The minimum viable brand: what pre-seed founders actually need

The most common brief we turn down goes like this: pre-seed team, good product, eighteen months of runway, and they've been quoted $80k for "a full brand identity" they read about on a design blog. They can't afford it, and — this is the part nobody says — they shouldn't buy it. A full identity programme is a wonderful thing for a company that knows what it is. A pre-seed startup does not know what it is yet. It's a hypothesis with a pitch deck.

What it needs instead is a minimum viable brand: the smallest set of decisions and artefacts that makes the company look deliberate, lets it sell, and doesn't box in the pivot that's probably coming. Here's what that actually contains, what you can fake gracefully, and the signals that it's time to invest properly.

What a minimum viable brand is for

Founders tend to want a brand for the wrong reason first: to feel legit. Legitimacy is real, but it's a side effect. The three jobs a brand does at pre-seed are more practical:

Make the pitch deck, site and product look like they came from the same company. Investors and early customers are pattern-matching on coherence. Coherence is cheap; polish is not.
Let the team ship fast without re-deciding everything. Every time someone asks "which blue?", a roadmap slips a little. A minimum viable brand is mostly a decision cache.
Give the sales story a voice. Two sentences about who you are for and why you're different, written well enough that a champion can repeat them in a meeting you're not in.

Notice what's not on the list: awareness, equity, emotional resonance at population scale. Those are series-A-and-beyond problems. Buying them early is buying a tailored suit for someone still growing.

The three artefacts that matter

When we run a minimum viable brand engagement — usually two to three weeks, inside our brand & identity practice — the output is deliberately small. Three things.

1. The positioning sentence (and its two proofs). One sentence: for [this customer], we are the [category] that [difference], because [proof]. Forcing one sentence surfaces every disagreement the founding team has been politely avoiding. We've watched co-founders discover in week one that one of them is building for ops teams and the other for CFOs. Better now, on a sentence, than in year two, on a website. Write the two proofs under it — the things you'd show a sceptic. If you can't name proofs that exist today, the sentence is lying.

2. The visual core: three decisions, not thirty. A primary typeface (a good free one — there are several excellent options with proper italics and full weight ranges), a palette of one ink, one paper, one accent plus tints, and a wordmark set in your typeface. That's it. No custom logomark yet — a well-set wordmark with one memorable quirk (a ligature, a coloured character, an unusual case) carries a startup further than a generic geometric mark ever will. Type as brand isn't a compromise at this stage; it's the smartest version of the thing. The palette decision is strategy: don't pick the category-default colour unless you own the category.

3. The one-page usage doc. Not a 48-page PDF. One page: the sentence, the wordmark files, the three colours with hex values, the typeface with two sizes shown, three example headlines in your voice, and a don't-list with five items. This document exists so the engineer building the landing page at 11pm makes the right call without asking anyone. When it grows past its welcome, graduate to living guidelines — later.

What to fake gracefully

Plenty of grown-up brand surface can be borrowed honestly at this stage.

Photography. Use no photography before using stock photography. A confident typographic site reads as intentional; a handshake stock photo reads as a cry for help. If you need imagery, abstract texture crops and your actual product, screenshotted beautifully, beat a $4 stock subscription every time.

Motion. Skip it. One CSS transition used consistently (the same 200ms ease on every hover) reads more designed than a dozen one-off animations.

Tone of voice. Don't write a voice guide; write five example sentences. "We say: 'Your books reconcile themselves.' We don't say: 'Leverage seamless automation.'" Show-don't-tell beats rules, as we argue in our piece on teachable voice.

The logo animation, the merch, the brand video. No. Crate-digging through a founder's desire for embroidered caps is a real part of our job, and the answer is: sell first, stitch later.

What you must not fake

Three things can't be improvised, and startups regret all three within a year when they skip them.

A name that clears search. Before you fall for a name, do the basic screens: trademark registers in your markets, obvious domain availability, and the pronunciation test. A rename at series A costs ten times a naming sprint now. Our naming field guide has the whole funnel, and the page is free.

Accessibility of the visual core. Check your palette's contrast ratios before you print business cards. An accent colour that fails WCAG at launch will haunt every deck you send for two years.

Honesty of scale. A three-person startup whose website implies forty staff and a global office network is one due-diligence call from embarrassment. Small, stated confidently, is a brand position. "Two of us, both obsessive, reachable" closes more early customers than "enterprise-grade" from a Gmail address.

When to invest properly

The minimum viable brand has an expiry date, and founders usually feel it before they can name it. The signals:

Sales keeps re-explaining who you are. When every call starts with ten minutes of category-setting, the positioning sentence has failed or — more commonly — been outgrown.
The team forks the palette. When the product, the site and the deck are three different blues, the decision cache is full and needs a rebuild.
You're hiring strangers. Ten people can absorb a brand by osmosis. Thirty can't. The moment culture must be transmitted rather than absorbed, the brand needs to be written down properly.
The pivot landed. The minimum viable brand is built to survive one pivot, not three. Once the product's settled and revenue's real, invest — and do it before fundraising, because a coherent brand measurably shortens diligence conversations.

When we worked with a fictional but typical fintech at this transition point, the trigger wasn't aesthetics at all — it was that their own staff described the company three different ways. That's a brand problem wearing an ops costume.

The budget, said out loud

For a pre-seed team in 2026, a minimum viable brand — positioning sentence, visual core, one-page doc, and the assets to execute a landing page and deck — should cost somewhere between a senior hire's monthly salary and nothing at all, depending on whether you buy help or steal these frameworks. It should take weeks, not quarters. Anyone quoting you a six-month "brand journey" before you have ten customers is selling you a journey, not a brand.

Spend the difference on the product. Come back when the sales calls get repetitive. We'll be here, and we'll bring the good typefaces.

Key takeaways

A minimum viable brand is three artefacts: a positioning sentence with proofs, a visual core (one typeface, three colours, a wordmark), and a one-page usage doc.
Coherence beats polish. Looking like one company matters more than looking like a big one.
Fake photography, motion and brand video gracefully; never fake the name screens, colour accessibility or your actual size.
Invest fully when sales re-explains you, the team forks the palette, hiring outpaces osmosis, or the pivot has landed.
Timeline: weeks. Budget: one senior salary-month. Scope: enough to sell.

FAQ

Can we just use an AI logo generator? For a placeholder while you validate, sure — it roughly equals "wordmark in a decent free font". The risk isn't quality, it's that generator output tends toward category-generic marks, and distinctiveness is the one thing you can't afford to lose. Read our full position on AI in brand work before you put generated assets on anything permanent.

Our investors said we need to "look more enterprise." Do we? Usually what they mean is "look inevitable," which is a coherence and confidence problem, not a navy-blue problem. Fix the deck's story and the site's consistency first. If it still feels junior after that, the fix is in the typography, not the sector cosplay.

Do we need a brand name different from the product name if there's only one product? No. One name at pre-seed, always. Naming architecture is a problem you earn by shipping a second product, and the architecture decision is much easier with evidence of what the first name buys you.

How often should the visual core be revisited? Not on a schedule — on the trigger signals above. Teams that review annually tend to tinker; teams that review on triggers tend to change things when change pays for itself.