Concept Car, Manifesto, Trojan Pony: Three Ways to Build a North Star
Ben Shown is head of design at Blink UX, a studio whose work sits inside Microsoft Office, Google NotebookLM, the New York Times, Amazon Alexa and NASA. His Nivan Live session began with an engagement ring and a diamond marketing campaign, then delivered the most immediately applicable framework of the future portal: three distinct ways to build a North Star, and three reasons they fail.
Ben Shown at Nivan 2026
Two months salary
Ben opened by asking the room how much should be spent on an engagement ring. The answer arrived before the slide did. Before the nineteen thirties diamond engagement rings were rare, and a company devastated by the Depression tied diamonds to commitment and to a fraction of annual salary. One month became two and eventually three. No frame of reference existed, so one was manufactured, and an entire market oriented around it.
What a North Star actually is
That is the function of a North Star. It creates a frame of reference for a team and a company, and it challenges them to think bigger than the next release. It is not a roadmap or a specification and it does not answer how something works. You are not supposed to reach it. Like a ring, it is a symbol rather than the thing itself, and it points at who you are becoming rather than who you are.
What teams are actually asking for
Ben reported what he hears from almost everyone in his orbit. Help us make decisions with confidence. Help us stop reacting to market noise. Tell us whether we are moving towards something or merely moving. Underneath sits short term pressure drowning out long term thinking, and the real request is for something to move towards instead of something to react to. These are teams whose roadmaps have roadmaps. They do not need another plan, they need orientation.
Why this got easier in the last year
Ben named a genuine change rather than a general one. AI has lowered the barrier to North Star work specifically. Ideas can be tested faster, prototypes reach higher fidelity sooner, and thinking can be stress tested cheaply. Work that once required a large budget and a long window is now within reach of teams that could not previously attempt it.
The concept car
The first archetype came from childhood auto shows and the GM Impact, a fully electric concept car shown in 1990, decades before the current market existed. Before it, electric vehicles were barely on the company's radar. After it, they could not be ignored, and it led to the first mass produced electric vehicle. A team makes a large strategic bet as a high fidelity prototype or product film. Compelling parts migrate into the development lifecycle. The North Star itself never ships.
Two concept cars in practice
For the New York Times in 2022, a flagship application was buckling under an expanding offering of sections, podcasts, video, games and cooking, most of it buried in menus and lists. The Flex Model gave readers dynamic panels a swipe away, rallied leadership, the newsroom, product and engineering, and became the largest app redesign in sixteen years when it shipped in 2024. For Nextdoor, a platform meant to build community had drifted towards a complaint engine, and the concept car was a ground up redesign presented as a physical exhibit in their offices so anyone at the company could walk in and kick the tyres.
When to use it
The concept car suits teams trying to change the game and organisations with hard silos that need rallying. Two questions determine feasibility. Do you have permission to think two to three years out, a window that used to be five and has narrowed. And do you have roughly three months to assemble a team to do audacious work that remains grounded in user needs.
The manifesto
The second archetype is lighter and reachable by product leaders without the leverage to orchestrate something large. It designs the future through persuasive language rather than prototypes. A good manifesto distils a product into an elevator pitch and answers why we exist, tying actions and goals to that purpose. Ben was emphatic that this is not a corporate poster from leadership, which he considers useless.
Two manifestos in practice
Amazon's working backwards press release is his favourite example. You write a release from a future in which the product has launched and the problem is solved, starting from customer need rather than from what you are building. It is lightweight, honest and uncompromising, and it becomes a filter for design decisions afterwards. The second is a seller mindset diagram built for eBay, where research established the five things every seller wakes up thinking about, and made visible why the outer ring of ambitions stays out of reach until the inner ring is solved.
Why manifestos matter more now
Ben made a point specific to this moment. A manifesto is built for people to absorb and repeat, and it has gained a second function as a primary source for grounding AI agents and generative workflows. Asking a model to align outputs with your stated North Star, or to review your work against it, only works if the document exists in a form precise enough to reason from.
The Trojan Pony
The third archetype, and his favourite, is a focused North Star aimed at a single user journey or high traffic feature, chosen so that its innovations spread outward through the product. He renamed the Greek original as part of a personal effort to reduce military metaphors. The premise is that you cannot boil the ocean to design the future, but you can go extremely deep on something specific. For Stibo Systems, a fleet of dated enterprise data management products, one product and a few key moments were redesigned into the Instrument UI, which then travelled to other teams and raised the appetite for user experience across the company.
Why North Stars fail, and what comes next
Three failure modes recur. They are built for boardroom audiences only, when frontline change agents should be stakeholders. They are divorced from user reality, created inside a design silo rather than driven by real user needs. And they miss the ambition sweet spot, landing either too futuristic or too close to a modest evolution of what already exists. When one succeeds, Ben advocates back casting rather than forecasting: begin from the achieved future and work backwards to the decisions required to reach it, sharing it at the right altitude inside the organisation before analysing which concepts pull teams forward. He closed on the ring he actually bought, which made an abstract promise feel real and worth fighting for.
