Creative studio · Original interactive experiences
Don’t tell them. Let them find out.
We design and build original interactive experiences that help people see not just how a complex innovation works, but why it matters. We’re starting with science and technology.
Complex innovations tend to get stuck in one of two places.
Stuck at how
They can’t picture it.
The mechanism lives in a lab, a model or a line of code, and nothing in your audience’s experience looks like it. So they nod politely and move on.
Stuck at why
They follow it perfectly, and still don’t see why it matters.
A 30% gain that sounds like a rounding error. A new method that sounds like the old one with extra steps. A future that sounds like one more slide. This gap is hard to see from the inside, because to your team the value is obvious.
An idea can be brilliant. That doesn’t make its value obvious to everyone else. We work on both, and it usually becomes urgent at a particular moment: a launch, a new website, a raise, a conference, a change of direction.
02Prototypes
Our own experiments: original pieces on ideas we think deserve a better way in. Each one is built the way we’d build yours. Prototype_001 takes an advantage that’s easy to state and hard to feel, and lets you find it out for yourself.
Prototype_001
The Loop
A study in learning from reality.
Billions of possible enzymes. Enough lab time to test a few hundred. Predict which of three strategies will find the best one, then run all three and watch what happens when the results come back. (predict first.)
Field
Enzyme engineering · machine learning
Format
Playable, in the browser
Time
About six minutes
Best on
A laptop (works on a phone)
Inside
A real enzyme in 3D, from its published crystal structure
A lab tray that becomes a map of every possibility
Your prediction, then three strategies on the same budget
A question that checks whether the idea carried over
A sample report of what a real audience might reveal
From Prototype_001 · One tray of 96 testsYellow means a more active enzyme, as in the real lab test. Values from the prototype’s own simulation.
Independent THING3 prototype, built from public research. The simulation and the report’s audience figures are illustrative; your own answers are real and stay in your browser.
Not an explainer video with buttons. Each piece gives people something to do (predict, compare, adjust, decide) and explains at the moment they’re ready for it. Explanation still matters. It just lands better once someone has found something out.
Three things we haven’t made yet
Illustrations of where an idea could go. Every project starts from the idea, never from a format.
A mechanism no one can picture
An explanation you operate.
Set the process going yourself and see exactly where it parts ways with what came before.
An advantage that sounds small
A difference you can feel.
Try the usual way and the new way on the same problem, and watch a modest-sounding gain become the one that matters.
A future that’s hard to imagine
A vision you can step into.
Face the decisions your customers will face, with and without what you’re building.
They could live on your website, at a launch or on a conference stand, or in an investor’s inbox after the meeting.
What making one involves
Understanding
What has to land, for whom, and where it currently gets lost.
Concept
An original creative direction for your idea, with alternatives explored where it helps.
Design
The interaction, visuals and words, in your brand.
Production
Built for the screens it will run on (phone, laptop, the display on your stand).
Accuracy
Scientific and commercial claims agreed and reviewed with your experts, and simplifications stated plainly.
Refinement
We try it, watch it being tried, and improve it before it goes out.
Depending on the project Structured audience testing · built-in questions and an aggregate read-out · a review of your current material · versions for other formats or events · hosting and upkeep. Scoped when they’re useful, not by default.
Our rule If a clear sentence would do the job better, we’ll tell you.
04How we work
Creative direction with experimental discipline.
Imagination decides what the experience could be. Evidence decides whether it’s working.
Your idea
01
Find what has to land.
We learn the idea, the audience and where explanations currently lose people. Often it isn’t the mechanism. It’s why it matters.
02
Find the way in.
Two or three original concepts for this idea and this audience. Each one has to earn its interaction.
03
Make it.
Designed and built for the browser, with the science and the claims checked as we go.
04
See what lands.
We watch what people actually take away, then refine. For a handful of investors that might mean a few conversations; for thousands of customers, short questions built into the piece.
Psychology informs the design: what to show first, what to let people try, and what to leave out.
05Working together
Bring us the idea that keeps getting lost.
Every project starts with a conversation about the idea and who needs to get it. Then it goes one of three ways:
“We know what needs to land.”
Straight to concepts, then design, production and testing.
“Something isn’t landing, and we don’t know why.”
We start by finding where it breaks: we go through what you use now and talk to a few people who don’t get it yet. You get a short read-out and a recommendation, then decide what to make.
“We’d like to see the idea before we commit.”
A concept stage: two or three directions as sketches, storyboards or a rough playable fragment. Then you decide on the build.
Scope, timing and fees are agreed up front, for each project.
06The studio
Good ideas don’t need dressing up. They need the right light.
See what’s already there.
flick the switchnow change the angle
This figure needs WebGL 2. The rest of the site works without it.
40°88°
Show the numbers
Fig. 01Dispersion in SCHOTT N-SF11 glass
Glass bends shorter, bluer wavelengths more than longer, redder ones.
Lamp off: flick the switch to see the light
Incidence
63.2°
Deviation · 587.6 nm
—
Spread · 400–700 nm
—
Flick the lamp’s switch, then move across the image or use the slider; the scale on the prism reads the angle of incidence. The beam is broad-spectrum white light, modelled as 400–700 nm. A 2D ray model of a 60° prism using Snell’s law, Fresnel reflection and SCHOTT’s published N-SF11 glass data. 587.6 nm is the helium d line, the reference wavelength glass makers use. Not modelled: absorption, scattering, polarisation, the third dimension. Colours are approximate: screens can’t show pure spectral light.
Fig. 01 · A prism doesn’t add colour to light. It shows what was already there. We try to do the same for ideas.
Every angle real: Snell’s law and published glass data.
THING3 is a founder-led creative studio, started in 2026 by Chan, who brings together creative direction, psychology and hands-on building. Science and technology are where we’re starting, because that’s where important ideas are most often hard to see. The approach won’t stay there: we’re interested in any idea worth paying attention to, and in formats beyond the screen.
The name
Science is one thing. Art is another. THING3 is what happens when they talk.
The rule
Accuracy first. We simplify; we don’t exaggerate.
The stage
We’re new. Instead of a client list, there’s the work: start with Prototype_001.
07Contact
What do you wish people understood about your idea?
Tell us what it is and who needs to get it. (a few lines is plenty.) If you’d rather talk it through first, we’re happy to start with a 30-minute call. No pitch, no obligation.