More insights.
The learning never stops.
Strategy, creative and numbers feed each other every week. What the loop teaches us lands in your inbox.
Most agencies bleed quality at the handoff.
Strategy gets done in one room. Production happens in another. The strategist briefs the designer, the designer makes their best guess, and somewhere in the gap between intent and execution, the thinking that justified the work falls out. You can see it in the output: ads that could belong to any brand in the category, a website saying one thing while the packaging says another. Each piece fine in isolation. Together, not a brand.
This isn't a craft problem. It's structural. When the people making the work weren't in the room where the strategy was decided, they're guessing, and guesses don't compound. So we built production the other way: the same people who run the strategy are involved in the making, the brief carries the thinking in full, and the shoot is planned around the same personas the ads are testing. Here's how each part runs, concretely enough to copy.
The six surfaces
Statics and digital ads, the volume work that carries the strategy in the form audiences see most. Brand films, built for the platform, shaped by where the brand is in its journey. Photo shoots, planned around messaging the strategy already locked. Websites, designed in Figma to handoff, developed by our long-term partner. Brand identity, as a focused digital package rather than a shelf-bound book. Packaging, because the shelf and the feed have to say the same thing. One team across all six is the point: the person making the static knows the brand voice from the website and the visual language from the identity, which is why the output reads as one brand without a style guide enforcing it.
How to plan a shoot that pays for itself
A shoot starts six to eight weeks out, and the sequence matters more than the budget:
1. Name what the shoot is testing. Before location, model or photographer. A film testing a new persona at problem-aware level is a different production than thirty variations of a hero frame for a proven angle. Every later decision cascades from this one.
2. Pull references from the angle, not from what's trending. The reference board serves the message being tested.
3. Choose the location to support the message. Light, airy denim proves itself outdoors. Calm, premium loungewear proves itself in draped muslin. The place is an argument.
4. Cast for the persona, not for general aesthetic appeal. The model is the customer the ad is hunting.
5. Plan every setup to be cut many ways. One day on set should return the film, its breakdowns, GIFs and stills. If a setup only works once, it doesn't earn its hour.
6. Keep strategic context on set. Not to direct, but so that when the in-the-moment call comes, close-up or pull back, it gets made with the strategy in mind.
7. Hand over the full footage library. The shoot's output isn't the film. It's the library the team keeps producing from for months. That's how one call sheet becomes a year of content.
The Lama film, as proof
Two years into working with Lama, they asked for a brand film with, in their words, no obvious purpose for the next four months. A traditional agency would have asked for the KPI. We made a skit: comedic, no product hero, no CTA, because Lama isn't a serious brand and pretending otherwise would have produced a film that didn't sound like them. It became the most-watched thing the brand has put out, 22.3M views, shared in WhatsApp groups, referenced in reviews by people who hadn't bought yet. We could only make it because we'd been inside the same conversation for two years. That's production as part of the strategy: the film wasn't a creative leap, it was the obvious next move.
Volume is a structure, not an output
A scaling D2C brand needs 30-50 ads a month, films every quarter, a site refresh a year. The trap is outsourcing that volume to whoever produces cheapest. The agencies that win on volume run it on a system: three personas, four awareness levels, six message angles, three formats is 200+ distinct creative directions before anyone designs anything.
Volume isn't a creative output. It's a strategic structure that the production team executes against.
Which is why the cadence is tied to the strategy phase, not a content calendar. When CPMr rises and the map points to a new persona, production responds with creative for that persona. The output is reactive to the signal.
Websites: start from the ad account
The website is the brand's most carefully read writing: the visitor arriving from an ad is deciding, and scanning for proof. Most agency-built D2C sites fail because the designers never read the ad account, so they default to "premium quality, crafted with care," words that mean nothing because everyone says them.
The build order that fixes it: pull the converting messages from the ad account. Pull the recurring objections from reviews. Write the copy from both. Structure the page around the decision path the account has already revealed. Ground the visuals in the identity work. Then hand to development. The site ends up saying what six months of spend proved people respond to, instead of what sounded professional in a workshop.
Identity, built for the surfaces that use it
A digital-first identity has to survive contexts the founder never anticipated: a 1:1 crop, a 9:16 Reel, a mobile checkout, a WhatsApp broadcast, a label that becomes a sticker. So our identity work is more rigorous about the system than the look: the rules that keep the brand coherent after fifty contractors have touched it. The output is a focused digital identity package, built for the surfaces the brand will actually appear on. There's a full article on this site about how the identity work itself runs.
The principle
Production is the easiest place to lose the strategy: every step from brief to delivery is a chance for the thinking to fall out. The prevention is involvement, not information. The people producing the work were in the strategy conversations, know which angles came from which reviews, and know what the brand is trying to become. That's the only way the work compounds, and twelve years of producing for the same brands is the evidence.
If you want to see what this looks like applied to specific brands, the work is here. If you want to talk about whether it fits what you're building, book a call.