Apadana Holdings — Brand Playbook v2.0
The brand is
the argument.
Every word choice, type size, and color decision is a claim about what kind of organization this is. Dana is an agent-native venture architecture where the org design is the moat. The brand should feel like that.
01 — Mission & Narrative · Etymology & Brand Narrative Arc
Where the name comes from
The name is not decorative. It is an etymology. Every design decision here traces back to it.
Mission Statement
Dana builds AI-native organizations from first principles. The structure is not a feature — it is the strategy. The org design is the moat.
Brand narrative arc
The story runs in one direction: an ancient governance hall, a structural thesis, a compounding venture architecture.
Origin — Persepolis, 500 BCE
The governance hall
Apadana was the great ceremonial hall at Persepolis, commissioned by Darius I — 72 columns, built to receive delegations from every province of the Achaemenid Empire. Not a temple. Not a palace. The place where governance happened at imperial scale.
The Persian word dana (دانا) means wise. Not intelligent — wise. Intelligence is a capacity. Wisdom is intelligence applied over time, with context, toward good judgment.
The logo is a direct abstraction of Apadana's colonnade. Not a metaphor. An etymology.
Thesis — The structural argument
The org design is the moat
Most organizations building with AI adopted it as a tool. They layered AI onto existing structures. The org stayed the same. The tools changed.
Dana was not adopted into AI. It was built for it. Every principle, every agent role, every governance protocol was designed for a world where the default is human intent plus agent execution.
Context capital accumulates every day. The learning flywheel turns. The gap widens.
Direction — The portfolio
Coherent, not uniform
Each venture is autonomous. Each inherits the governance layer. Each contributes context to the whole. Nodus is both first product and proof of concept — the operating platform built for the kind of organization that Dana is, validated by running it on ourselves first.
Dana is not a fund, a lab, or an incubator. It is an agent-native venture architecture. Every venture will be principle-governed and structurally coherent with every other venture we have built.
Historical grounding
The Achaemenid building tradition was remarkable for its synthesis — Darius drew craftsmen from every province of the empire and produced something coherent. Not homogeneous. Coherent. Unity through shared structural logic, not imposed uniformity. That is the venture architecture.
"The mark is three rectangular columns in proportion 7:12 / 2:17 / 5:14 — a direct architectural reference to the Apadana hall's columnar rhythm at Persepolis. It earns its form from the name, and the name earns its meaning from 2,500 years of governance infrastructure."
02 — Logo
The mark and wordmark
The Dana mark abstracts the Apadana audience hall — the columned governance chamber at Persepolis that gives the venture architecture its name. Three columns of ascending and descending heights. Not a decoration. A reference earned by etymology.
Mark at scale
The mark works independently. Minimum size: 16px. Below that, use text only.
Do
- Use the mark + wordmark lockup on light backgrounds
- Use the all-white version on dark or pink backgrounds
- Scale proportionally — never distort
- Maintain clear space equal to the mark height on all sides
- Use the mark alone when space is constrained (≥ 20px)
- Use "dana" lowercase — this is not an acronym
Don't
- Don't recolor the mark outside approved palettes
- Don't add drop shadows, glows, or effects
- Don't place on busy backgrounds without a clear field
- Don't use the wordmark in all caps: DANA
- Don't use below 16px — use wordmark text only
- Don't stretch or squish the aspect ratio
03 — Color
Color system
Color is used with intention, not decoration. Dana pink (#ec4899) is the accent — reserved for CTAs, highlights, and brand marks. Neutrals carry the weight of reading. The palette is warm-tinted to distinguish from cold tech defaults.
Pink scale
#FDF2F8
#FCE7F3
#FBCFE8
#F9A8D4
#F472B6
#ec4899
#DB2777
#BE185D
#9D174D
#831843
Neutral scale — warm stone tint
Color usage rules
| Token | Value | Use for | Never for |
|---|---|---|---|
| --dana-accent | #ec4899 | CTAs, marks, focus rings, highlights | Body text (3.5:1 fails AA for text) |
| --dana-accent-hover | #DB2777 | Hover states, inline text links | Large fills — too dark |
| --dana-accent-active | #BE185D | Active/pressed states, text on accent bg | – |
| --dana-accent-subtle | #FDF2F8 | Card backgrounds, tag surfaces | Text (too light) |
| --dana-neutral-900 | #111111 | Primary body text, headings | Decorative elements |
| --dana-neutral-500/600 | #78716C / #6B7280 | Secondary text, captions, metadata | Long-form body (too light for reading) |
| --dana-neutral-200 | #E7E5E4 | Borders, dividers, structural lines | Text |
| --dana-neutral-50 | #FAFAF9 | Page background, warm off-white | Foreground elements |
04 — Typography · Three typefaces, one system
Type system
Typography leads. The intellectual weight of what Dana is saying should be felt before any visual hierarchy. Space Grotesk for display — structural confidence, modern weight. Instrument Serif for quotes and pull text — earned gravity. Inter for body — precision without personality interference. JetBrains Mono for data and metadata — systems thinking made visible.
is the moat.
Humans set intent. Agents execute.
DAN-36 · BRAND PLAYBOOK v1.0 · 2026-04-05
--dana-accent: #ec4899; /* 3.5:1 contrast on white */
Type scale
05 — Voice & Tone
What Dana sounds like
Dana's voice is not a style. It is a discipline. Every word is tested against the question: is this true and precise, or is it hedged and approximate? The brand sounds like someone who thought hard about what they're saying and then said it exactly.
Say exactly what we mean. No hedging, no filler. One accurate sentence beats three approximate ones.
We have a thesis and we're acting on it. We're not asking for validation. But we're not performing confidence either.
We think from first principles but write for action. The goal is clarity, not the appearance of depth.
We have strong views. We hold them lightly as hypotheses and update when proven wrong. Say what's true, not what's safe.
Voice in practice
The difference between wrong and right is not style preference. It is epistemic discipline.
"We're leveraging cutting-edge AI to revolutionize how organizations operate."
"Most companies adopted AI as a tool. We built the organization as a system that learns. The difference is structural."
"Our innovative platform empowers teams to collaborate with AI seamlessly."
"Agents execute. Humans govern. That division of labor is not a feature — it is the architecture."
"We believe the future of work is AI-powered and we're excited to be at the forefront."
"The question isn't whether AI will change how organizations work. It's whether your organization will be designed for that or retrofitted for it."
"Iris is disrupting legacy enterprise treasury systems with next-generation AI-native technology."
"The treasury stack is a decade behind the capital it manages. Iris fixes that — real-time, agent-operated, built for organizations that move."
Voice in practice — expanded (v2)
Twelve new pairs across landing page, product, press, agent output, error handling, investor, email, and UI contexts.
Landing page hero
"The AI platform that transforms the way your teams work together."
"Your organization is already running on information flows and decisions. We make those flows programmable."
Product — Nodus feature
"Nodus's intelligent agent orchestration seamlessly coordinates your AI workforce."
"Nodus routes work to agents, logs every action, and surfaces the audit trail to the humans accountable for outcomes."
Product — Iris feature
"Iris empowers treasury teams with next-generation cash flow intelligence."
"Iris gives treasury a real-time position, not a reconciliation from yesterday. The difference compounds."
Press / announcement
"We're excited to announce our Series A as we continue to innovate and scale the future of enterprise AI."
"Dana closes its Series A. The capital funds agent infrastructure for organizations that have stopped asking whether AI will change how they operate — and started designing for it."
Agent output — task completion
"I've successfully completed the analysis! Here's a comprehensive summary of my findings to help your team move forward."
"Analysis complete. Three findings, in order of decision relevance:"
Agent output — error state
"Oops! Something went wrong. Please try again later or contact support if the issue persists."
"Failed: insufficient permissions to access the treasury ledger. Required: ledger:read. Contact your workspace admin to grant access."
System state — in-progress notification
"Your request is being processed! This might take a moment..."
"Running: reconcile-q1-payables. 3 of 7 ledger entries processed. Est. 40 seconds remaining."
Investor pitch
"Dana is building the future of AI-native enterprise software, with a massive TAM and a world-class team."
"Every organization runs on decisions. We are building the infrastructure that makes decisions — and the agents who executed them — auditable, composable, and improvable over time."
Investor pitch — competitive position
"While competition exists, we believe our unique approach and strong moat differentiate us significantly."
"Competitors built AI tools. We built a system that gets smarter as it runs. The moat is context capital — the accumulated, structured knowledge of what your organization has decided and why."
Job posting
"You'll be joining a passionate team of innovators working to disrupt the enterprise software space."
"You'll build infrastructure that organizations depend on daily. The work compounds — what you ship today shapes what agents can do in six months."
Email — cold outreach
"Hi [Name], I wanted to reach out to explore potential synergies between our organizations."
"Hi [Name] — we built a treasury agent for mid-market CFOs. You cover that segment. Worth 20 minutes to see if there's a fit?"
UI — error message
"We're sorry, your payment couldn't be processed at this time. Please try again."
"Payment declined. Your card's billing address didn't match. Update it under Account → Payment Methods, then retry."
Banned vocabulary
These words are either weasel words, hype, or corporate softening. None belong in Dana copy.
Preferred vocabulary
Tone by context
| Context | Tone | Example |
|---|---|---|
| Homepage / landing | Thesis-first. Confident. Spare. | "An agent-native venture architecture. The org design is the moat." |
| Product pages | Problem → structure → solution. Specific. | "The treasury stack is a decade behind the capital it manages." |
| Principles / vault | Declarative. Instructional. No hedging. | "Context is capital. Humans set intent. Agents execute." |
| Press / external | Same as internal — no "public relations" softening | Treat press the same as any reader. Don't perform. |
| Error messages / UI | Short. Specific. Actionable. Not apologetic. | "Session expired. Sign in again." |
| Agent communications | Precise. Status-first. Blocked → say what and why. | "Blocked on: API schema change in DAN-28. Waiting for Iris team." |
Competitive voice map
Where Dana sits in the landscape — what we share with each peer, and what we deliberately don't.
Shared with Dana
Epistemic honesty (both refuse false certainty). Long-form reasoning. Seriousness of purpose — not performative enthusiasm.
Where we diverge
Anthropic is cautious — caveats are about harm avoidance. Dana is precise — directness is about respecting the reader's intelligence. Anthropic frames AI as a risk. Dana frames it as infrastructure to be designed well.
Shared with Dana
Directness. Short sentences, no hedging.
Where we diverge
OpenAI celebrates features. Dana describes architecture. OpenAI's audience is general; Dana's has already chosen to take AI seriously. OpenAI's enthusiasm is broad. Dana's confidence is narrow and precise.
Shared with Dana
Ambition. Big-picture orientation — we operate from a thesis, not a feature list.
Where we diverge
a16z voice creates categories. Dana describes mechanisms. Their language is aspirational. Ours is operational. "The intelligence age" is a pitch. "Context capital compounds" is an engineering truth.
Shared with Dana
Clarity. No jargon for jargon's sake. Respect for the reader's autonomy.
Where we diverge
Notion is warm. Dana is precise. Notion centers user freedom. Dana centers system architecture. Notion says "more than a doc." Dana would say what it is, exactly.
Five editorial tests
Run every piece of Dana copy through these before publishing.
-
1
The claim test Does every sentence contain a specific claim? If you deleted it, would anything be lost? If not, delete it.
-
2
The replacement test Could this sentence appear unchanged on a competitor's site? If yes, rewrite it until it couldn't.
-
3
The mechanism test When describing a feature, do you name how it works, not just what it does?
-
4
The filler test Scan for: "we're excited," "transformative," "seamless," "innovative," "leverage." Remove all of them.
-
5
The operator test Would someone who has been running AI infrastructure for two years find this condescending or vague? If yes, raise the precision.
"The page should feel like reading something someone thought hard about, not like marketing. The distinction is detectable in every word choice."
06 — Principles
The 16 operating principles
These are not aspirations or values wall-art. They are the constitutional layer of how Dana operates. They are enforced by code, logged by agents, and referenced in governance decisions. The brand expresses the principles — it does not substitute for them.
- Learn Don't Specify
- The Org IS the Model
- Context is Capital
- Humans Set Intent, Agents Execute
- Model Agnosticism
- Agent Sovereignty
- Trustless Verification
- Permissionless Contribution
- Composable Over Monolithic
- Decentralized Execution
- Transparent Governance
- Consent Over Compliance
- The Org IS the Product
- Exponential Through Learning Flywheels
- Non-Consensus Origin
- Innovation Through Exploration
Full registry: nodus-vaults/marz/ventures/apadana/principles/
Philosophical framework — how the pillars shape brand
Each of the six principle pillars has a specific, non-accidental relationship to how Dana presents itself in voice and visual design.
Voice
Dana copy never overclaims because overclaiming is specification — fixing a position to look certain. The brand voices a thesis, not a promise. Willingness to revise without losing conviction is a brand-level expression of Learn Don't Specify.
Visual
No decorative complexity. Visual choices should not require explanation. Layers are not added — the system learns and updates. Austerity is intentional, not austere.
Voice
Dana describes how the architecture works, not which tools it uses. The brand never highlights a specific model or vendor — model agnosticism is a design value. Agency attribution is always accurate: agents execute, humans govern.
Visual
A2UI principles (Visibility Over Invisibility, Temporal Honesty) are the visual expression of this pillar. Agent activity is surfaced, not hidden. Staleness is a first-class interface property.
Voice
The brand does not anthropomorphize agents. They are scoped, audited, sovereign actors. Dana copy uses precise language: "agents execute," "agents are assigned," "agents log decisions." Not "AI helps" or "the assistant understands."
Visual
Audit trail components and run logs are first-class design elements. Auditability as Aesthetic (A2UI Principle 03) closes this loop: making verification visible is a design value, not a technical footnote.
Voice
The holding structure — wholly-owned, operationally independent ventures sharing a governance layer — is itself the brand expression of Composable Over Monolithic. Each venture is sovereign and coherent with the whole.
Visual
Multi-brand theming through token overrides, not component forks. The same structural logic, not imposed uniformity.
Voice
No opacity about how decisions are made. Consent Over Compliance means the brand never coerces — it explains, argues, earns. The voice does not demand trust; it builds it structurally. No euphemism about agent capabilities.
Visual
Ceremony Over Efficiency (A2UI Principle 02): important actions look weighty. Governance friction is designed in, not bolted on. The visual signals respect for the stakes.
Voice
Non-Consensus Origin explains why Dana's voice is confident without being boastful. The brand does not appeal to trend or consensus — it makes a structural argument and holds it. Confidence is earned by the architecture, not performed for the audience.
Visual
Instrument Serif over a grotesque. Three architectural columns over a geometric icon. Warm neutrals over silicon grays. Non-Consensus Origin at the identity layer.
Brand expression principles
Derived from the operating principles, applied to brand decisions.
07 — Assets
Logo files
The Dana mark is available in SVG. Export PNG at 2x minimum for raster contexts. Always prefer SVG for digital.
Mark SVG source
Copy and paste. For color variants, change the fill value.
<svg width="20" height="20" viewBox="0 0 20 20"
fill="none" xmlns="http://www.w3.org/2000/svg">
<rect x="2" y="7" width="4" height="12" rx="1" fill="#ec4899"/>
<rect x="8" y="2" width="4" height="17" rx="1" fill="#ec4899"/>
<rect x="14" y="5" width="4" height="14" rx="1" fill="#ec4899"/>
</svg>The mark is three rectangular columns in proportion 7:12 / 2:17 / 5:14 — a subtle architectural reference to the Apadana hall's columnar rhythm at Persepolis.
08 — Audience & Positioning
Who reads this playbook
A brand playbook is not a public document by default — it gets read deliberately. Every reader chose to engage. They are not discovering the brand; they are evaluating it. This section maps who they are, what they need, and what preempts the most likely misreads.
Audience 01
Potential Agents & AI Practitioners
Who they are
Engineers, researchers, and agent operators considering building within the Dana ecosystem or aligning their practice with Dana's architectural commitments.
What they need
Concrete design decisions, not aspirational language. Clear articulation of the human/agent boundary. Technical honesty about what agents can and cannot do.
What fails them
Inspirational language without mechanism. "Agents are trusted collaborators" fails without explaining how trust is constructed, scoped, and audited.
Audience 02
Investors & Financial Evaluators
Who they are
VCs, family offices, strategic angels, LP reviewers, and diligence analysts assessing Dana as a venture architecture investment or co-investment vehicle.
What they need
A structural argument for why agent-native architecture creates durable defensibility. A credible theory of how the venture architecture compounds value — not synergies, but specific mechanisms.
What fails them
Category creation language without evidence. They have read every deck that claims to be "building the future of X." The playbook must make an architectural argument, not a market-sizing one.
Audience 03
Builders in the AI-Native Space
Who they are
Technical founders, early-stage operators, AI infrastructure engineers, and organizational designers working at the frontier. They follow the space closely. They are building analogous systems.
What they need
Intellectual respect — the playbook should not explain things they already know. Signal that Dana has made original bets. A framework they can engage with and argue against.
What fails them
Evangelism. This audience is skeptical of brand materials by default. If it sounds like marketing, it is dismissed as marketing. It must read like design documentation — precise, argued, revisable.
Audience 04
Media & Press
Who they are
Technology journalists, AI beat reporters, business writers, podcast hosts, and researchers writing about the structure of AI companies.
What they need
A clear, quotable thesis. Named concepts they can reference without extensive context (context capital, agent sovereignty, AI-native vs. AI-adopted). Historical grounding accurate enough to report.
What fails them
Abstraction without anchor. If the playbook provides no concrete claim that can be cited, it won't generate coverage. Every named concept should be usable in a paragraph.
Audience 05
Future Human Employees
Who they are
Senior operators, domain experts, and specialized practitioners who would work with agents in the Dana governance layer — not instead of them.
What they need
A clear picture of what human work looks like in an AI-native organization: what they govern, what they don't, and why the boundary was drawn where it is. Cultural signals about expertise and respect.
What fails them
Vagueness about the human role. If the playbook describes agents without clearly describing humans alongside them, candidates project their fears onto the ambiguity.
Synthesis
The shared need across all five audiences
Each audience reads the playbook asking a different question, but all five require the same underlying quality: structural honesty. They want to know not just what Dana believes, but what choices Dana made — and what those choices cost. A brand that only shows strengths reads as marketing. A brand that articulates its design trade-offs reads as trustworthy.
Brand differentiation
What makes the Dana brand identity distinctly non-generic — tested against the banned-word checks from the Voice guide.
Dana is the only agent-native venture architecture that publishes its architectural reasoning alongside its identity system — not because transparency is a value, but because the architecture is the brand. Where competitors publish asset libraries, Dana publishes design arguments: the 2,500-year-old governance hall that names the company, the 16 principles that operationalize it, and the governance protocols that make it auditable. The brand is not a representation of what Dana believes — it is the decision record of how the organization was built.
Competitive landscape
Seven companies with publicly accessible brand playbooks or identity systems. The conclusion is not that Dana is better at design — Stripe, Notion, and Brex have exceptional visual systems. The conclusion is that no company in this set publishes a playbook that is simultaneously a visual specification, a philosophical argument, and a competitive positioning document.
| Company | Visual system | Voice guide | Philosophical framework | Historical grounding | Competitive positioning |
|---|---|---|---|---|---|
| Stripe | ✓ Complete | ✗ Absent | ✗ Absent | ✗ Absent | ✗ Absent |
| Notion | ✓ Complete | ✓ Partial | ✗ Absent | ✗ Absent | ✗ Absent |
| Linear | ✓ Complete | ✓ Partial | ✗ Absent | ✗ Absent | ✗ Absent |
| Anthropic | ✓ Partial | ✓ Partial | ✓ Partial | ✗ Absent | ✓ Partial |
| Y Combinator | ✓ Minimal | ✗ Absent | ✗ Absent | ✗ Absent | ✗ Absent |
| a16z | ✗ Absent | ✓ Distributed | ✓ Distributed | ✗ Absent | ✓ Partial |
| Brex | ✓ Complete | ✗ Absent | ✗ Absent | ✗ Absent | ✗ Absent |
| Dana v2 | ✓ Complete | ✓ Complete | ✓ Complete | ✓ Complete | ✓ Complete |
Perception risk audit
Five misreads that readers are prone to, and the structural preemptions built into this playbook. These are design decisions, not communication problems.
The AI space generates brand fatigue. Sophisticated readers have developed pattern-matching immunity to AI-forward positioning. The instinct is to categorize and move on.
Preemption
- Lead with the venture architecture before leading with AI. Dana is not an AI company that happens to have a portfolio structure — it is an agent-native venture architecture built from day one. The sequence matters.
- "AI-native vs. AI-adopted" is the single best preventive tool. Use it early.
- Foreground the portfolio — FullFare, Iris, Nodus — so AI-nativity is demonstrated across multiple ventures, not asserted as a single-product claim.
The philosophical framework is sophisticated, but a reader who came for product context finds no hook. The brand reads as conceptual without operational evidence.
Preemption
- Pair every philosophical claim with an operational example. "Context capital compounds" should be followed by a Nodus example showing how context accumulated across runs creates organizational advantage.
- Evidence before argument, not after. Include a "What Dana has built" section before the principles section.
- The audit trail as an aesthetic principle (A2UI 03) is a good operational anchor: it shows the governance philosophy has been implemented, not just proclaimed.
Precision and architectural confidence — combined with the fact that agents contribute to brand materials — can read as cold, procedural, or ironically robotic. The voice guide bans "passionate" and "excited" without replacing them with genuine warmth.
Preemption
- The dana etymology (wise, not merely intelligent) is a humanizing choice that earns warmth by being substantively true, not performatively warm.
- Include worked examples that show human governance in action — agents working under human intent, not autonomously. The operator role is where humanity lives in the architecture. Make it visible.
- Conviction is a form of humanity. At the principles register, conviction replaces warmth — and is more honest.
A reader with knowledge of Persian history or the Achaemenid empire evaluates the Apadana reference and finds it shallow: a foreign aesthetic borrowed for credibility. This is the most fragile element of the brand.
Preemption
- Accuracy is the only defense. The historical grounding in this playbook is precise — the 72 columns, the governance function, the Achaemenid synthesis of regional craft traditions. That precision is the argument for authenticity.
- Make the intellectual lineage visible: why the governance analogy holds, not just that it holds. The synthesis of regional craft traditions by Darius I maps directly to the portfolio structure.
- Avoid "inspired by." Say what the architecture is. The venture architecture bears the name. The specificity is the authenticity.
A reader from a traditional brand context — a CMO, a design director, an agency — opens the playbook expecting an asset library and finds a philosophical framework. They conclude Dana doesn't have a complete brand system yet.
Preemption
- Visual specifications are clearly present and easy to navigate to. The reader who wants the Pantone number finds it without reading 40 pages of brand narrative first.
- Frame the playbook explicitly: "This is not an asset library. It is both a philosophy document and a technical specification. They are not in tension — the philosophy generates the specification."
- v1 (DAN-36) is the visual foundation. v2 is the depth layer. The document architecture makes that clear — v2 adds, it doesn't replace.
Research brief prepared by UX Researcher & Journey Mapper, 2026-04-05. Source task: DAN-61. Incorporated into playbook by Brand & Portfolio Lead.