Constellations, patent pending

The Go Live Cliff · Problem

Why Rebrands Fail

Not craft. Not boldness. A perfectly reasoned solution to a person the room misunderstood, and nobody checked.

Quick answer

Famous rebrand failures are almost never failures of craft. The work is usually well made, internally coherent, and defensible in the room where it was approved. What fails is the read of the person on the other end. A team builds a perfectly reasoned solution to a human being they misunderstood, and the misunderstanding only becomes visible after launch, when everyone can suddenly see it at once. The lesson the listicles draw, don't change too much, is the wrong one. Plenty of dramatic rebrands work. What separates them is whether anybody checked how the audience read the change before the money was spent.

The listicle lesson is wrong

Search for failed rebrands and you get the same article twenty times: before and after images, a short paragraph of mockery, and a closing line about respecting your brand equity and not changing too much.

It is a comforting lesson because it is a rule you can follow without learning anything. It is also contradicted by every successful dramatic rebrand, and there are many.

The real variable isn't the size of the change. It is whether the change was tested against the people it was aimed at, or reasoned into existence by a group of smart people in a room with a deck.

Reasoned solution, misunderstood person

Every one of these projects has a rationale. That is what makes them interesting.

Somebody looked at the market, identified a problem, and constructed an answer that follows logically from their premise. The strategy document is coherent. The design executes the strategy. The presentation is persuasive, everyone signs off, and it goes live into a world that never saw the premise and reads the result completely differently.

Figure 1 · Where the fault is, and where the blame lands

A rebrand pipeline from premise to launch, with the failure at the first step Five stages in a row: premise, strategy, design, approval, launch. A marker above the first stage reads the fault is here. A marker above the last reads the blame lands here. A bracket underneath notes that every stage is well made and internally coherent. Premise how people feel Strategy follows from it Design executes the strategy Approval everyone signs off Launch the market reads it THE FAULT IS HERE THE BLAME LANDS HERE Every one of these is well made and internally coherent which is exactly why nobody in the room catches it
Everything downstream of the premise does its job correctly, which is what makes the failure invisible in the approval room. A wrong premise, executed well, fails at full scale and looks like a design failure.

The failure happened at the premise, not the execution. And a premise about how people feel is exactly the kind of thing a room full of experts can't check by thinking harder about it.

There is a longer version of this argument in Joseph Fioramonti's essay on the Tropicana carton, which is the most examined case of the pattern there is and worth reading if you want the full teardown.

Cracker Barrel: a forced choice that wasn’t real

The Cracker Barrel identity change is the cleanest recent illustration.

From the outside, the assumption underneath it reads clearly: the brand believed it was facing a choice between an older demographic it already had and a younger demographic it wanted, and that serving one meant releasing its grip on the other. Everything about the resulting work follows from that premise. Simplify, modernize, shed the specificity.

The premise was the mistake. Nobody had established that the choice was real.

Figure 2 · The choice, and the question that comes first

Two audiences shown as separate circles, then as overlapping circles On the left, the audience you have and the audience you want are drawn as two circles that do not touch, with a note that serving one means releasing the other. On the right the same two circles overlap, and the overlap is highlighted with the question of which forms read as warm and enduring to both. THE ASSUMED CHOICE THE QUESTION NOBODY ASKED FIRST The audience you have The audience you want Serving one means releasing the other Have Want Which forms read as warm and enduring to both?
The left panel is an assumption that has never been checked, and it determines everything the project does next. The right panel is the same question asked in a way that has an answer, and the answer is available before a single comp is built.

The question worth asking first is where those two groups overlap. Which forms, textures and typographic choices read as warm and enduring to both? What do a 32-year-old and a 68-year-old both connect to a place their family has stopped at on road trips for decades? That is a question you can answer with a segmented perception test before you commit, and the answer determines whether you are facing a tradeoff or an opportunity.

The upside on the other side of that question was significant. A brand that visibly built a bridge across generations, rather than picking one, would have been celebrated for it, and the commercial result would have followed. Instead the choice was made on an untested assumption and the market delivered its verdict in public, for free, after the money was already spent.

What “reasoned into existence” looks like at full volume

The most instructive artifact in this whole category is the strategy document Arnell produced for the Pepsi globe redesign, which is public and worth looking at once.

It is a collection of esoteric material with no connecting thread: the Mona Lisa, the golden ratio, the magnetic fields of the earth, gravitational pull, a timeline that runs from the dawn of time to the future and beyond, from underground to deep space. It is archetypal thinking with the brakes off, applied to a soda can.

The document is easy to laugh at. The part worth sitting with is that a room of professionals received that presentation and approved it. That happens when the only evidence in the room is the persuasiveness of the person presenting. Nobody had anything to hold the argument against, so the most confident narrative won.

That is the failure mode in its purest form, and it is operating at lower volume in most brand projects.

Hindsight, available at the beginning

There is an autopsy for every failed rebrand and a case study for every successful one, and both read as obvious. Of course that didn't work. Of course that did. Hindsight makes everything crystal clear.

The thing worth understanding about a perception test is that it produces that clarity at the start of the project instead of the end.

Figure 3 · The same clarity, two prices

Cost to change rising across a project, with the same insight available at two points A curve rising from low to high across test, design, production, rollout and launch. A marker early on the curve reads the pattern is obvious, nothing is built yet. A marker at the top reads the pattern is obvious, the rollout is already paid for. HIGH LOW COST TO CHANGE The pattern is obvious nothing is built yet The pattern is obvious the rollout is already paid for Test Design Production Rollout Launch
The insight is identical at both ends of the curve. What changes between them is whether acting on it costs a few days or a full production run, and the curve is steepest exactly where most projects first hear from the audience.

You put a broad range of visual questions in front of a real audience, and the patterns that come back have a specific quality: they cannot be unseen. Nobody has to argue for them or spin them. The pattern is there, everyone in the room sees it at the same time, and it becomes painfully obvious in retrospect that it was always going to be that way.

That is the same feeling as a post-mortem. It just arrives before the hundreds of hours of production work and the millions in rollout, when acting on it is still cheap.

What to check before you commit

Three questions, in this order, before a rebrand goes into production.

  1. What does the audience currently associate with what you have? You can't decide what is safe to change until you know what is carrying meaning. Some of what looks dated is holding equity, and some of what looks like equity is just old.
  2. Is the strategic premise something you know, or something you assumed? Write it in one sentence. If that sentence is a claim about how a group of people feel, it is testable, and it should be tested before anything gets designed against it.
  3. Where do your segments overlap? Most forced choices between audiences dissolve when you look for the elements both groups respond to. If the overlap genuinely doesn't exist, you want to find that out now, with evidence, rather than discovering it in public.

None of the three is a question about taste, and none of them has to wait for a comp.

Frequently asked questions

Why do most rebrands fail?

Because the strategic premise about the audience was never checked. The design executes the premise faithfully, so when the premise is wrong the work fails at scale, in a way that looks like a design failure and isn't.

Is it the logo change that causes the backlash?

The logo is where the backlash lands because it is the visible part. The cause is usually the reading of the audience that produced the logo change, and the fact that nobody outside the approval room got a vote before launch.

How do you test a rebrand before launch?

Break the direction into elements rather than finished comps, put that range in front of a real audience segmented the way your market is segmented, and read where responses cluster. It is a pre-direction step, not a final approval step, and results come back in under 48 hours.

Can a dramatic rebrand ever be the right call?

Yes, often. The size of the change isn't the risk factor. Changing something that carries meaning you didn't know was there is the risk factor, and that is knowable in advance.