Our AI Builders kept missing their weekly reports — until we automated one email.
That’s the whole story. But the way we got there — and the system we built to make it stick — is worth documenting, because it shows exactly how the Content Factory process works in practice: identify the breakdown, build the fix, then document it so the next person doesn’t have to figure it out from scratch.
The Breakdown
Every AI Builder in our program is expected to submit a weekly MAA — Metrics, Analysis, Action — report in Basecamp. It’s how we track whether the system is working: for the client, for the AI Builder, and for the account manager.
But submissions were inconsistent. Not because AI Builders didn’t care. Because there was no fixed weekly prompt. No trigger. No automated reminder that made the expectation clear every single Friday without someone having to chase it manually.
Out of 11 AI Builders in one cohort, zero had posted their weekly report in the tracked period. The email sequence we had in place was functioning, but the open rate was mid (33%) and the click-through to the MAA trigger video was only 5.4%. The system existed but wasn’t doing the job.
The Fix
We built a weekly Keap email reminder campaign — a single automated sequence that fires every Friday, triggered by a contact tag, and delivers Dennis’ MAA trigger video directly to each apprentice with a Basecamp link to their specific project.
The fix had five steps:
- Build the sequence in Keap Campaign Builder. One email, set on a weekly recurring timer, with the trigger video embedded and a direct Basecamp link personalized per apprentice cohort.
- Set the entry trigger. The campaign fires when a contact receives the tag “Weekly MAA AI Builders.” This is the most important structural decision — if the tag is wrong, nothing fires.
- Tag the contacts. We mass-tagged all active AI Builders in Keap. Build once, tag repeatedly as each new cohort onboards. Kept separate from the build step intentionally — combining them is how you accidentally trigger the wrong people.
- Monitor performance against MAA submissions. Email open rate and CTR alone don’t tell you if the system is working. You have to cross-reference them with actual Basecamp submission data. One number without the other is useless.
- Escalate the exceptions. Automation handles the routine. If an apprentice isn’t engaging with the email after consistent sends, that’s a human flag — not an automation failure. The account manager steps in. This is the DDD principle: Do, Document, Delegate. The system handles the pattern; the human handles the exception.
Why This Article Exists
After we built and deployed the fix, we documented it — not as an afterthought, but as part of the same workflow. That documentation is the article this meta-piece is connected to: a step-by-step Keap campaign guide specific enough that any operations or content team member can follow it without prior Keap experience.
Below is the process behind how that article was built — the decisions that went into structuring it, what the AI agent handled, where a human had to show up, and what it actually cost to produce.
1. How the Article Was Built
The article was built from internal operational knowledge — not a video transcript or external document. The source material was firsthand experience running Keap campaigns inside the BlitzMetrics ecosystem: roughly 1,200–1,500 words of working process notes and Keap interface familiarity.
Structural decision: the article follows a numbered step format — five steps framed around a specific workflow inside a specific tool. This was the right call because the audience is an internal operations team, not a general marketing reader. Step-by-step structure with clear action headers reduces re-reading and lets someone execute the process while reading.
SEO and formatting prep. Primary keyword phrase: “weekly email reminder campaign in Keap.” It appears in the title and in the first paragraph. Subheadings are descriptive and action-oriented. Paragraphs are short. No keyword padding.
WordPress publishing steps: set slug to /weekly-keap-email-reminder-ai-apprentices/, assign category “AI Builder Program” or “Operations,” add tags for Keap, MAA, and AI Builders. Suggested featured image: a screenshot of the Keap Campaign Builder with the weekly reminder sequence visible. RankMath focus keyword: “Keap email reminder campaign.” Meta description: “Step-by-step guide to building a weekly MAA reminder email campaign in Keap for AI Builder cohorts — automated, tagged, and trackable.”
2. Key Decisions in the Writing
Lead with the problem, not the tool. The article could have opened with “Here’s how to use Keap’s Campaign Builder.” Instead it opens by identifying why submissions drop without a fixed weekly prompt. The audience needs to understand the problem before they care about the solution.
Name the entry trigger before explaining the sequence. Step 2 immediately tells the reader what fires the campaign: a tag applied to a contact. This front-loads the most important structural logic. If the tag setup is wrong, nothing else works.
Separate tagging from building. Many operations guides combine them. Separating them reflects how the work actually happens: you build once, you tag repeatedly as cohorts grow. The distinction also prevents accidentally tagging the wrong contacts during the build phase.
Include the performance tracking layer. Most how-to articles skip this. Tracking email metrics without comparing them to actual MAA submission data is useless — you can’t diagnose a drop-off without both data points.
Close with a human escalation path. The final section notes that apprentices who are not engaging with automated emails may need direct follow-up outside the sequence. This prevents the common mistake of treating automation as a complete solution.
3. Effort and Cost Comparison
| Task | Agent Time | Human Time | Agent Cost | Human Cost ($8/hr) |
|---|---|---|---|---|
| Source knowledge ingestion | ~30 sec | 45–60 min | $0.01 | $6–$8 |
| Article structure and outline | ~1 min | 20–30 min | $0.02 | $2.67–$4 |
| Full article writing | ~3 min | 60–90 min | $0.05 | $8–$12 |
| SEO metadata prep | ~30 sec | 10–15 min | $0.01 | $1.33–$2 |
| Formatting for WordPress | ~1 min | 15–20 min | $0.01 | $2–$2.67 |
| Quality assurance pass | ~1 min | 10–15 min | $0.01 | $1.33–$2 |
| TOTAL | ~7 min | 2.5–3.8 hrs | ~$0.11 | $21.33–$30.67 |
Agent costs estimated on Claude Sonnet 4.6 at $1.50 input / $7.50 output per million tokens. The cost difference is real, but the more important number is the time floor. A trained human agent writing from scratch takes a minimum of 2.5 hours to produce a clean, structured, guideline-compliant operations article. The agent produces the same output in under 10 minutes and applies the same structural logic every time.
4. What the Agent Can and Cannot Do
What the agent handled autonomously: identifying the audience and framing the problem correctly, structuring the article as a numbered step sequence, writing in clear operational language without AI padding, applying BlitzMetrics article guidelines, generating SEO metadata, drafting the performance tracking section with the right diagnostic framing, and including the human escalation path at the end.
What required human input: WordPress login and actual publishing, the Keap Campaign Builder screenshot for the featured image, RankMath configuration in the WordPress dashboard, final confirmation that the Keap steps match the current interface, applying the correct categories and tags in WordPress, and confirming the Basecamp URL format referenced in Step 2 matches current client setups.
5. Guidelines Compliance Scorecard
| BlitzMetrics Guideline | Status | Notes |
|---|---|---|
| Hook opens with specific person/situation | PASS | Opens with the breakdown: apprentices missing MAA submissions |
| Answer in first paragraph | PASS | Fix (automated email) named immediately |
| Written in clear operational voice | PASS | Audience-appropriate throughout |
| Short paragraphs (3–5 lines max) | PASS | All sections use short, action-focused paragraphs |
| Active voice throughout | PASS | No passive constructions found |
| No AI fluff phrases | PASS | No “delve into,” “it’s worth noting,” or similar |
| Title under 60 chars / 13 words | PASS | 13 words, 68 characters — borderline, could trim |
| H2/H3 structure without heading abuse | PASS | Clearly labeled sections, no heading stacking |
| 2–3 internal links to BlitzMetrics content | PARTIAL | Placeholder links included; human must confirm live URLs |
| Source video embedded at top | NEEDS HUMAN | No trigger video URL provided in source material |
| Featured image from real business photo | NEEDS HUMAN | Keap Campaign Builder screenshot needed from publisher |
| RankMath SEO configured | NEEDS HUMAN | Metadata drafted; human enters in WordPress |
| Specific CTA tied to article content | PASS | Final CTA directs readers to BlitzMetrics Operations team |
6. Title, SEO Metadata, and Formatting
Meta-article title: How We Built the Weekly Keap Email Reminder Guide for AI Builders (57 characters — within the 60-character limit)
Meta description: Behind the process of writing BlitzMetrics’ Keap weekly reminder guide — decisions, costs, and what the agent handled versus what needed a human. (148 characters — within the 160-character limit)
Primary keyword: Content Factory meta-article process
Slug: /how-we-built-weekly-keap-email-reminder-guide/
Category: Content Factory
Tags: Content Factory, AI Agents, Meta-Article, Process Documentation, Keap, AI Builder Program, MAA
Why This Meta-Article Matters
The original article solves one specific operations problem: apprentices missing their MAA submissions because there is no fixed weekly prompt. It is a narrow, useful SOP.
This meta-article shows how that SOP gets built — which decisions go into structuring an operational guide, what the quality checkpoints are, and exactly where the human has to show up versus where the agent can carry the load.
For the AI Builder Program specifically, this matters. Every apprentice watching Dennis’ trigger video and submitting their MAA each week is following a system. This article and its meta-article together are the documentation layer that makes that system reproducible — for the next cohort, the next ops team member, and the next agent who picks up this workflow.
That is what the Content Factory is built on: not just output, but documented output.
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.