A screen that shows information does not change behaviour. A message that is read, understood and acted on does. The difference between the two is not the hardware — it is what the message says, when it appears, and which part of the brain it is written for.
People in a crowd are not reading carefully
Someone walking through a busy site is not weighing options. They are tired, distracted, following the person in front of them, and processing almost everything automatically. Daniel Kahneman’s distinction is useful here: System 1 is fast, automatic and effortless; System 2 is slow and deliberate. Under crowd conditions you are writing for System 1, whether you intend to or not.
That has practical consequences. A sentence that needs a second reading has already failed. A sign that offers three options gets none of them chosen. An instruction that arrives after the decision point is decoration.
Tone decides whether people comply or resist
Eric Berne’s ego states explain something every steward already knows: the same instruction lands differently depending on the voice it is delivered in. Address a crowd as a stern parent and a predictable share of them will push back like children. Address them as capable adults and you get cooperation instead of friction.
This is not politeness for its own sake. In a safety situation, resistance costs time you do not have.
The moment matters as much as the wording
Behaviour is easiest to influence at the point where a choice is actually being made — a fork in a route, the exit of a stage area, the moment a crowd realises it must go somewhere. Certain triggers reliably switch attention on; most content ignores them and simply repeats a message everywhere.
One position at a genuine decision point does more than three along a route where nobody has a choice to make.
Why this is half the product, not the packaging
CrowdCows is not a screen supplier that happens to write text. The message library, the scenario builder and the software that sends messages to the channel are the product; the HIKER units and the broadcast signal are how it reaches people. Both halves are needed, and the psychology half is the one that is hard to copy.
Our content is built on priming, nudging and framing — a route that is not only correct but feels obvious, a warning that steadies people rather than alarming them. To the visitor it reads as good service. For a safety team it is precise direction.
What we will not do with it
We apply these techniques to safety behaviour only: moving people out faster, spreading density, keeping routes flowing, getting emergency instructions followed. We do not apply them to commercial behaviour. Steering a crowd towards a bar is a different product with a different ethical weight, and it would undermine the safety claim the moment anyone noticed the combination.
We also do not publish effectiveness percentages. Our measurement programme is still being built, and a figure that ends up in a client’s safety documentation has to survive a critical reader. Until it can, we would rather say nothing than say something convenient.
Frequently asked questions
Does this work on people who are not paying attention?
That is the point of designing for System 1. The message has to work without demanding attention first.
Who writes the messages?
We do, with you. Your safety scenarios become communication scenarios per position, agreed in advance rather than improvised on the night.
Can we use our own wording?
Yes, and we will tell you where we think it will not land, and why.
Is this manipulation?
It is the same choice every sign-maker faces: wording that works or wording that does not. We are explicit about the boundary — safety behaviour only, never commercial.
Read on: what crowd communication is · which scenarios you should have prepared for
