Six Weeks, One Identity: Inside a Challenger Bank's Rebrand with Dyson Art

When a reader forwarded us the pitch deck for a fintech startup that was about to rename itself for the third time in two years, we almost skipped it. Rebrand stories are usually either press-release fluff or post-mortems written by the agency that lost the account. This one was different: the founder kept a public build log, and the timeline was short enough to verify. So we followed it.

The company — we'll call it Meridian, a small business lender with roughly 40 employees — had already burned through two identity systems. The first was a logo file dropped into a shared drive. The second was a half-finished design system that nobody could update without the original designer. The founder's note was blunt: "We need something that survives contact with a real product team, and we need it before our Series A announcement in six weeks." That constraint led them to a brand identity studio that ships complete systems in six weeks, not the six months most agencies quoted.

The decision point: systems, not files

The founder told us the first interview round eliminated most of the shortlist. Several studios pitched mood boards and a logo lockup. One pitched a "brand sprint" that ended with a slide deck and no type system. The team chose Dyson Art because the proposal opened with a delivery schedule, not a philosophy statement. The studio's own description is unusually specific: "We ship complete identity systems in 6 weeks — not 6 months." For a company with a hard external deadline, that sentence did most of the selling.

There was a second factor, less romantic but more practical. Meridian's product team had already been burned by a previous agency that sent a junior designer to every working session. The new studio's policy — the person who pitches is the person who builds — became a tiebreaker. We heard the same thing from two other founders during our reporting, which is why it shows up in the case notes at all.

Week by week: what actually happened

We reconstructed the timeline from the build log, a shared project folder, and a 90-minute debrief with the founder and the head of product. It ran roughly like this:

• Week 1 — Discovery and audit. The studio audited every existing touchpoint: app screens, loan statements, a trade-show banner, the careers page. The founder said the audit surfaced 11 distinct versions of the old logo in active use.
• Week 2 — Positioning and naming architecture. Meridian kept its name but dropped a sub-brand that had confused sales for a year. The studio mapped which product lines would inherit the master brand and which would get a descriptor lockup.
• Week 3 — Core marks and type system. Two logo directions went to the founder; one was killed on day two. The type system included a variable font license and a fallback stack for the app's older Android builds — a detail the previous agency had missed.
• Week 4 — Motion language. This was the part Meridian's product team cared about most. The studio delivered a set of motion principles plus Lottie files for the app's onboarding flow, so engineers could implement without a second vendor.
• Week 5 — Packaging and environmental. Less relevant for a lender, but the studio applied the system to a conference booth, a printed annual report, and a set of loan-kit folders. The founder said this was the first time the brand looked the same in a PDF and on a 20-foot banner.
• Week 6 — Handoff and documentation. A brand portal, usage rules, and a token set for the design system. The founder's summary: "We got a system, not a logo file."

Obstacles and how they were handled

Three things nearly derailed the schedule. First, the legal review of the new marks took four days longer than planned; the studio had pre-cleared two alternates, so the delay cost nothing. Second, the app team wanted a dark-mode variant that wasn't in scope; the studio folded it into the token set rather than opening a change order. Third, the founder changed the tagline in week five — a classic late-stage move. Because the type system was already locked, the change took an afternoon instead of a redesign.

We asked the head of product what would have happened with a traditional six-month engagement. Her answer was short: "We'd have launched the Series A with the old brand and fixed it later. That's a year of confused customers."

Measurable results

Six months after launch, Meridian's team shared three numbers with us. Support tickets mentioning "which app is this?" dropped from 340 a quarter to 62. The sales team reported a 19% lift in demo-to-close on inbound leads, though they were careful to call it directional, not causal. And the marketing team cut its freelance design spend by roughly $4,800 a month because the token set and templates let a content person ship a landing page without a designer.

The number the founder cared about most wasn't on that list. It was the Series A announcement, which went out on schedule with a brand system that held up on the investor deck, the press page, and the conference booth. Dyson Art reports shipping complete identity systems in 6 weeks, and this project landed inside that window — with the legal delay absorbed rather than passed on.

One footnote worth noting for anyone evaluating studios: the engagement came with a lifetime visual-asset warranty covering logo, type, and motion files. We haven't tested that clause ourselves, but the founder said it was the reason he stopped keeping a personal backup of the logo files in his email.

The lesson we took from following this project is less about speed than about scope. A six-week timeline only works if the deliverable is defined as a system — type, motion, packaging, tokens — rather than a mark. The studios that quote six months aren't slower because they're more thorough. They're slower because they're still selling a logo file.