Counted on 2 November 2026, in the games' code as of 30 October 2026.
Where this comes from.
This work starts from one idea: people are not built to act emotionally on a dashboard. A line that falls two percent has to be read, compared and explained. A dying flower needs no legend. We understand it at a glance, and we want to do something about it.
Paul Slovic writes that numbers “fail to spark emotion or feeling and thus fail to motivate action”[1], and that images carry what numbers cannot. He is writing about loss of life on a large scale, not about invoices. We borrow the mechanism, not the scale.
In experiments by New, Cosmides and Tooby, people spotted changes to animals faster and more accurately than changes to plants, objects and even vehicles[2]. They read this as attention shaped by what mattered to our ancestors. A crocodile coming closer should therefore be harder to miss than a row in a table. We have not tested that.
A garden has also been tried before. The UbiFit system showed a person's physical activity on their phone as a garden of flowers. 28 people used it for three months, over the winter holidays. Those with the display kept up their activity, while the activity of those without it dropped significantly[3].
The question.
Administration fails quietly. An invoice slips past its due date, a deal goes a week without a follow-up, a compliance review lapses. In a list or a dashboard, each of these looks the same on the day it falls due as on the day it was created.
We wanted to know whether a different kind of interface changes that. In it, the things that are due come closer, and handling them visibly changes the place.
The question is narrow. If the same records are shown as a world that changes with them, do people handle what is due earlier, and miss fewer dates? We do not have the answer yet. This piece describes the experiment before it runs.
Why not points and badges.
The usual approach is gamification, which Deterding and colleagues define as “the use of game design elements in non-game contexts”[4]. In practice that often means points and badges added on top of a list that does not change.
A review of the empirical studies by Hamari, Koivisto and Sarsa found positive effects overall. It also found that those effects depend heavily on the setting and on the people using the system[5].
We took a different route. Nothing is added on top of the work: the records are the world. A deal is a flower, an invoice is a ship, a compliance deadline is a crocodile. Points exist, but they are incidental. A world turns green only when the real work is done.
Three worlds.
Each world reads one part of the platform and nothing else. Visitors without an account see invented sample companies.
Sales pipelineThe Garden
Each open deal is a flower on the terrace of its stage. A deal with no activity for seven days wilts. Won deals become trees; lost deals go to compost, with the reason they were lost.
- Weather
- Progress against the month's sales target
- Actions
- Follow up, move stage, mark won, mark lost
MoneyThe Harbor
Open invoices, bills, expenses, subscriptions, payroll and tax are ships sailing towards their due date. Money coming in and money going out sail in separate lanes.
- Tide
- Cash runway: how many months the cash lasts at the current monthly spend
- Actions
- Acknowledge, mark received, mark paid
ComplianceThe Swamp
Regulatory milestones, controls, incident reports, vendor reviews, evidence and policy reviews are crocodiles swimming towards the warden's dock. Their jaws open as the date nears.
- Fog
- The compliance posture: the higher the score, the clearer the water
- Actions
- Acknowledge, or resolve in the way each kind needs
One loop.
All three worlds share one loop. What is due comes closer or wilts, the player handles it, and the place recovers. Each world's health blends two parts: whether everything due has been handled, and one real business outcome.
The weights are the platform's own, copied one to one, and the tests fail if they drift. The second part is the important one: a world cannot reach green on tidiness alone. Acknowledging every invoice does not fill an empty bank account.
Something due within seven days is in danger; within thirty days it is on watch. The Harbor shows obligations up to 120 days ahead. The Swamp looks 540 days ahead, because regulatory deadlines are announced long in advance. Distance on screen follows a logarithmic curve, so next week takes up far more room than next year.
The rules we held it to.
- 01The game never changes a record.Every action the player takes is sent to the platform, which decides whether to record it through the same steps as its own buttons. The game holds no keys to any database.
- 02Unknown is not good.When the cash runway has not been measured, the Harbor shows it as unknown, never as endless, and an unknown value cannot lift the world's health. When no compliance posture has been measured, the Swamp shows its health as unknown.
- 03A record that leaves was not necessarily paid.When an acknowledged item drops out of the data, the game shows it as managed. It never celebrates it as paid.
- 04Every picture maps to data.Nothing invented is shown as if it were real. Sample companies in guest mode are made up, and nothing a guest does is saved.
- 05It has to run on ordinary hardware.The target is 60 frames a second on a 2020 laptop. Phones get a lighter version, and people who switch off motion in their system settings get no animation.
What we will measure.
We chose these measures before any customer uses the worlds, so we cannot pick them afterwards to suit the result. Each will be compared with the same team's records over the four weeks before they started.
Each number we publish will say how many records it counts and the day it was measured, as set out in How we measure.
What this does not prove.
Nothing on this page shows that the worlds work. We have built them and tested their rules. No customer has used them on real records yet.
A before-and-after comparison on a few teams cannot separate the interface from everything else that changes in four weeks. Novelty may lift the first weeks and then fade.
A picture can also hide a number as easily as it shows one. That is why reading accuracy sits next to speed in the list above. A world that gets things handled faster but leaves people unsure what they owe would be a failure.
The idea behind this piece is also less settled than a dying flower makes it look. The attention study found its effect for animals, and plants were among the things people noticed less. The research on feeling and numbers is about human lives. The garden study compared a display with no display, not flowers with charts. Whether a wilting flower moves someone more than a falling line is what we have to measure.
Where it stands.
All three worlds are playable on invented sample companies, in the browser, with no install. Open the demo.
The connection is specified and tested. A world runs inside the platform's page, receives the user's records and sends back each action.
Customers cannot use it yet. The first measurements come after one team has used a world on its own records for four weeks.
We will publish those measurements here, including the ones that do not go our way.
- [1]Slovic, P. (2007). "If I look at the mass I will never act": Psychic numbing and genocide. Judgment and Decision Making, 2(2), 79–95. doi:10.1017/S1930297500000061read in the originalpeer-reviewed
- [2]New, J., Cosmides, L. and Tooby, J. (2007). Category-specific attention for animals reflects ancestral priorities, not expertise. Proceedings of the National Academy of Sciences, 104(42), 16598–16603. doi:10.1073/pnas.0703913104read in the originalpeer-reviewed
- [3]Consolvo, S. and colleagues (2008). Flowers or a robot army? Encouraging awareness & activity with personal, mobile displays. Proceedings of UbiComp 2008, pp. 54–63. doi:10.1145/1409635.1409644abstract onlypeer-reviewed
- [4]Deterding, S., Dixon, D., Khaled, R. and Nacke, L. (2011). From game design elements to gamefulness: defining "gamification". Proceedings of MindTrek 2011.abstract onlypeer-reviewed
- [5]Hamari, J., Koivisto, J. and Sarsa, H. (2014). Does gamification work? A literature review of empirical studies on gamification. Proceedings of the 47th Hawaii International Conference on System Sciences, pp. 3025–3034. doi:10.1109/HICSS.2014.377abstract onlypeer-reviewed
- [6]iSyncSO. The three worlds' code, scoring rules and tests, counted on 2 November 2026.our own measurement
