April 14, 2026
Spent the first two days this week on a brand architecture document for a DTC healthcare client. Hierarchy of names, tone guidance, the reasoning behind typeface choices. Wrote it as I would write any creative brief: the goal was to give future work a decision filter, not a decoration guide.
By Wednesday I was in the Liquid templates. Not handing it off. Opening the repo myself.
“What matters is that the brief and the component share a brain.”
When I wrote that the product name should feel "clinical but not cold," I was also the person deciding what that means in a font stack, a spacing scale, a color temperature. No one had to translate.
That translation layer is where a lot of work goes sideways.
April 16, 2026
Ran into a conversion tracking discrepancy for a different project. The server-side event data was not matching what the ad platform reported. I know what these gaps usually mean technically - event deduplication, browser-side timing, missing parameters. I also know what they mean from the marketing side: the attribution story breaks and campaign decisions get made on bad data.
Most of the time this problem lands on two separate desks. Someone in media buying flags the discrepancy. Someone in engineering investigates. They speak different enough languages that the fix takes two or three rounds before anyone agrees on what the problem actually is.
I fixed it in a day because I could read both sides of the symptom at once. I know what a broken attribution stack looks like because I have lived on both ends of it.
April 18, 2026
A founder asked me this week whether I was a "designer who learned to code" or a "developer who learned design." I understand why the question gets asked. It is how most people make sense of hybrid work.
The honest answer is neither. I was trained as a graphic designer, formally. I taught myself to code because the work kept requiring it. Over time the two skills stopped feeling like separate tracks and started feeling like one practice.
Design is decisions under constraints. So is engineering. The vocabulary is different but the underlying habit is the same: you are trading something to get something else, and the quality of the trade depends on how well you understand both sides.
Where the hybrid actually wins
There are specific situations where a creative director who codes changes the output, not just the timeline.
These are not exotic situations. They happen in most DTC businesses, most marketing technology stacks, most brand-to-build projects. The hybrid is useful in ordinary work.
Where the hybrid loses
This is worth being clear about.
Pure creative directors who do nothing else are better creative directors. Their craft goes deeper because it is all they do. Same for engineers. The specialists who focus entirely are sharper at their edge than any generalist.
The hybrid is not a claim to be best at both. It is a claim to be genuinely capable at both, which is a narrower claim but a more practical one for a certain kind of problem.
If you need a brand identity campaign that requires months of visual development, working with a dedicated creative director is probably right. If you need a complete build - brand positioning, design system, technical implementation, marketing infrastructure - and you want it to hold together without a committee, the hybrid is where that becomes possible.
What this means week to week
I publish what I ship. Every Monday, a dated account of what ran: endpoints, dashboards, builds, fixes. The hybrid work shows up there in ordinary form - not as a thesis, just as a tally. The brand architecture work that came out of one of those weeks is a reasonable example of what this looks like in practice.
The goal is not to demonstrate range for its own sake. The goal is to do work that would otherwise require three handoffs in one conversation.
That is a useful thing to be able to do. It is also a finite thing. There is a real ceiling on how deep any one person goes in two disciplines, and being honest about that ceiling is part of what makes the work trustworthy.
What this week taught me
The people who benefit most from a creative director who codes are not people looking for a bargain. They are people who have been burned by the gap between brief and execution, and who would rather work with one person who understands the whole thing than three people who each understand a part.
The services trap is real, and I am not exempt from it. But the hybrid skill set is not the trap. The trap is selling hours when you should be selling outcomes. The capability itself is not the problem.
FAQ
Is a creative director who codes actually effective at both, or is one skill shallow?
It depends on the person and the timeline. In my case, I have shipped production-grade code and held senior creative direction roles across a 20+ year career in both lanes. Both skills are real. The tradeoff is breadth of surface versus the very narrow specialization of a single-track senior, and for DTC operator work the breadth is usually the win. That is an honest description of what you are getting.
What kind of projects benefit most from this combination?
DTC brand builds that need a system from positioning through Shopify implementation. Marketing technology setups where creative decisions depend on what the event data supports. Any project where the brief and the build need to stay connected to the same reasoning.
Does this mean I am working with one person instead of an agency?
Yes, with some nuance. I use AI agents to multiply output. A DTC analytics audit I run solo produces something comparable in scope to what a small agency team would produce. The difference is that there is one brain coordinating all of it, which changes the coherence of the result. If you want to see what the Shopify + creative side of the hybrid looks like in practice, the theme build case study is a reasonable place to start.
Where does this approach break down?
Anything that requires deep collaboration with a large internal team. Long creative campaigns that need sustained senior creative leadership across a big org. Projects where the technical requirements are at the frontier of engineering depth. I know what I am not and try to be clear about it upfront.
How do you decide when to bring in specialists?
When the work goes past what I can do well. That line exists. I run the creative and technical work I know, and I flag the edges. Working with me is not a bet against specialists; it is a bet that the coordination problem is as real as the execution problem.
Sources and specifics
- Experience grounded in commercial brand and digital projects across DTC healthcare, ecommerce, and B2B SaaS sectors, 2019-2026.
- Attribution discrepancy scenario described is drawn from a client engagement; identifying details generalized to pass GC-wince test.
- "73 API endpoints in Q1" figure referenced in the build log is from public shipping notes, already published at /writing/shipping-with-agents-week-14.
- No specific client names, team sizes, or revenue figures appear in this article per anonymization protocol.