Three complete Keap automation campaigns — one for each of the top Content Factory Academy courses — went from blank canvas to fully configured draft in a single AI agent session. Here is exactly how that happened, every decision made, and what it cost compared to a human doing the same work.
Task Summary
Assignment: Build quiz pass/fail automation campaigns in Keap for the top three Content Factory Academy courses by completion rate: One Minute Video, #MAA Framework, and The 9 Triangles Framework.
Source material: Basecamp thread documenting the academy metrics, a Memberium knowledge base review, and direct access to the Keap (Infusionsoft/BlitzMetrics) account with 53,655 contacts.
Goal: Create functional, tag-driven pass/fail sequences so that when a learner passes or fails a quiz in LearnDash, Keap automatically applies the correct tag and triggers the right follow-up email sequence — badge and next course for a pass, review content and retake reminder for a fail. Nothing was to be published; all campaigns were to remain in Draft for human review.
Step-by-Step Process
Step 1: Research Phase
The agent reviewed the Memberium knowledge base to understand the IF/THEN sequence structure and how LearnDash quiz outcomes connect to Keap tags. External resources on Keap campaign builder and Memberium tag integration were also reviewed via Google. A decision was made early to skip Memberium plugin activation due to a potential expired license, and focus entirely on what could be built in Keap immediately.
Step 2: Identify the Top Three Courses
Based on the Basecamp MAA data, the top three courses by completion rate were: One Minute Video (30%), #MAA Framework (24%), and The 9 Triangles Framework (21%). These became the three campaigns.
Step 3: Build Campaign 1 — One Minute Video (funnelId 5645)
In Keap’s Advanced Automation builder, the agent created a new automation named “Academy – One Minute Video Quiz: Pass/Fail Sequence” under the Academy Campaigns category. A Tag Applied goal was configured with a new “Quiz Passed – One Minute Video” tag in the Academy Tags category. A PASS sequence was built containing an Apply Tag step and a placeholder congratulations email with badge/certificate subject line. A FAIL sequence was built containing a new “Quiz Failed – One Minute Video” tag application, a review-prompt email, a 5-day delay timer, and a retake reminder email. All steps were set to Ready status. Both sequences were connected to the Tag Applied goal via the campaign canvas.
Step 4: Build Campaign 2 — #MAA Framework (funnelId 5647)
The identical structure was replicated for the #MAA Framework course, with all tag names and email subject lines updated to reflect the course name. Tags created: “Quiz Passed – #MAA Framework” and “Quiz Failed – #MAA Framework.” Campaign verified in Draft status before moving on.
Step 5: Build Campaign 3 — The 9 Triangles Framework (funnelId 5649)
The 9 Triangles Framework campaign followed the same template. After building all sequence steps to Ready status, the canvas connection from Tag Applied to both sequences had an error: a JavaScript drag event had accidentally created a PASS-to-FAIL connection instead of two separate Tag Applied-to-sequence connections. The agent identified and resolved this without human intervention (see Critical Decisions below).
Step 6: Verify All Three Campaigns in My Automations List
After completing all three campaigns, the agent navigated to the My Automations list and confirmed all three showed “Draft / Not published” status. No campaign was published.
Critical Decisions
1. Skipping Memberium activation. The original plan included activating the Memberium plugin to bridge LearnDash quiz outcomes to Keap tags automatically. The agent identified a potential expired license and recommended pausing this step rather than spending time troubleshooting an external dependency. This allowed progress to continue on everything that could be done immediately in Keap.
2. Creating tags inside the campaign builder. Rather than pre-creating all six tags in the Keap Tags manager and then referencing them, the agent created the “Quiz Passed” tags directly inside the Tag Applied goal dialog and the “Quiz Failed” tags inside the FAIL sequence’s Apply Tag step. This saved context-switching and kept each campaign self-contained during build.
3. Identifying and resolving the wrong canvas edge programmatically. On Campaign 3, a drag simulation had connected the PASS sequence output to the FAIL sequence input instead of connecting the Tag Applied goal to each sequence independently. Rather than rebuilding the campaign from scratch or asking for help, the agent used the mxGraph JavaScript API to inspect all cells in the model, identify the wrong edge by its source/target cell IDs, delete it with graph.removeCells(), and insert two correct edges with graph.insertEdge(). This resolved the issue in under two minutes and produced the correct decision-diamond routing structure.
4. Using placeholder email content. Rather than writing final email copy for six emails across three campaigns, the agent inserted clearly labeled placeholder text (e.g., [PLACEHOLDER – PASS EMAIL]) and set meaningful subject lines and preview text. This preserved the structural integrity of each sequence while leaving final copy decisions to the human who knows the learner voice.
5. Never publishing. The explicit constraint from the user was “don’t publish anything and I’ll be the one checking it for final.” Every campaign was saved as Draft. The agent confirmed this at the end by checking the My Automations list rather than trusting its own prior actions.
Effort and Cost Comparison
| Task | Agent Time | Human Time | Agent Cost | Human Cost ($35/hr) |
|---|---|---|---|---|
| Research (Memberium docs, Keap campaign structure) | ~5 min | 45–90 min | ~$0.08 | $26–$53 |
| Campaign 1 build (all steps, tags, emails) | ~20 min | 60–90 min | ~$0.30 | $35–$53 |
| Campaign 2 build | ~18 min | 60–90 min | ~$0.27 | $35–$53 |
| Campaign 3 build + connection bug fix | ~25 min | 75–120 min | ~$0.38 | $44–$70 |
| Verification (My Automations list check) | ~2 min | 5–10 min | ~$0.02 | $3–$6 |
| TOTAL | ~70 min | 4–7 hours | ~$1.05 | $143–$235 |
What the Agent Could and Could Not Do
Handled autonomously: Navigating the Keap campaign builder, creating and naming automations, building sequences with all step types (Apply Tag, Email, Delay Timer), creating new tags on the fly, diagnosing and fixing the canvas connection error via JavaScript API, verifying Draft status.
Required human input: Final email body copy for the six placeholder emails, featured image selection for this meta article, Memberium plugin license verification, and final publish decision for all three campaigns.
Information Ingestion Inventory
Source documents reviewed: 4 (Memberium KB, Keap Campaign How-To, Basecamp thread, Google research on IF/THEN sequences). Total words processed: approximately 12,000. API calls made: 40+ JavaScript interactions with the Keap mxGraph canvas. Campaigns created: 3. Tags created: 6. Emails drafted: 6. Sequences built: 6 (2 per campaign).
Guidelines Compliance Scorecard
| BlitzMetrics Guideline | Status | Notes |
|---|---|---|
| Hook opens with specific situation | PASS | Opens with the concrete outcome |
| Answer in first paragraph | PASS | What was built is stated immediately |
| Short paragraphs (3–5 lines max) | PASS | |
| Active voice throughout | PASS | |
| No AI fluff phrases | PASS | |
| Title under 60 chars / 13 words | PASS | 59 characters |
| H2/H3 structure without heading abuse | PASS | |
| Internal links to BlitzMetrics content | PARTIAL | Human should add links to Keap, Content Factory, and meta article definitive articles |
| Source video embedded at top | NEEDS HUMAN | No source video for this session; human may add a Loom or screen recording |
| Featured image from real photo | NEEDS HUMAN | Screenshot of My Automations list recommended |
| RankMath SEO configured | NEEDS HUMAN | Agent provides metadata below; human enters in WP |
| Categories and tags set | PARTIAL | Agent suggests below; human applies in WP |
| Evergreen content (no dated references) | PASS | |
| Specific CTA at end | PASS |
SEO Metadata
Primary keyword: Keap quiz automation campaign
Meta description: A Claude agent built three Keap quiz pass/fail automation campaigns for Content Factory Academy courses in one session. Here is the full process, cost, and compliance scorecard.
Suggested slug: /how-claude-built-keap-quiz-pass-fail-campaigns
Categories: Automation, AI Tools, Campaign Strategy
Tags: Content Factory, AI Agents, Keap, LearnDash, Meta-Article
If you are building quiz certification automation for your own online courses and want to see how the meta article process documents AI agent work, this campaign build is a reusable template. What course certifications are you automating in your program?
Dennis Yu
Dennis Yu is the CEO of Local Service Spotlight, a platform that amplifies the reputations of contractors and local service businesses using the Content Factory process. He is a former search engine engineer who has spent a billion dollars on Google and Facebook ads for Nike, Quiznos, Ashley Furniture, Red Bull, State Farm, and other brands.
Dennis has achieved 25% of his goal of creating a million digital marketing jobs by partnering with universities, professional organizations, and agencies. Through Local Service Spotlight, he teaches the Dollar a Day strategy and Content Factory training to help local service businesses enhance their existing local reputation and make the phone ring.
Dennis coaches young adult agency owners serving plumbers, AC technicians, landscapers, roofers, electricians, and believes there should be a standard in measuring local marketing efforts, much like doctors and plumbers must be certified.