A 90-minute hands-on session anchored on one through line: a pre-built Campaign Bill of Materials codification that the team loads, runs, and takes apart to see why each piece matters. The addendum closes with a preview of how the same codification extends across the rest of the stack.
AI-assisted workflows do not fail because the model is weak. They fail because the underlying process was never codified. Where the shape of the work lives in one person's head and varies every time, AI cannot run it reliably. The craft that makes AI dependable is the craft of turning tacit process into explicit structure: three artifacts that work together.
This 90-minute session teaches that craft by showing it in action. Daniel brings a pre-built codification of a real marketing workflow, the Campaign Bill of Materials. The team loads it, runs it, and takes it apart. By the end, every participant owns a running codified workflow, a mental model for how the three artifacts fit together, and a preview of what it looks like when the codification extends into the rest of the stack.
Every codified workflow is a suite of three documents that work together. Each answers a different question. Take one away and the AI workflow stops being reliable.
The ordered process. Steps, sequence, handoffs, decision points. The answer to what happens in what order, and who does what. Without it, the AI does not know where to begin or when to stop.
The reusable template. Structural patterns, field definitions, example inputs, the shape the output must take. The answer to what the work looks like when it is done right. Without it, every run starts from zero.
The assessment criteria. What the output must contain, what counts as complete, what fails. The answer to is this good enough to ship. The trick is to write the rubric so it catches what matters, not everything measurable.
Every real campaign is a bill of materials: a structured decomposition of everything that has to exist for the campaign to run. Strategic rationale. Hero offer. Distribution channels. Content assets. Sales enablement. In most marketing teams, that decomposition lives in someone's head and gets rebuilt from scratch every campaign. The workshop codifies it once, together, and shows what the difference feels like.
The campaign BOM lives in Slack threads, old decks, and institutional memory. Every campaign rebuilds it.
The BOM is an SOP + Standard Work + Quality Rubric. Claude runs the SOP, fills the template, passes the rubric gate.
The session is not a presentation. Every movement follows the same rhythm: a brief demonstration, independent work inside Claude Cowork, and one-on-one coaching in real time. Every participant receives direct guidance at their screen.
Daniel opens each movement with a short demonstration, typically five minutes or less, to establish the concept and show what the exercise looks like in practice.
The team works inside Claude Cowork on their own material. No hypothetical examples. Every participant builds from the workflows and pain points they brought in through the questionnaire.
Daniel moves between individuals, coaching over the shoulder, helping each person refine their inputs, troubleshoot issues, and sharpen their outputs in real time.
Ninety minutes is a tight window. To keep the workshop honest, the default is one shared codification that everyone loads and runs together. When participants engage the pre-work early, there is room to personalize: same codification shape, different substrate. The path is disclosed up front so the team can see exactly what they are getting either way.
Daniel pre-builds one Campaign BOM codification against a representative campaign brief. Every participant loads the same SOP, Standard Work, and Quality Rubric, and runs the same workflow. Discussion happens against a shared reference.
If a participant submits the pre-workshop questionnaire by the deadline and arrives with an open mind, Daniel pre-builds a Campaign BOM codification seeded from that participant's actual campaign context. Each person loads and runs their own. Same structural pattern, personalized substrate.
The gate is low and human. This is not a proficiency test. It is a commitment to engage. Path B is work on Daniel's side, so it is reserved for participants who meet the team halfway. Everyone still leaves with a working codified workflow either way.
The arc is simple on purpose. Five minutes to frame, twenty to see the shape, twenty-five to run the codification, thirty on why it works and what extends next, ten to close. Targets, not handcuffs. The facilitator adjusts based on how the room is moving.
Daniel loads the pre-built Campaign BOM codification in front of the room and walks through it end to end. The SOP: steps in order, who does what, where the handoffs live. The Standard Work: the reusable template, the layer structure, the component tables, the cross-reference columns. The Quality Rubric: the criteria that tell you whether the output is complete, specific, and ready to ship.
Each artifact gets named and defined once. Participants see what each one contributes. They see what would go wrong if any one were missing. The abstraction gets grounded in a concrete example before anyone is asked to run it.
Participants load the same three artifacts into Claude Cowork on their own laptops. On Path B, each participant loads their own personalized version. They feed it a campaign brief and run the workflow. The SOP directs execution. The Standard Work shapes the output. The Quality Rubric scores the result.
This is where the codification stops being abstract. Participants see a complete Campaign BOM emerge: strategic rationale, hero offer, promotional distribution, content assets, sales enablement. Filled out, organized, rubric-verified. What used to take a day in a deck now takes ten minutes and passes its own quality check.
First half, the craft: each artifact's specific contribution, the failure modes when any one is absent or weak, the heuristics for recognizing the next process in your own work that is ready for codification. The team identifies candidates together. Daniel surfaces patterns across the group.
Second half, the horizon: a preview of what happens when the codification extends beyond Claude. The BOM becomes structured records in Airtable. Components route into ClickUp as production tasks. Make.com triggers downstream scenarios when components ship. Cloudflare handles asset deployment. The codification is the spine. The stack is the reach.
On Path B, each participant loads a Campaign BOM codification tailored to their role. Same three-artifact shape. Same arc. The substrate comes from their own context: the campaigns they actually run, the brands they actually market. Select a role below to see what the pre-built codification looks like for that seat.
A Campaign BOM codification oriented to content production. SOP directs the content-asset decomposition. Standard Work defines per-asset spec rows (format, owner, brand voice, SEO target). Quality Rubric scores content-layer completeness and brand-voice alignment.
A running codified workflow that takes a campaign brief and produces a rubric-verified content bill of materials: every asset named, scoped, and tied to the campaign architecture. The craft decisions stay hers. The structural scaffolding stops being rebuilt every campaign.
Ninety minutes is not enough to wire a codified workflow into the rest of the stack live, and that is fine. The last stretch of Movement 3 previews what it looks like when the codification extends, using the exact connectors Daniel has on hand. This is not a demo the team has to reproduce. It is a picture of the upside, concrete enough to plan against.
The Campaign BOM becomes a queryable table. One row per component, layers as views, brand ecosystems as filters. The BOM stops being a slide and starts being a system of record.
Components route into tasks with owners, dependencies, and due dates. The rubric's coverage check becomes the backlog for what still needs producing before the campaign ships.
State changes fire scenarios. A component marked ready triggers the next handoff. A rubric flag pings the right reviewer. The codified process becomes an orchestrated one.
Static campaign assets, from landing pages to microsites to gated resources, deploy through the same codified pipeline. The BOM ships its own distribution layer.
Connectors shown are the ones Daniel has on hand and can demonstrate against. The team's own stack may differ. The principle, a codified workflow as the spine for automation, holds regardless of which tools occupy each slot.
At the end of the offsite day, the facilitator returns and synthesizes what the team produced in the 90-minute session alongside what emerged across the rest of the day. The recap serves three purposes.
Which processes showed up for more than one person? Which are sitting in everyone's head right now with no SOP, no template, no rubric? The workflows that repeat across seats are the highest-value candidates for the next codification.
A recommended order for what to codify next, ordered by frequency, return, and feasibility. The Campaign BOM is the first. The team leaves with a shortlist of the next three or four, sequenced.
A frame for arguing internally that codification is the force multiplier: the return on an hour spent codifying compounds every time the workflow runs, and the same spine carries through the rest of the stack.
Each participant leaves with a running codified workflow and a mental model for the shape. Esty leaves with a team-level view of what to codify next and a frame for making the internal case.
Each participant completes a structured questionnaire 7 to 10 days before the workshop. It takes 15 to 20 minutes. Responses are used solely to personalize the session and are not shared with the broader team. The questionnaire also determines the path. Submit on time and arrive with an open mind, and Daniel pre-builds a Campaign BOM codification seeded from your context. Miss the window, and you still run the workshop on the shared Path A codification.
Path B is not a proficiency test. It is a commitment test: questionnaire submitted by the deadline, and an open mind in the room on the day. Both are low bars and both are honest. Participants who clear them receive a codification built specifically from their own campaign context. Participants who do not still run the shared Path A codification and leave with the same craft; only the substrate differs.
This addendum accompanies the RegScale Marketing AI Enablement proposal dated April 15, 2026.