Blog & Content Engine

When the blog function landed on my plate, there was no clear system — no templates, defined process, or reliable way to scale.

Before anything could be automated, the blog needed a repeatable shape. I built brief templates, brand-specific voice guides for blog content, and an image-generation pipeline for on-brand illustrations — groundwork that made consistent output possible even before an agent was writing anything.

Once that foundation existed, it became the basis for extending Voice Velma into the blog pipeline. She treats each row in the content database as a full creative brief. A single row carries the keyword, search intent, angle, word count target, and brand — provided by our SEO manager — so she has everything she needs to write without back-and-forth. Before drafting, she pulls the relevant brand voice guide and blog format rules from the same reference library that governs all her writing, then writes to that brand's tone and structure using only verified product facts and credible research. Once a post is written, she creates the database entry herself, links it back to the originating pipeline row, and updates that row's status automatically. No manual handoff between "drafted" and "tracked."

We grounded the process in real customer questions by closing the loop with Customer Insights, Customer Success, and Reddit — working from their insights instead of keyword guesswork alone.

Consistently publishing 5+ posts a week without adding headcount, up from 1-4 posts a month when I started. With the templates, pipeline, and back-end databases now in place, that scale was possible without sacrificing quality.