cart is the efficient reallocation target.

Review system

What 10X checks

These checks sit after the main explanation so a reviewer can scan the evidence requirements without breaking the article flow.

Evidence checks

Questions to answer

  • Decisiondocs
    What decision should this article help the team make?

Evidence inputs

Data sources that must stay attached

These inputs keep the recommendation grounded before anyone changes the page, campaign, query target, CRM step, or growth priority.

  • 10xDotIn Docs
  • Playbooks Index
  • Playbook: Platform-Wide Feature Exploration Lab
  • Industry Best Practices
  • E-Commerce and Retail
  • Creator Economy
  • SaaS and B2B
  • Affiliate and Partnerships
  • Events and Ticketing
  • Apparel Prospect Demo

FAQ

Questions before using it

FAQ rows sit near the end, where they help unblock the next action without interrupting the diagnostic flow.

When should I use this article?

Use it before approving growth work that needs evidence, ownership, and a clear next action.

10X

Use 10X with the docs open

Keep product context attached while you review growth work.

Open 10X ->