A career framework,
applied to travel
A personality quiz result is easy to make and easy to forget. If the output was a one-word label, the product would be played once and closed, and a newcomer to Australia would be no closer to knowing where to go. The result had to earn its usefulness or the whole thing was a horoscope.
The brief: design a quiz that helps newcomers to Australia, including those in Melbourne and Sydney, identify their travel style, using Holland's personality model as the framework.
Two things in that brief made it harder than a normal quiz.
What the research
changed
Five people, who potential to be a future Australia traveller, are interviewed and pattern of a expectation from result are found from them.
Moving to the Australia aesthetic recognisation, Wildlife and a natural scene were named as Australia's signature in every interview, but a wildlife recognisation stopped at two animals.
I do not know, maybe a big spider.
Prospective Australia traveller, asked to name wildlife beyond koalas and kangaroos
Wildlife as identity,
scene as backdrop
This quiz's result had to represent the person, not label them.
Decision one: split the two symbols of Australia by job.The interviews named two things as Australian identity: wildlife and natural scenes. Scenery became the ambient backdrop of the quiz, while a native animal including Platypus, Brolga, Kangaroo, Wombat, Southern Right Whale, or Dingo, carried each personality type's identity on the result page, where the personal weight had to live.
Decision two: every pairing had to hold under two lenses.Each animal was matched to its type through natural behaviour and meaning in Aboriginal belief, and kept only if it held under both.
Decision three: AI surfaced candidates, sources decided.Research was AI-assisted, but every symbolism claim was cross-checked against Indigenous-affiliated, institutional, and academic sources before inclusion.
Beyond
the animal
Every piece on the result page was curated for the type, not filled. Take the Curious Explorer (Platypus): three pieces of content sit alongside the animal, each shaped to the type rather than dropped in.
Two render strategies
for two moments
The project design under the tech stack, including Next.JS, Typescript, Prisma, TailswindCSS. The quiz and the result page do different jobs, so they got different render strategies, which resonace with an assorted techstack.
Shipped and
learned
The overall quiz performance is good both in visual and media, but on the results the wildlife-and-destination pairing was the strongest signal. However, tips and trivia got less attention, likely because they were text-only.
That was where I wanna go!
User who initially doubted her personality match, on reading her recommended destinations
Reflection
- 01The differentiator lived in curation, not quiz mechanics. Personality labels alone weren't enough; destinations, tips, and trivia had to answer what travellers actually asked for in research.
- 02Wildlife carried identity; scenes carried atmosphere. Splitting those jobs kept the quiz immersive without making the result feel abstract.
- 03Cultural symbolism required traceable sources and honest scope limits. AI helped surface candidates, but shipping claims needed Indigenous-affiliated, institutional, or academic backing.
- 04Two render strategies matched two moments. Preloading for the quiz; server-rendered results for the payoff page.
- 05A fuller version needs more than one internship. Scaling the matching responsibly wants a cultural advisor co-authoring from the outset, and the premade quiz questions some users found inconsistent need a revision pass.
