INFO 7375 — Branding and AI at Northeastern University — did not produce a slideshow about brand strategy. It produced a brand strategy system.
Nina is live. You can use her now: Nina Brand Identity System →
Twenty-six commands. A complete pipeline from intake to style guide. An AI that introduces itself before you say a word. A readiness score with a hard threshold. A jargon audit that deletes “omnichannel” on sight and asks you to name the actual channels.
We built this today. In class. Together. That fact matters more than what the system can do — because what it can do depends entirely on the fact that it now exists.
The system is called Nina. It is named for a persona, not a person — but the persona was built to carry the weight of someone real.
That person is Nina Harris: Brand Director and Creative Director with over twenty-five years of practice across Charles Schwab, Publicis, McCann-Erickson, and Saatchi & Saatchi. At Schwab alone, she led a creative team of twenty, oversaw brand identity systems across every business line, and directed the production of more than ten thousand proprietary photographs. She co-teaches INFO 7375 with me at Northeastern. She sits on the board of Humanitarians AI. She is not a persona. She is the standard the persona was built to approximate.
The AI version of Nina enforces a structural argument before she entertains any deliverables: discovery before strategy, strategy before identity, identity before build. The sequence is the product. Violate it and you get a beautiful logo for a brand nobody understands.
What we built today is a first pass. The real Nina Harris will continue to add her expertise as we refine it — her judgment about what actually works in the room with a skeptical CMO, what a creative team actually needs to hear before they open a design tool, what twenty-five years of brand failures and successes actually teaches you about the order of operations. The AI encodes the structure. Nina provides the truth that makes the structure worth following.
What the class built is not a prompt. It is an architecture — a sequencing constraint imposed on a model that would otherwise skip the sequence entirely if you let it. The intake exists because you need to be stopped from treating it as optional. The archetype shadow exists because without it you will identify the Hero and never notice you are building the Bully. The single hard no in every creative brief exists because a brand with no hard no has no values, only positioning.
None of this was guaranteed at the start of the session. It required decisions: what to include, what to retire, where to put the threshold, which failure modes to name. Those decisions are documented in the system. The system is the receipt for the thinking.
The thinking happened today.
What follows is the full architecture of what we built, documented for anyone who wants to use it — or build something better.
The Architecture of Belief: What Happens When a Brand Strategist Becomes a System
You open a new conversation. A message appears before you’ve typed a word.
Hello. I’m Nina.
I’ve spent 25 years making brands matter to the people who need to find them.
You pause. Something is off — in the best way. You expected a prompt asking what you need. You got, instead, a person telling you who she is. The AI spoke first. It didn’t wait. And it didn’t say “Great!”
This is the first design decision in the Nina Brand Identity Prompt Set, a complete command library for building brand identities from intake to style guide. It’s a system built on a single structural argument that most AI tools refuse to make: the order in which you do things determines whether they work. Discovery before strategy. Strategy before identity. Identity before build. Not because a consultant said so. Because a brand that designs its logo before it understands its audience has built a beautiful answer to a question it never asked.
The architecture is rigid by design. Resist it and you build noise. Follow it and you build belief.
The Problem That Preceded the Tool
Consider what most AI-assisted brand work actually produces. You type “help me create a brand identity for my consulting firm.” The model generates a name, a color palette, a tagline, a mission statement, and perhaps a logo description — all within ninety seconds, all before it has asked you a single question about who your clients are or what you do that your competitors don’t.
The output looks complete. It has all the pieces. What it doesn’t have is coherence.
This is not a model failure. It’s a prompting failure, and more fundamentally, a thinking failure. The model gave you what you asked for. You asked for deliverables. You should have asked for a foundation.
Nina’s twenty-six commands exist because the deliverables are not the point. The deliverables are the proof that the thinking was sound. Every section of the prompt set is structured around this principle: nothing ships until the reasoning that justifies it exists.
The intake command, /n1, is not asking you for information to populate a template. It is asking you eight questions in sequence, waiting for your answer each time, refusing to proceed to the next phase until it has built a summary that you confirm is accurate. “The brand is...” / “The tension to resolve is...” / “The opportunity is...” Those three lines, once confirmed, become the load-bearing wall for everything that follows. Change them and the whole structure shifts.
This is unusual. Most tools flatten the process. Nina verticalizes it. You can feel the hierarchy when you use it — not as bureaucracy, but as insistence that your thinking be precise before it becomes expensive.
The Jungian Engine Underneath the Tagline
Slide past the intake and you reach the archetype command, /n2, and here the framework reveals its theoretical stakes.
The twelve archetypes — Innocent, Everyman, Hero, Outlaw, Explorer, Creator, Ruler, Magician, Lover, Caregiver, Jester, Sage — are Jung’s original taxonomy, imported wholesale into brand strategy in the late 1990s by Carol S. Pearson and Margaret Mark in The Hero and the Outlaw. The framework argued that brands resonate at the same psychological frequency as myths. An Outlaw brand doesn’t just sell nonconformity; it activates the psychic pattern of the rule-breaker, the liberator, the one who names what everyone else is afraid to say.
Nike is the Hero. Dove is the Innocent. Harley-Davidson is the Outlaw. The framework works well enough that entire brand consultancies have been built on it. It also fails in a specific, predictable way: practitioners identify an archetype and stop there. They say “we’re a Sage brand” and then write copy in the Sage register and design a logo in muted academic tones and believe they’ve done the work.
The Nina implementation does something different. It demands three outputs that most archetype exercises skip.
First: the shadow risk. For every archetype, there is a failure mode — the thing the archetype becomes when it overreaches. The Hero becomes the Bully. The Sage becomes the Pedant. The Caregiver becomes the Martyr. Name the shadow and you name the creative guardrail. You know what to avoid not because someone told you to be careful but because you understand the mechanism of failure.
Second: the secondary archetype. Single-archetype brands are either genuinely rare or simply underdeveloped. Most brands that resonate hold productive tension between two archetypes. Apple is both the Outlaw and the Creator — the rebel who makes beautiful things. Patagonia is both the Explorer and the Caregiver — the adventurer who feels responsible for the terrain. The secondary archetype doesn’t dilute the primary; it creates the texture that makes a brand feel like a person rather than a persona.
Third: the archetype brief, formatted precisely:
“[Brand] is a [Primary] with a [Secondary] edge. It believes [core belief]. It speaks to people who [audience truth]. It will never [hard no].”
The hard no is the most important clause. A brand’s “will never” is more revealing than its mission statement. What you refuse to do tells you what you actually believe. A brand that “will never speak down to its audience” has made a commitment. A brand that “will never sacrifice performance for aesthetics” has named a value hierarchy. A brand with no hard no has no values — only positioning.
The Single-Minded Problem
The creative brief command, /n4, runs to seven sections, but the test that matters is at the end.
Can you reduce the entire brief to one sentence a creative team could carry in their head all day?
David Abbott, the British copywriter, used to say that a brief should fit on a matchbox. Not because complexity is bad but because complexity is how agencies hide from decisions. The brief expands to accommodate every stakeholder’s priority until it contains everything and means nothing.
The Nina brief borrows from the Single-Minded Proposition tradition — the discipline, codified at agencies like DDB and Ogilvy, of forcing a brief down to a single, unchallengeable claim. The UVP command, /n5, scores any proposition on four dimensions: focus, clarity, distinctiveness, and inspiration. But distinctiveness is the operative test:
Would this sentence be false if a competitor said it?
If the answer is no — if “we help businesses grow” or “we put customers first” passes through your lips — you don’t have a UVP. You have a sentiment. Sentiments are not differentiators. Differentiators are claims only you can make because only you have the evidence, the history, the capability, or the courage to make them.
The /n5 command generates three alternative framings of every UVP: functional, emotional, and provocative. Then it makes a recommendation. For THIS audience, I’d use the [X] version because... It doesn’t offer all three as equally valid and ask you to choose. It makes a call. This is rarer than it sounds. Most AI writing tools are professionally noncommittal — they generate options and attribute the decision to you. Nina’s architecture is built to force the opinionated move, then explain it.
Voice as Philosophy
The voice guide command, /n6, contains one requirement that repays more attention than it initially gets.
The IS / IS NOT table must be constructed so that each “IS NOT” is the corruption or overreach of its paired “IS.” Not an unrelated trait. The failure mode of the virtue itself.
Direct is NOT Blunt. Not “boring,” not “rude,” not “unpolished.” Blunt is what Direct becomes when it loses its care for the listener. The distinction is precise enough to be actionable: a Direct brand chooses honest words. A Blunt brand chooses honest words and doesn’t consider their landing.
This structure — archetype, shadow, IS/IS NOT corruption — is the same move repeated at different levels of the framework. The system is teaching you to think in failure modes because brands fail the same way people fail: by taking a virtue too far, or by mistaking its corruption for its expression.
The framework also contains a list of words to retire. Not banned for vagueness but banned for saturation: words so overused in a given category that they’ve lost semantic content. “Authentic” has been used to sell everything from fast food to pharmaceutical companies to banking apps. At this point “authentic” means nothing except “we want you to believe us.” It is a signal of insecurity, not honesty.
The Nina jargon audit, /jargon, extends this logic to the full document. Every flagged term gets a rating — red, yellow, green — and a translation. “Brand archetype” becomes “brand personality.” “Touchpoints” becomes “every place a customer encounters the brand.” “Omnichannel” gets deleted entirely, with an instruction to name the actual channels. The translation guide is not just semantic cleanup. It is a test of whether you understood what you were saying in the first place.
The Readiness Score
The /ready command scores any deliverable on five dimensions, twenty points each, with a threshold: eighty points to ship.
The scoring categories are standard enough — strategic clarity, distinctiveness, audience fit, internal consistency, execution readiness. What distinguishes the implementation is what the system deducts points for.
Under audience fit, you lose points for “demographic generalizations” and for “failure to name a specific human truth the audience holds.” Not for missing a target age range. For failing to name something true about how a person sees the world. The framework cares about psychology, not census data.
Under execution readiness, you lose points for “vague photography direction” and for “tone descriptions that require interpretation.” A style guide that says “warm and approachable” has not guided anything. A style guide that names the three scenes a photographer could shoot — specific moments in real-world locations with specific lighting — has given someone something to do.
The final deliverable from a /ready run is exactly one priority fix: “Before this goes to a client, change [X].” Not a list of improvements. Not a score with a note that “several areas could be strengthened.” One thing. The most important thing. The structure refuses to let you distribute responsibility across a comma-separated list of concerns.
What Nina Is Not
The system is comprehensive, but comprehensiveness is not the argument for it.
The argument is that most AI-assisted creative work suffers from a sequencing problem disguised as a capability problem. People believe the model isn’t good enough to write great brand copy. Often the model is entirely capable. The problem is that it was asked to write great brand copy before anyone answered the question of who the brand is for and what it is trying to make that person believe.
Nina is a sequencing constraint imposed on an AI system that would otherwise skip the sequence entirely if you let it.
The intake exists not because Claude needs to be told what a brand intake is, but because you need to be stopped from treating the intake as optional. The archetype shadow exists not because the shadow is exotic strategic content, but because without it you will identify the Hero and never notice you’re building the Bully.
The system’s real product is not the style guide at the end. It is the quality of thinking that made the style guide possible. The guide is the receipt. The thinking is the purchase.
The Manifesto Clause
Every brand framework eventually reaches for the manifesto — the document that precedes all execution, the belief statement written as if no one has ever said this before.
The Nina prompt for /manifesto contains one test that functions as its thesis:
If someone read this and thought “that’s not for me,” that’s correct. A manifesto that offends no one believes in nothing.
This is the structural argument for specificity at every level of the framework. The brand that refuses to specify its audience keeps trying to convert the Tertiary persona — the person it was never built for. The UVP that won’t name what it won’t do is afraid to lose the wrong customer. The voice guide that refuses to retire any words is afraid to sound like something.
Fear of exclusion is the root cause of most brand incoherence. Nina’s architecture is designed to make exclusion feel like strategy rather than failure. You are not losing the customer who doesn’t fit. You are finding the one who does.
That is the belief the system is built on. The twenty-six commands are the machinery. But the belief is what makes any of it work.
The Nina Brand Identity Prompt Set is available as a full command library for building brand identities from intake to style guide. It is designed to be used in sequence.
This was such a fun class. Such a pleasure learning from you Professor!