Brand assets a startup needs at seed stage: the smallest system that holds

Most seed stage teams buy a forty page brand book or nothing at all. Here is the kit in between, and the order to build it in.

September 6, 2026 · 10 min read

The first thing a lot of seed stage teams buy is a brand book. Forty pages, a mission pyramid, a section on tone of voice, three pages on what the logo must never do. It arrives as a PDF, it gets linked in Notion, and six weeks later the deck for the next partner meeting is still being assembled from whatever slides happened to exist before.

The second thing teams buy is nothing at all. Assets get improvised: a founder makes the conference banner, the first growth hire makes the launch post, someone drops a product screenshot into a template that shipped with the tool. Each one is defensible on its own. Together they read like four different companies, and the bill arrives later, when a fundraise or a launch needs the company to look like a single thing.

There is a version in between, and it is much smaller than most teams expect. A seed stage brand is not an identity project, it is an operating kit: the fewest decided elements that let people who are not designers produce consistent work quickly, and that survive the next eighteen months without being torn up.

What a seed stage brand actually has to do

Before listing files, it helps to be honest about the job. At seed stage a brand system has three jobs and no others.

1. 01Make you recognisable across surfaces you do not control. An avatar at 32 pixels, a link preview in a group chat, a logo on a sponsor wall, a slide screenshotted into someone else post. Recognition happens in those places long before anyone visits your site.
2. 02Be usable by people who are not designers. At this stage the person making the asset is a founder at 11pm or an operations hire between two calls. A system nobody can execute is a system that does not exist.
3. 03Hold up when the company changes shape. Seed stage positioning moves. The parts that should not move are the ones tied to who you are rather than to what you currently sell, which is why the logo and the type survive a repositioning and the tagline usually does not.

What is not on that list matters just as much: winning a design award, expressing the founding story in a shape, and being finished. A seed stage identity that looks finished is usually one that was over specified for a company that has not settled yet.

The minimum kit, item by item

This is the list we build for pre seed and seed companies. It is deliberately short. Every item earns its place by being used weekly, not by rounding out a document.

That is the whole kit. Seven items, all of them executable by the team without a designer in the loop for every request. If you are weighing this against hiring, our piece on the in house designer versus a creative agency choice covers where each one pays off at this size.

What can wait, even when it feels urgent

Most of the budget teams lose at seed stage goes to work that is genuinely good and genuinely early. These are the items we push out, and the reason we push them out is always the same: they are expensive to build and they are the first things a repositioning invalidates.

• The full brand book. Write the two page sheet now and let the long document be a Series A problem, when there are enough people to need it.
• A custom typeface. It is a real differentiator at a scale you are not at yet, and licensing a good family costs a fraction of drawing one.
• An illustration system or a mascot. It needs consistent volume to pay off, and it is the hardest thing to keep on model when several people are producing.
• A motion identity. An animated logo before you have a stable static one is a decision made twice.
• Original photography direction. Useful when you have a team, an office or a physical product to shoot. Rarely before.
• The manifesto film. It costs more than everything above it combined and it asks strangers to care about your worldview before they have used anything you built.

The exception is the launch slot. If the product ships in a fortnight and there is no film, we should disclose an option of ours rather than pretend it does not exist: ShipTeaser is an automated tool built by the same founder as this studio, and it turns a product URL into a 15 second 1080p motion graphics teaser made for muted feeds, with an optional music bed and no voice-over. The first one is free without a card. What it cannot do is give you a narrative spine or a visual system, so treat it as the post that fills launch week, not as the brand work you still owe yourself.

Build it for the people who will use it

This is where most seed stage brand work quietly fails. The assets are good, the handover is not, and within a month the team is back to improvising because the correct file was too hard to find or too hard to edit.

• Deliver editable templates in the tool the team already opens, not only in the design tool the studio prefers. A perfect file nobody can edit produces zero assets.
• Ship export sizes ready to use: avatar, link preview, the two post ratios, slide background, email header. Resizing is where consistency dies.
• Name files so the right one is obvious at a glance. Logo dark, logo light, logo mark. Not v3 final.
• Put everything in one folder with one link, and make that link the only answer to "where is the logo".
• Name one owner. Not a committee, and not the founder by default: whoever actually publishes most often.
• Include the source files and the usage rights in the handover. Assets you cannot edit later are rented, not owned.

The order to build it in

Sequence matters because each step constrains the next. Building the deck before the type is decided means rebuilding the deck.

1. 01Decide the one sentence the company says about itself. Not a tagline, a plain claim. Everything visual is downstream of it.
2. 02Type and colour next, because they carry most of the recognition and they are cheap to test. Put them on three real screens, not on a moodboard.
3. 03Logo set after type, so the mark sits in the same world as the words around it.
4. 04The deck template, because it is the asset with the most immediate use and it stress tests the system against real content.
5. 05Social frames and the screenshot treatment, derived from the deck rather than designed fresh.
6. 06The two page rule sheet last, written from what the build actually settled instead of from theory.

Two weeks of focused work covers this for most seed companies. That is the shape of our Launch Sprint: a narrative spine and a visual micro system first, then the assets that come out of it. If a fundraise is the forcing event, the deck video work in our piece on a video for a seed fundraising deck picks up exactly where this kit ends.

How to tell the system is holding

You do not need a brand audit to know. Four checks, all of them free, all of them fast.

• The scroll test. Put your last ten posts in a row and look at them small. If they read as one sender, the system works. If three of them could be anyone, the system is decorative.
• The non designer test. Ask someone on the team to produce an announcement post without help. Time it. Anything over ten minutes means the templates are not doing their job.
• The crop test. Take the mark down to 16 pixels and put it on a light and a dark ground. Most seed logos fail here and nobody notices until a conference or a partner asks for the file.
• The reuse rate. Count how many assets last month came from the kit versus from scratch. That ratio is the only honest measure of whether the money worked.

Run these quarterly rather than continuously. Brand systems fail slowly, and the failure shows up as drift across a batch of assets, not inside any single one. When the kit is complete and the output still looks inconsistent, the missing layer is usually art direction: the rules that decide how the pieces are allowed to be used.

How we build one at BuildLore

We start with the claim, not with sketches. One working session produces the sentence the company stands behind, and nothing gets designed until it is agreed, because every visual decision after that point is downstream of it and reversing it later is what turns a two week build into a five week one.

From there it runs in the order above, and the system carries straight into whatever comes next: the launch page art direction, the announcement assets, the first film. That continuity is the point. A kit that has to be reinterpreted for every new surface is not a system, it is a folder. You can see how these hold across tech companies and protocols in our selected work, and the video side of it is set out in our piece on brand video for tech startups.

One habit is worth stealing whether or not you work with a studio: build the deck template early and treat it as the proof. It is the asset with the most real content, the most internal users and the least tolerance for a system that only works on a presentation slide.