Almost every product I've watched die didn't die from being badly made. It died from arriving late to the only conversation that mattered: the one with the market. It was polished for months in private, around an idea of the customer nobody had verified, and when it finally shipped it discovered —all at once, and expensively— what it could have known in three weeks. This article is the route I follow to avoid that: launch before you're ready, read what the market gives back, and understand that what you iterate is not only the product. It's the brand too.
Launching late is the expensive decision
There's an asymmetry that changes everything once you see it: launching incomplete costs embarrassment; launching late costs the business. A rough product in ten people's hands returns information within days. A perfect product still sitting on your machine returns nothing, and meanwhile it consumes time, money and —worst of all— conviction. Every week without contact with the market is a week in which your assumptions harden without ever being tested.
That's why an MVP is not "the product, but smaller". A small product is just a small product. An MVP is an instrument of measurement: the minimum version capable of provoking a real response, with money or time on the line, about one concrete hypothesis. If you can't say which hypothesis it measures, it isn't an MVP yet — it's a demo.
And this doesn't mean shipping anything at all. It means being crystal clear about which part has to be finished: the part that holds up the promise. If you promise speed, speed is non-negotiable even if the interface is ugly. If you promise care, the finish matters even if there are only three features. The "minimum" is cut from surface area, never from the axis that gives the offer its meaning.
The mistake I see most
Confusing "it isn't ready yet" with "I don't dare yet". The first is a technical constraint and can be named. The second is fear, and it disguises itself as perfectionism for months.
A value proposition and Product Market Fit are not the same thing
They get used as synonyms and they aren't. The difference between them is the difference between what you claim and what the market grants.
A Value Proposition is a hypothesis you write: what main benefit you offer, to which segment, and why that is preferable to their current alternative — including the alternative of doing nothing, which is the most underrated competitor there is. It's drafted around the jobs that customer is trying to get done, their pains and their expected gains. It's a sentence you could have written today, and it's falsifiable: that's what makes it worth something.
Product Market Fit is the verdict on that hypothesis. You don't declare it; it's granted to you. It exists when demand pulls the product instead of you having to push it: people come back without being chased, they recommend it without being asked, and the cost of getting the next customer stops climbing. Before that point, all growth is proportional to your effort. After it, it stops being.
The practical consequence is uncomfortable: Product Market Fit isn't designed, it's discovered. You get there by iterating on three variables —who you talk to, what you offer them, and how you tell it— until one combination no longer needs you to keep moving. And to iterate you need the one thing only launching provides: a response from the market, feedback, usage patterns, real objections phrased in the customer's words instead of yours.
The business model is flexible; the brand's core is not
As long as it isn't validated, the business model is one more variable. Subscription, one-off payment, freemium, service with an entry product, licence by volume: changing it isn't a failure, it's part of the experiment. I've watched teams cling to the model they started with —"we're a SaaS"— while the market was shouting for another way to pay. Revenue structure is a hypothesis, exactly like the value proposition.
The brand is a different matter. The brand is respected from day one because it's what makes iterations accumulate: if every version looks like a different company, neither of them leaves a mark. But "respecting the brand" doesn't mean freezing the logo. It means holding the core steady —who you are, what you promise, how you behave— while the execution adjusts.
And here's the misunderstanding I want to attack: a brand is not its aesthetics.
Quality axes: what people remember when they don't remember your logo
I call quality axes the dimensions a customer judges your offer by before having any opinion about your design. They're not abstract; they're the things a customer can describe without technical vocabulary:
- Attention — whether there's someone on the other side, and whether that person understands what you're telling them.
- Responsiveness — how long you take, not to solve it, but to show signs of life. Silence is an answer, and it's almost always the worst one.
- Keeping the promise — whether what happened matches what you said would happen.
- Consistency — whether the second time resembles the first. An erratic product breeds more distrust than a mediocre but stable one.
- Recovery from failure — what you do when something goes wrong. It's the axis that generates the most loyalty and the only one that can only be proven at the worst moment.
The sum of those axes is Perceived Quality, and it decides whether they buy from you again. A product with an impeccable visual identity and erratic attention produces a badly remembered brand; the customer won't say "their typography was inconsistent", they'll say "they never replied". That's brand too.
Which doesn't mean the graphic line is decorative: it is literally the mechanism by which people remember you, associate you and tell you apart from the shop next door. Without it the quality axes have nowhere to accumulate: you do excellent work and nobody knows to credit you for it. What doesn't work is a graphic line on its own, with no discourse behind it. Aesthetics hold memory; quality axes hold reputation; the discourse —the Brand Promise, the tone of voice— is what fuses the two. When one of the three is missing, the brand reads as a template, as a whim, or as a lie.
Presence: where the brand has to be in order to be found
Positioning needs somewhere to happen, and that somewhere changes radically depending on the model. It isn't the focus of this article, but without clarifying it the rest stays theoretical.
If the product is physical, your digital presence isn't there to sell online: it's there to push someone into a physical space. That reorders every priority. Geolocation rules —showing up when someone searches "near me"—, reviews rule, because they're the social proof that stands in for trust you haven't earned yet, and local SEO rules, with a well-fed business listing. Social is worked by area, not by global audience: content that someone who lives there recognises. And above all, local identities of trust carry the weight: other neighbourhood brands, users with more presence than you, niche experts the community already validates. Being named by someone local and credible is worth more than a thousand bought impressions, because it transfers reputation, not attention.
One detail of this route gets badly underrated: a review isn't a vanity metric, it's the asset that compounds. Each one improves your position on the map and, at the same time, stands in for the trust conversation you'll never get to have with every person before they walk in. That's why asking for them is part of the product, not part of marketing. Same with the business listing: real opening hours, photos that look like the actual place, replies written by you. And with local alliances —the shop next door, the neighbourhood event, the supplier everyone knows— which are cheap distribution dressed up as courtesy.
If the product is digital, the map inverts: the goal is global and the problem isn't distance but finding whoever has that taste. There the work is locating the target by affinity, understanding digital consumption patterns —where your audience consumes, in what format, at what hour, with what words they name their problem— and instrumenting the journey. That means measuring CAC against direct purchases and not against likes, wiring the Meta Ads pixel and the Google Ads tag to reconstruct the full Funnel, and using that journey to retarget whoever already showed intent instead of paying again for cold attention.
With two warnings I see fail constantly. First: CAC only means something next to what that customer leaves over time; a high acquisition cost isn't bad if the purchase repeats, and a low one isn't good if nobody comes back. Second: the attribution window lies to you if you read it per campaign instead of per cohort — people take time to decide, and last click takes credit that isn't its own. And underneath both, what actually moves the result is where you show up: the niche communities where your audience is already discussing the problem tend to convert far better than the best-segmented cold audience, because you arrive with context instead of with an interruption.
The underlying difference is what's scarce. In the physical world trust is scarce and you buy it with borrowed reputation. In the digital world qualified attention is scarce and you buy it with journey data. Confusing the two strategies is why so many local businesses burn budget on global campaigns, and so many digital products obsess over a single city.
Three tools for iterating without shooting in the dark
Iterating isn't "let's try something". It's a process you can steer. These three are the ones I use most, and they belong to different moments.
De Bono's Six Hats — so criticism doesn't kill the idea too early
De Bono's Six Hats split thinking into six modes and forces the whole team into the same mode at once: white (data and facts), red (intuition and emotion, unjustified), black (risks), yellow (benefits), green (new alternatives) and blue (running the process). The value isn't in the colours, it's in the forced sequence: most product meetings are one person in a black hat against another in a green hat, and neither moves forward. When everyone criticises at once and then everyone proposes at once, the conversation stops being a defence and becomes an analysis. This is the tool for reading launch feedback as a team without it turning into a trial.
Blue Ocean — to stop competing by comparison
Kim and Mauborgne's idea: there are red markets, saturated, where everyone competes on the same attributes and difference is paid for out of margin; and there are blue oceans, where you redesign the axes of value and direct comparison stops applying. The mechanic is concrete: take the factors your category competes on and decide which to eliminate, which to reduce, which to raise and which to create. It's the tool for when the market doesn't reject you but doesn't choose you either, because it's indifferent — which is exactly the problem we'll meet below under parity.
SCAMPER — to force variants of something that already exists
Seven operations on an offer you've already built: Substitute, Combine, Adapt, Modify (or magnify), Put to other uses, Eliminate and Rearrange. It's deliberately mechanical, and that's its virtue: it doesn't depend on inspiration. When the launch returned an ambiguous signal, running the offer through the seven verbs produces ten variants in an afternoon that you can test against what the market actually said. It's the fine-iteration tool, the one that feeds the small gradations of a relaunch.
The three follow a natural order: hats to understand what happened, Blue Ocean to reframe where you compete, SCAMPER to generate the concrete variants you'll test.
The route: two steps, three readings and a cycle
Here's what put my own head in order. After launching, the market can only give you three readings. None of them is an ending: all three flow into a relaunch, and the only thing that changes between them is how much has to be rewritten.
La ruta del lanzamiento: dos pasos comunes, tres lecturas posibles del mercado y un relanzamiento que solo cambia de amplitud. Arrástralo o ábrelo a pantalla completa.
The first two steps are common to everyone. The launch is the moment the MVP exists and the customer can reach it — not when it's announced, but when it's reachable. Positioning is the set of strategies that bring that customer there: presence, channel, price, discourse. After that, information starts arriving, and that information reads in one of three ways.
Market receptivity: you're growing, and now you have to decide
The offer starts growing, in volume or in price, and the market accepts it. It's the best case and also the one most people manage badly, because growth is an anaesthetic. Two warnings before celebrating: check that the pull isn't coming from novelty or from your inner circle —a false positive looks a great deal like Product Market Fit for a few months— and assume that all growth plateaus. The question isn't whether it will happen, it's what you'll do when it does.
From there two paths open, and both are legitimate. The first is to hyper-specialise: keep iterating on experience and marketing to stay relevant, and go deeper into the specific need you detected until you've covered every edge of it within the same line. The second is to leverage the brand you've built to raise different models on top of it. That's where Line Extension (variants within the same category), Brand Extension (using the name to enter a new category) and, higher up, architecture decisions come in: Branded House, House of Brands, Subbrands — that is, how you organise your Brand Portfolio.
The classic examples are useful precisely because they grew in opposite ways. Nivea extended its name into dozens of personal-care categories leaning on a single meaning —accessible, trustworthy care— without fragmenting its identity. Virgin did the opposite and the same at once: it carried an attitude —challenge the incumbent— into unrelated industries, from music to airlines, because what it extended wasn't a product but a behaviour. And Procter & Gamble chose the opposite road: a portfolio of autonomous brands where the shopper buys Ariel or Gillette without knowing, or needing to know, who's behind them. None of the three is the right answer; the right one is whatever your Brand Equity can carry.
Dissatisfaction: there's signal, and the signal is data
This is the common outcome, and it's worth saying without drama: the product is good, just not good enough. They arrived, tried it and weren't convinced. It stings more than flat rejection because it's lukewarm, but it's the best position to learn from, because people interacted and therefore evidence exists.
Here the opening is wider than anywhere else: you can iterate the Target —maybe the offer is right and the segment isn't—, the product, or the marketing. And that's exactly why the tools above pay off most here: it takes an honest process of self-assessment and rediscovery, and that needs separated modes of conversation (hats), a reframing of the axes of value (Blue Ocean) and concrete variants to test (SCAMPER).
What matters is the amplitude of the move that follows. This relaunch is subtle: the same offer with slight gradations, under the same brand. You don't change the name, you don't change the discourse, you fine-tune. Then you position again and check whether the reading improves. A loud relaunch here backfires: you spend the attention you'll need when you genuinely have something different to say.
One nuance that took me a while to see and is worth separating out: silence is not dissatisfaction. If nobody arrived, you don't have a product problem, you have a positioning or channel problem, and the right iteration happens one step earlier —in how you attract— not in the offer. Plenty of people wreck a perfectly good product because they read an absence of data as if it were data.
Negative positioning: when the problem is no longer the product
This is the most atypical case —statistically it's far more likely to go unnoticed than to earn a bad reputation— but it's the one to address fastest. A public event discouraged buying, and now the brand works against the product.
The base move is the same as in dissatisfaction: listen, iterate, relaunch. The difference is amplitude: here the relaunch is complete and under a different brand, because the name drags the association along and no product improvement gets perceived while the customer sees the same logo.
With one filter before spending that money, which is what I'd add: ask whether the association is attached to the name or to the behaviour. If the problem was a one-off reputational event, a new identity wipes the slate. If the problem is the quality axes —you don't reply, you miss deadlines, the promise doesn't hold— the new brand inherits the same behaviour and ends up in the same place with a different logo. Rebranding isn't an apology; it's a reset of perception, and it only works if whatever caused the damage is already gone.
There's a variant of this case that isn't a scandal at all and that shows up constantly: the reputation you earned isn't the one you wanted. You aimed premium and the market read you as cheap. That condemns how you can grow: if your product is handmade, you can't grow on volume —production capacity forbids it— and you need to grow on price, but price doesn't rise on a perception of cheapness. There the relaunch is also one of identity: not because you did anything wrong, but because it takes an image reset to move perception to where the model needs it.
And it starts again
All three exits arrive at the same place: relaunch. Extending, tuning or restarting, but relaunching. That's why I draw it as a cycle and not as a funnel: there's no finish line at the end, there are laps, each better informed than the last. What accumulates between laps isn't only product. It's brand.
The differentiation ladder: where you are and how far you can go
If the cycle above describes what you do, this ladder describes where you stand in the customer's head. Bottom to top:
| Position | What the customer thinks | What to do |
|---|---|---|
| Negative positioning | "Of these, better not" | Restart the identity if the damage is in the name; fix the behaviour if it's in the quality axes |
| Brand Parity | "I don't care whether I buy from you or the shop next door" | Pick an axis of differentiation and build it. This is where Blue Ocean pays off |
| Identity Brand | "This is who they are" (origin, values, history) | Sustained coherence over time; identity isn't declared, it's demonstrated |
| Attitude Brand | "This is how they behave" (tone, stance, ritual) | Consistent behaviour at every point of contact |
| Analogous Brand | "This is like X, but in my category" | Import codes from another category that already has equity |
| First-Order Brand | "This is X" | Hyper-specialise in the need until you're the synonym |
Two readings strike me as the important ones.
First: parity is the default state, not an accident. If you don't actively pick an axis of differentiation, the market files you under parity, and there the only lever left is price — the lever that destroys margin. Getting out of parity has three routes, not one: by who you are (identity), by how you behave (attitude), or by analogy with a universe that already has equity. They aren't successive rungs: they're three doors onto the same floor, and choosing which one is yours is a strategic decision, not an aesthetic one.
Second: the First-Order Brand is the realistic ceiling for a niche brand, and it's a very high ceiling. It's the point where you stop being an option within the category and become the synonym for the object —the product— or for an experience —modernity, quality, care. When someone asks for your brand instead of asking for the category, you've won. And that's where the one lovely risk in this article shows up: taken to the extreme, that synonym turns generic —what happened to Kleenex, Band-Aid or Aspirin— and it can cost you the Registrability of the name. It's a problem anyone would want to have, but it's a real problem, and it's managed through language: you defend the brand as an adjective, never as a noun.
In short
Launch before you're ready, with the minimum that holds up your promise. Write your value proposition knowing it's a hypothesis and that Product Market Fit is the verdict the market gives, not you. Keep the business model flexible while it isn't validated and the brand core steady from day one, remembering that the brand is also attention, responsiveness and follow-through —the quality axes— fused with a graphic line that makes them memorable. Put your presence where your model needs it: local reputation if the product is physical, measured journeys if it's digital. And when the response arrives, read it without fear: there are only three readings, all three are resolved by relaunching, and the only thing that changes is how much you rewrite.
The product gets finished eventually. The brand doesn't: it gets iterated.
