The data disagreed with the brand.
A single-product DTC brand built end to end, from positioning and identity to a custom storefront, pricing and the first paid campaign. The campaign changed the plan.
The call
If the product is for older buyers, change the product, not the brand.
The whole brand was built for women aged 26 to 35. Positioning, naming, voice, imagery, the storefront. Then the first paid campaign ran, small and deliberately cheap, and returned one sale. One sale is not a trend, so I looked at the clicks instead. People 55 and over clicked through at twice the rate of everyone younger, a gap too large to be noise (p ≈ 0.012), and I checked the obvious objection, accidental taps on mobile, before believing it. The comfortable move was to blame the targeting and spend more. I took the finding at face value instead. If the people who actually want this are older, the honest response is to reposition the product as a finishing and makeup item rather than an anti-ageing skincare item, and keep the brand I had already built. That decision is still being worked through.
The situation
Volée started as a single product and the question of whether a brand could be built around it that would outlast it. Everything except the physical product was mine to decide: the name, the voice, the identity, the price, the storefront, the supplier, and how the first dollar of advertising would be spent.
The parts nobody puts in a portfolio
Four identity iterations produced the final direction, supported by a brand voice with its own written style rules. The identity carried into product imagery, video and a custom Shopify theme, built section by section rather than dropped in from a template.
Underneath that, the less photogenic work. Pricing set against real unit economics rather than a competitor's number, which is how a BOGO model that would have killed repeat purchase got rejected. A supplier chosen, then reconsidered, with the reversal written down. Legal pages. A financial model with a traffic-light dashboard. And later, a real customer chargeback, disputed with a full evidence package. A storefront that takes money is also a storefront that occasionally has to give it back.
What the log is for
Twenty-eight decisions are written down with the options that were rejected and the conditions under which each one should be reversed. That last part matters more than it sounds. Writing the reversal condition at the time you decide is what makes it possible, weeks later, to change your mind on evidence rather than on mood.
Where it stands
An early-stage brand with a built identity, a live storefront, and a product angle that the first real data called into question. The interesting outcome here is not revenue. It is that the campaign was cheap enough to be wrong on, and the decision log made it cheap to change course.
Logbook
How it got here
Scroll to move through the steps. Where there is a before and after, drag the handle to compare.
The brand first
Four identities, one promise.
Before a logo, a buyer study with thirteen numbered insights that settled every tie later on. Four identity directions were explored before the one that carries the promise "Your face, finished."
01 / 07Wordmark explorations, and the logo that was chosen.