TL;DR: I run six content properties publishing 30+ pieces of content per week across kokasexton.com, chiefcontentmarketer.com, mayorofwalnutcreek.com, visibilitycreatesopportunity.com, signalscout.kokasexton.com, and bizflix.kokasexton.com. No writers. No editors. No content managers. Just automated workflows, precise editorial standards, and the right integration between AI and governance. Here is the engine that makes it possible, what each property does differently, and the governance layer that keeps quality from degrading at scale.
Why 6 Properties?
Most people think running six content properties means six teams. Or at minimum, six writers and an overworked editor. That is the traditional model. I went a different direction.
Each property serves a distinct audience with different needs, different vernacular, and different expectations. One voice does not fit all, and one content operation cannot serve a B2B CMO, a Walnut Creek resident, a startup founder, and a content marketer with the same format or tone.
Here is how it breaks down:
- kokasexton.com is my personal brand. First-person practitioner voice. “Here is what I built and what I learned.” It attracts B2B marketing leaders, founders, and operators who want frameworks and playbooks. Three posts per week across six topic pillars.
- chiefcontentmarketer.com is an editorial publication. Third-person byline as Chief Content Marketer. Research-backed, data-cited, tactical. Five articles per week targeted at content marketers and demand gen operators. Publication-grade standards.
- mayorofwalnutcreek.com is hyperlocal news. Short pieces, community-focused, 800 to 1,200 words. Tuesday and Friday cadence. Completely different audience from the B2B properties.
- visibilitycreatesopportunity.com is my consulting program. Every piece reinforces the VCO equation: Visibility x Time x Relevance = Opportunity Density. The audience is founders and executives deciding whether to invest in systematic LinkedIn presence.
- signalscout.kokasexton.com is product. Dark theme, data-dense, benefit-driven. Different voice, different design language, different conversion goals.
- bizflix.kokasexton.com is a video content hub. Cinematic, visual-first. Content here is description, curation, and context for video assets rather than long-form articles.
Six voices. Six audiences. Six content strategies. One engine underneath all of them.
The Automated Pipeline: 5 Stages
Here is the system that makes six properties sustainable without a content team. It runs on 82 automated cron jobs spanning system health checks, content creation, social posting, and CRM enrichment. Each stage feeds the next.
Stage 1: Content Sourcing
Ideas do not come from brainstorming sessions. They come from systems.
My sourcing pipeline pulls from property sitemaps, signal monitoring on competitor and industry content, and structured topic pillar frameworks. Each property has a topic architecture, a set of pillars that define what it covers and what it does not cover. No ambiguity, no “what should we write about this week” paralysis.
The sitemaps serve as content inventory: what exists, what is gapped, what needs updating. Signal monitoring surfaces what competitors and industry voices are writing about right now, so I am not operating in a vacuum. The combination means I never start from a blank page.
Stage 2: Topic Selection and Assignment
A content calendar rotation handles assignment. CCM runs on a 4-week rotation across 5 content types: thought-leader analysis, tactical how-to, data breakdown, framework overview, and research synthesis. KSB rotates through 6 topic pillars, publishing 3 times per week. MWC hits Tuesday and Friday. Each property has its own cadence and its own content mix.
The rotation does the work of deciding what to write. I do not wake up and pick a topic. The system picks it. This removes the single biggest bottleneck in content production: decision fatigue. When you remove the question “what should I write?” you remove the friction that kills most content operations.
Stage 3: AI-Assisted Writing
This is where most people start and stop with AI content. They prompt, they publish, they wonder why their content sounds generic.
The difference in my system is that the AI is working against a style guide, not a blank page. Each property has defined voice parameters, structural requirements, word count minimums, and forbidden language lists. The same approach to AI-assisted writing I detailed in my B2B AI content strategy framework applies here: the AI is a junior writer, not a replacement for judgment. The AI is constrained by rules, not free to write whatever it wants.
For KSB articles: first-person practitioner voice. 1,500 to 2,500 words. No em dashes. No corporate throat-clearing. No “game-changing” or “revolutionary.” Paragraphs must lead with the point in the first 2 to 3 words. For CCM articles: third-person editorial, same word count floor, different visual requirements, different schema, different featured image protocol. Every property gets its own parameters.
The AI produces the first draft. That is it. Never the final.
Stage 4: Editor Review: The Governance Gate
Every piece of customer-facing copy goes through an editor review: revise, edit, proofread. Three passes, never combined. The editor checks for voice violations, banned words, structural issues, and AI tells. Em dashes get removed. Hedging language gets tightened. Paragraphs that do not lead with the point get restructured.
This is the layer that turns AI output into publication-ready content. The AI writes the first draft. The governance makes it good.
What makes it work is that the rules are specific enough to be checked systematically. “No em dashes” is checkable. “No paragraph over 4 sentences” is checkable. “First 2 to 3 words must carry the point” is checkable. When your standards are specific, automation can enforce them. When your standards are vague, “make it sound better,” automation fails and you need humans.
The automation is not the hard part. The editorial standards are. AI writes the first draft. Governance makes it good. Most teams skip the governance layer and wonder why their AI content reads like AI content.
Stage 5: WordPress Publishing and Social Distribution
Articles flow from the editor to WordPress via Plesk deployment. PHP scripts handle post creation, Yoast SEO configuration, JSON-LD schema injection, featured image assignment, and category placement. Every article gets its meta, its schema, its links, and its social copy configured programmatically.
From WordPress, articles feed into the social queue. Each article generates platform-specific social copy for LinkedIn and Twitter. These go into Airtable Juggernaut as Idea status records, waiting for approval before posting. The publishing pipeline is not one piece. It is a sequence of small, automated steps that compound into a complete content operation.
Property-by-Property: The Voice Differences
Each property has a distinct voice codified in a style guide. Here is a quick hit on what makes each one different:
- kokasexton.com: Authoritative practitioner. “Here is the system I built.” First-person. The goal is frameworks and playbooks that B2B marketing operators can apply immediately.
- chiefcontentmarketer.com: Editorial thought-leader. Third-person byline. Publication-grade standards with a light callout box, a pull quote, and a data visualization in every article.
- mayorofwalnutcreek.com: Hyperlocal journalism. Factual, accessible, community-focused. Shorter pieces for a reading audience that wants local information, not B2B strategy.
- visibilitycreatesopportunity.com: Consulting-grade. Executive-to-executive framing. Every article reinforces the VCO equation and is backed by specific, named results and outcomes.
- signalscout.kokasexton.com: Product marketing. Technical-but-accessible. Benefit-driven copy designed to demonstrate product value without feeling like a sales pitch.
- bizflix.kokasexton.com: Visual-first curation. Content serves as context and framing for video assets. Minimalist text, media-heavy presentation.
Six properties, six voices, one engine. That is the architecture. The voice rules are not creative preferences stored in someone’s head. They are documented parameters fed into every AI writing session. The AI does not get to choose the voice. The rules do.
Why Governance Matters More Than AI
Everyone focuses on the AI. The model. The prompt. The output quality. None of that is where the value lives.
The difference between generic AI content and publication-grade output is not a better model. It is a better set of constraints.
The editorial governance layer does five things that raw AI output does not:
First, it enforces voice consistency. The AI does not drift from practitioner to academic mid-article because the review catches drift and corrects it. A KSB article stays a KSB article. A CCM article stays a CCM article. The voice rules are the fence, and the governance checks the fence.
Second, it removes AI fingerprints. Em dashes, hedging language, “In today’s fast-paced,” “It is important to note that.” These are tells. The governance layer strips them systematically. The result reads like a human wrote it because the rules were written by a human who knows what human writing looks like.
Third, it enforces structural integrity. H2 hierarchy, paragraph length, visual element placement. These are not creative choices. They are rules. Rules can be automated. A well-structured article reads better, ranks better, and performs better across every metric.
Fourth, it verifies technical requirements. Schema markup, meta descriptions, canonical URLs, internal links, OpenGraph tags. These are invisible to readers but essential to search performance. The governance layer checks every technical requirement before anything ships.
Fifth, it maintains property differentiation. A KSB article does not read like a CCM article. The governance layer checks property-specific voice parameters before anything ships. Without this, all six properties would converge into one generic voice. The differentiation IS the strategy.
What This Would Cost With Humans
Let me put numbers on this. Running six content properties the traditional way would require real headcount.
A traditional content team handling this output would need two full-time writers at $75,000 to $90,000 each, a part-time editor at $50,000 to $60,000, and a content manager at $80,000 to $100,000. For reference, the Content Marketing Institute’s 2025 salary survey puts the average content marketing manager salary at $95,000, which aligns with these estimates. That is $280,000 to $340,000 per year in salary alone. Before benefits. Before tools. Before the overhead of management, hiring, turnover, and the inevitable quality variance between individual writers.
The automated engine costs effectively nothing in comparison. The infrastructure exists. The AI credits are negligible at this scale. The real investment was building the system, not running it. The Content Marketing Institute’s B2B benchmarks show that while content volume is rising across the industry, only 29% of B2B marketers rate their content strategy as very effective. Most teams are producing more without producing better. That gap is exactly where governance lives.
But here is what the cost comparison misses: the system does not get tired. It does not call in sick. It does not produce great work Tuesday and mediocre work Friday afternoon. Consistency is not a management challenge. It is a system property.
Your 3-Step Starting Point
You do not need to build all of this at once. You need to build the foundation that makes expansion possible. Start here.
Step 1: Define your voice rules first
Before you automate a single piece of content, write down what good looks like. Not “high quality content.” Specific rules. Sentence length range. Forbidden words. Paragraph structure. Opening patterns. Closing patterns.
If you cannot articulate the rules, you cannot automate them. And if you cannot automate them, you cannot scale. Start with one property. Write the style guide. Make it specific enough that someone who has never met you could apply it. Then you can automate against it.
Step 2: Build your content sourcing mechanism
Stop brainstorming. Build a system that tells you what to write. Topic pillars, sitemap analysis, signal monitoring, content gap identification. The goal is to never wonder “what should I write this week?” again.
A content calendar rotation is the simplest version of this. Define your pillars, assign each a day or week, and let the rotation do the deciding. Decision fatigue is the silent killer of content consistency.
Step 3: Automate one property end-to-end before adding more
Pick your highest-value property. Build the full pipeline: sourcing, writing, reviewing, publishing, distributing. Get it working end-to-end before you even think about a second property.
The mistake is trying to automate everything at once and ending up with six broken pipelines instead of one that works. Prove the model on one property. Then the expansion is just configuration, not architecture.
The six-property engine I described did not appear overnight. It accreted one step at a time: first the voice rules, then the topic rotation, then the AI writing layer, then the editor review, then the publishing pipeline, then the social distribution. Each layer built on the one before it.
Start with step one. The rest compounds.
Want to build a governed content engine like this?
If you want to scale content across multiple properties without sacrificing standards, contact me here. I help teams build the governance, workflow, and AI routing that make multi-property publishing sustainable.
