Case Study — Sustify · Designing for factory workers in Asia

§ 01 — The premise

Making training accessible.

Sustify is an interactive learning app that teaches factory workers about workplace safety and human rights.

Sustify takes a bottom-up approach, bringing training directly to factory workers. The training runs on tablets in the factory and is designed to work across different literacy levels, using audio and culturally relevant illustrations alongside written content.

My work focused on how the learning content could be translated into simple, understandable interactions. A large part of that was finding the right visual support for the content, while considering different literacy levels and cultural contexts.

§ 02 — Designing beyond text

Designing beyond written language.

The training was designed for factory workers in India and China, with different literacy levels and cultural backgrounds. That meant I couldn't rely on written language alone — but replacing text with visuals wasn't automatically the answer either.

When text isn't enough

Important information needed to remain understandable without relying on reading alone.

When visuals need context

Icons and illustrations can be interpreted differently across cultures, so visual choices needed to be considered just as carefully as written ones.

§ 03 — Visual cues

From text to visual cues.

One fire safety module uses Raul, a fictional factory worker, to explain what a fire needs to start. Raul needs heat, oxygen and food to survive; a fire needs heat, oxygen and fuel to keep burning.

A frame from the Learn About Fire video: an illustrated worker, Raul, with arrows pointing to a sun, an O2 symbol, and a plate of vegetables.

Raul needs heat, oxygen, food.

A frame from the Learn About Fire video: a large flame on the left, with three arrows pointing to what it needs — a plug and cable throwing sparks, an O2 symbol, and a cardboard box.

So does fire.

Fig. 01 — The analogy carries the lesson, before the exercise begins.

In the exercise, learners drag a photo from everyday factory life into one of two categories: is this a source of heat that can ignite a fire, or something that can feed a fire further? Both categories were labelled with text alone. For workers with limited reading skills, that made an otherwise simple task difficult to understand.

The earlier sorting screen: the question at the top, a photo of a storage shelf in the middle, and two categories labelled only with the words HEAT and FOOD.

Fig. 02 — The earlier version: the two answer categories carried by text alone.

My task was to find visual cues that could communicate the same meaning without relying on text alone. The two categories turned out to be very different problems.

Straightforward

Feeds the fire

A flame was a simple and recognisable visual for this category.

Harder

Source of heat

A cable was technically related to the hazard, but recognising a cable didn't necessarily mean understanding “heat”. It still required the learner to interpret what the symbol was meant to represent.

Instead of introducing another visual abstraction, I reused the illustration from Raul's story. Its meaning had already been established in the lesson, so the exercise could build on recognition rather than asking learners to interpret a new symbol.

Analogy

Raul's story shows what a fire needs to start and keep burning.

Exercise

The same illustrations return as the answer categories.

Learner

Recognises the visual instead of interpreting a new symbol.

The redesigned sorting screen. The question reads: what is a source of heat to ignite a fire and what can feed a fire further? Below it, a photo of an overfilled storage shelf, with two answer categories either side — on the left, an illustration of a plug and cable being dragged along the floor throwing sparks; on the right, a flame.

Fig. 03 — Reusing familiar visuals from the lesson in the exercise.

Icons are rarely universal. If a visual has to work without words, the understanding has to be established before the user meets it.

What the fire safety module taught me

§ 04 — Recognition and context

When recognition depends on context.

In the first-aid exercise, learners need to recognise two possible sources of help: an ambulance and a first-aid officer. Unlike the fire symbol in the previous example, neither of these looks the same across cultural contexts.

The first-aid exercise: the question asks when to call the ambulance and when to contact the first aid officer. An illustrated ambulance sits on the left, a first aid officer on the right, and the case to sort in the middle.

Fig. 04 — The first-aid exercise: the same two-category structure, with the two sources of help as the answers.

So the question wasn't just how to draw an understandable icon. It was which visual details make something recognisable locally — and which parts of the illustration can stay the same everywhere.

Reference photograph of an Indian ambulance: a white van with an orange stripe and a blue star of life on the side.

Fig. 05 — Checking the illustrations against how these things actually appear on site. Stock photograph used as reference.

For the ambulance, I looked at which identifying details had to reflect the local context — the blue Star of Life on the side is one of them — and which parts of the illustration could stay as they were. The aim was to make the visual recognisable locally without redrawing the whole thing for every market.

§ 05 — Abstract or diversify

But what about the people?

Objects can borrow meaning from context. Human characters can't — every face carries features, clothing and skin tone that someone will or won't recognise as their own. There are two ways to handle that, and they pull in opposite directions.

Strategy A

Abstracting

Move away from realistic detail, so people can fit their own identity into the image. The first-aid officer takes this route: deliberately grey, with no skin tone that could be read as one ethnicity — human-like, but nobody in particular.

The first-aid officer icon: a grey human figure in a white coat with a red cross, holding a red first-aid case.

Strategy B

Diversifying

Show the range of differences people might identify with. The avatar set takes this route: differences between cultures, and differences within a single culture. It costs more work per market, but people see themselves rather than a stand-in.

Twelve illustrated avatars — an Indian group in the top row and a Chinese group in the bottom row, each in different traditional and everyday clothing.

Neither is the right answer everywhere. Abstracting scales — one figure works in every market. Diversifying recognises — but multiplies the work each time a new market is added. For Sustify, the deciding factor was scale with low effort for change, so we went with abstracting (the avatar set itself was designed by someone else on the team).

§ 06 — Learnings

What I took from this.

To be written

Candidates, based on what's still in the case study: whether abstracting would have scaled better than diversifying as more markets were added; and what borrowing meaning from context — rather than searching for a self-explanatory icon — changed about how I approach visual accessibility.

How I learned about the users My understanding of the users came primarily through the founder, who had visited the factories, knew the working conditions first-hand and tested the training on site. I combined that knowledge with the client's requirements and my own research into accessible and cross-cultural design.

Designing for a
different audience.

Making training accessible.