The meeting started fine. An hour in, the room's gone heavy, the air feels used, and the last twenty minutes are harder to follow than the first twenty. Nobody's opened a window. Nobody's going to.
There's a number for this, somewhere in UK law. It isn't the number that explains what just happened.
The Law Covers Ventilation. It Doesn't Cover Attention.
Under the Workplace (Health, Safety and Welfare) Regulations 1992, employers have a plain duty: "effective and suitable provision shall be made to ensure that every enclosed workplace is ventilated by a sufficient quantity of fresh or purified air." That's the process requirement. For the substance itself, the Health and Safety Executive publishes a Workplace Exposure Limit for carbon dioxide: 5,000 parts per million averaged over eight hours, or 15,000 ppm over fifteen minutes.
Read HSE's own explanation of why that number exists, and it's not about thinking clearly. It's about not getting hurt. At high concentrations, CO2 causes headaches, dizziness, confusion, and loss of consciousness, and in confined spaces, where it can displace oxygen, it has killed people. The legal limit is a safety threshold, built for the worst case. It was never designed to say anything about an ordinary meeting room on an ordinary Tuesday.
What a Controlled Study Found, Well Below That Number
In 2016, a Harvard-led study published in Environmental Health Perspectives put 24 office workers (architects, engineers, programmers, managers) through six full working days in a blinded, environmentally controlled space, varying CO2 while holding everything else steady. Three levels were tested: roughly 550 ppm, 945 ppm, and 1,400 ppm, each more than three times below the legal exposure limit.
Cognitive test scores were 15% lower at 945 ppm and 50% lower at 1,400 ppm than at 550 ppm. Modelled as a straight line, every 400 ppm rise in CO2 was linked to a 21% drop in scores. Two things are worth stating alongside that finding, not after it: the study had 24 participants, tested in a lab chamber, not a real office in Leeds or Liverpool, and it was partly funded by a gift from a building-systems manufacturer, which also had input on the study's design (specifically, adding an extra test day and a third CO2 level). The testing tool and the analysis were independent of that funder, and the result lines up with separate research below, but the funding is named here because it should be, not because it was hidden.
What Happens Outside a Lab, Including in the UK
A second study, led by the same Harvard research group and published in Environmental Research Letters in 2021, moved out of the chamber entirely. Over twelve months, 302 office workers across six countries, including the UK, took short cognitive tests on their phones while researchers tracked the CO2 and fine-particle levels around them in real working buildings. For every 500 ppm rise in CO2, response times slowed by 1.4–1.8% and the number of correct answers per minute fell by 2.1–2.4%.
The researchers looked for a safe floor, a concentration below which the effect disappeared. They didn't find one in this data. That's a more honest finding than it sounds: it means the relationship doesn't claim a clean cutoff point, only a direction that holds as CO2 rises, in real offices, not simulated ones.
What This Doesn't Tell Us
It doesn't tell us the CO2 level of any specific room, or that anyone should go and buy a monitor or crack a window on the strength of two studies. It doesn't tell us anything about hydration, caffeine, or what any drink does to attention in a stale room. Both studies measured ambient CO2 and cognitive test scores, nothing else, and that's exactly as far as they should be read.
What they do say, taken together, is that the number the law actually enforces and the number where thinking starts to slip are not the same number, and the distance between them is large enough that most meeting rooms will cross the second one long before anyone could plausibly claim a safety problem.
Why 0800Labs Is Interested
This is the same ground as our piece on workplace temperature, looked at through a different window: a legal limit that exists for a real reason, sitting well above the point where something quieter is already happening to the people in the room. Not a time of day, like the 3pm slump. Not a run of back-to-back meetings. The air itself, doing something nobody in the room is measuring.
We call the occasion underneath all of this Workday Hydration. It's our own term, not an established category, and not a claim about what either study found, or about what any drink does to a stale room. It names the moment we're interested in: the one that gets built for a desk day, whatever's actually in the air.
We're building something for that moment. If that's useful, the waitlist is below.