From search demand to pages ready to publish.

I built a content system that connects search research, company knowledge, writing, editorial review and publishing operations.

A closer look at the workExplore the case

Content MachineSystem overview

Choose the change

Search demand, buyer questions, and existing pages become a page-level brief.

Build the page

Research and company knowledge support writing, editing, and fact checks.

Review and publish

A person edits the HTML and approves publication; the system prepares the handoff.

Check the result

Verify the live page and links, then compare search and AI visibility.

Process illustration · Read the build decisions below

01 / The problem

The work between an idea and a published page

A page can be well written and still target the wrong demand, repeat an existing page or miss the company’s point of view. Producing useful content required several kinds of judgment, then a handoff between writers, reviewers and publishers. I wanted those decisions to remain connected as the work moved from an idea to the website.

02 / The build

How I built it

The system starts with the business and the existing website. It examines demand, competitors and page relationships, then proposes a cluster or page plan. Buyer questions can feed that planning through the company’s existing conversation intelligence.

For each page, specialist agents research the topic, use the appropriate company knowledge, write the content and check facts, voice and the brief. The output is an editable HTML page shaped around the website’s actual format. A content manager can review the page in context and return comments for the system to apply.

The workflow then prepares the publishing task and tracks the links that other pages need to add. Search and AI-visibility reporting help the team return to the original hypothesis after publication. People retain strategic direction, editorial decisions and the final publishing work.

My role

I designed and built the workflow and its supporting tools, working with the people responsible for content, product knowledge and publication. I improved the writing and review process through real editorial feedback. The system used existing company sources, including a conversation-intelligence platform built by a colleague.

03 / In practice

One page, with a complete handoff

Search opportunity

What the team receives

Buyer question, evidence, target page

Decision

Create or improve?

Content draft

What the team receives

HTML, source-backed claims, review notes

Decision

Ready for the editor?

Publication

What the team receives

Metadata, images, link tasks

Decision

Approved to publish?

Follow-through

What the team receives

Live-page checks and visibility comparison

Decision

Keep or revise the hypothesis?

Illustrative workflow example. This is a reconstruction of the process, not a production screenshot or company data.

04 / The result

10×publishing output

+300%AI visibility

The project reached 10× publishing output and +300% AI visibility. In routine editorial use, most materials could be reviewed and approved in approximately 15–20 minutes; more critical pages needed deeper work. The result was a usable content-production process with explicit human decisions at the points where they mattered.

Keep exploring

LinkedIn Signals

I built a system that turns LinkedIn engagement into qualified prospect worklists, with CRM context and follow-up measurement.

Next case