Journey North

Journey North was Cognito Foundation’s response to a piece of the Northern Economic Plan: give high school students in Northern Adelaide a way into entrepreneurship. We set out to build an online tool that mapped the South Australian ecosystem (training programs, scholarships, workspaces, microfinance, community groups, incubators) and pointed each student at the next opportunity that actually fit where they were.

The ambition underneath it was a culture change, not a directory. Develop a new wave of young entrepreneurs at scale, starting in the northern suburbs, so that eventually anyone who wanted a shot at it got one. That is where the name and the line we built the brand around came from: Journey North, guiding every kind of entrepreneur. “Every kind” was the part that mattered, and it set the bar for everything downstream.

The pieces already existed. TAFE programs, university initiatives, libraries, microfinance, accelerators, community groups, government schemes. The problem was that none of it was discoverable from inside a high school. A student leaving school had no single view of what was out there, no shared language for the stage they were at, and no obvious entry point into any of it.

The half of the problem that was easy to miss sat on the other side. The providers were not indifferent. Across local government, training, finance, and community sectors there was a group that actively wanted to reach young people and had no mechanism for doing it. Students bumped into one or two providers by accident, the long tail went unseen, and a willing multi-sector community stayed unconnected to the people it was set up to serve. What was missing was not supply or goodwill, it was connective tissue.

On top of that, the audience was teenagers. Anything that read like a careers portal or a government services list was going to lose them before the first scroll.

The first body of work was service design, not screen design. As Cognito Foundation, we worked with local schools, businesses, and students to inventory every organisation, program, and resource in the region, then categorised them by what they offered and by who they were for.

Roughly fifty organisations went onto the map across seven categories: local training and programs, online resources, business support, physical assets and co-working spaces, formal education, funding programs, and community experience. The range is the point. Khan Academy and LaunchX sat alongside the Elizabeth and Munno Para public libraries, Good Shepherd Microfinance, TAFE SA, Polaris, and the Rotary clubs of Playford and Salisbury. No student was ever going to assemble that list on their own, and no single provider had a reason to.

That work produced two artefacts the rest of the project leaned on. A timeline view that placed providers across the arc from latent entrepreneur to established entrepreneur, and a categorised ecosystem map that grouped them by service type. Both became reference material for the working group as much as inputs into the product, and the timeline in particular exposed where the ecosystem was thin: plenty of support once someone already had a business, far less aimed at a 15-year-old who had not started one.

The progression model is the piece of this project that still holds up, and the reason is that it was two things at once.

Underneath, it used the vocabulary the sector already had. The ecosystem timeline placed every provider across five recognised positions on an entrepreneurial pathway: latent, potential, nascent, new, established. That language let the working group argue about coverage precisely, and it let providers say which part of the arc they served without us inventing a taxonomy they would have to learn.

On the surface, students saw Cub, Apprentice, Grunt, Boss, Chief, each with an avatar. Deliberately informal, so a 15-year-old would recognise the journey instead of bouncing off it.

Same five positions, two audiences, and the platform did the translating. That mapping is what let one model serve a student choosing a next step, a provider tagging an opportunity, and a working group reasoning about gaps in the region, without any of the three having to speak the other’s language.

Two-sided product. Students needed surfacing, but providers needed a way to keep their listings honest. The platform was scoped as two tightly linked surfaces: a student experience that matched people to opportunities, and a provider tool that let a non-technical staff member at a library or training provider add and manage their own opportunities without help. Given that the ecosystem’s problem was willingness without a mechanism, the provider side was not a back office, it was half the product.

Tagging that survived the long tail. Every opportunity carried two layers of metadata: opportunity type (physical, online, support services, training, community, funding) and entrepreneur stage. That structure let the matching logic stay simple while keeping the long tail of niche providers reachable.

Matching first, search as the escape hatch. The default path was a matched feed, because a student who does not know what exists cannot search for it. But a keyword and opportunity-type search sat across the top of every signed-in screen, for the student who arrived knowing exactly what they wanted.

Workstreams ran in parallel rather than sequentially.

Student experience. Three steps: create a profile, work through the journey questionnaire, then land on a dashboard of matched opportunities. The questionnaire was the engine, a paginated set of questions whose answers placed the student on the progression model rather than asking them to self-select a stage. The dashboard carried a persistent progress rail showing where they sat, and each matched opportunity showed its type, the stages it suited, and a “Get in touch” path straight back to the provider.

The account kept their profile, their business if they had started one, and their completed questionnaire, so a student’s stage was something they could revisit as they progressed rather than a label fixed at signup.

Provider tools. Organisation profile, opportunity list, and add-opportunity flows designed for staff at libraries, training providers, and community groups. The forms were short, the tagging was opinionated, and the empty states pushed providers toward filling in the things that actually drove matching.

Prototyping and testing. Low and high-fidelity prototypes moved back to the Cognito working group at each stage. Concept testing with students focused on whether the progression model made sense, whether the surfaced opportunities felt relevant, and whether the language and avatars landed for the age group. Wireframes and process maps were approved by the working group, and the high-fidelity prototypes tested well with students before handover.

MVP build. Worked closely with the development team to bring the platform to life, making sure the progression model, the matching logic, and the provider tooling survived the build intact.

The wireframes, process maps, ecosystem model, and progression framework were approved by the Cognito working group. The high-fidelity prototypes tested successfully with students. The MVP was built. Before the platform reached public release, the Northern Economic Plan was defunded and the project was wound down. The artefacts (the ecosystem map, the progression model, the platform design) stayed behind as a clear vision for how entrepreneurship support could be made legible to young people in the region.

The useful lesson from this one was how much weight tone carries on educational products. The ecosystem and the opportunities were already there. What was missing was a way to talk about progression that a teenager would not bounce off. Cub through Chief did more work than any feature in the platform, because it gave students, schools, and providers a shared way to talk about where someone was and what to do next.

The version of that lesson I still use is narrower. The informal names worked because they were a translation of a model the sector already trusted, not a replacement for it. Renaming the layer a user touches is cheap and safe when the structure underneath keeps its own vocabulary. Inventing a new structure so the friendly names have somewhere to live is neither.

The other lesson was about working in service of a plan you do not control. The project did serious work and shipped real artefacts, then ended because the funding around it ended. That is part of the texture of public-good initiatives. The best you can do is make the thinking durable enough that the next attempt can pick it up.