We all know that occupant behaviour influences the way a building performs. Whether that is due to over-complicated control interfaces, or heating homes to greater than design temperature because it is now cost effective to do so, report after report mentions the performance gap between predicted and actual performance. The Ashford report 1 showed that measured water use in Code homes was anything between 40% to 500% of the projected amount. So why is this? Are we using incorrect data to make these predictions? How helpful is the term ‘average use’ anyway? After all, the phrase ‘lies, damn lies, and statistics’ has more than a grain of truth in it. I would argue that using statistical data is vital – we cannot design in a vacuum. But we need to know how robust the data is, who it is collected from, how it is collected and how many people were asked. And, fundamentally, we need to be prepared to change design practice if research shows that our original assumptions on behaviour are wrong.

The ‘five’ minute shower

Calculations about water use in the UK are still based on the ‘fact’ that the average shower is five minutes long. Except that it isn’t! At Ecobuild in 2010 I asked ‘Is the five minute shower an urban myth?’ presenting data ech2o had collected, where the average shower was a staggering 13 minutes. We were confident in the research, and only slightly surprised by the figure; anecdotal evidence had convinced us that average shower time was far higher than five minutes and our data set contained a high proportion of adolescent users. Our original report spawned a host of similar research all of which confirmed that the five minute shower was indeed an urban myth.2 Since then, the message of ‘take a shower instead of bath’ has been changed to ‘take a short shower instead of a bath’ and most Water Companies have a four minute shower challenge, complete with free shower timers.

The EST has recently published the biggest body of research on domestic water use in the UK to date.3 Data has been analysed from 86,000 self-selected households. Their research shows that 58% of all water is used in the bathroom, the shower is the biggest water user in the home taking 25% of all water and that the average shower time is 7.5 minutes. So, after three years of a concerted campaign by the water companies to ‘have shorter showers’ the average shower length is still 50% greater than the supposed norm. If we are collectively spending 50% more time in the shower than anticipated then not only are all calculations about water consumption flawed (both the Code and Part G calculations assume a five minute shower) but demand management plans by the Water Companies, CO2 emissions from heating water for showers, and the best way to reduce household bills are all based on incorrect information.

WC use in schools

This leads me onto another piece of research which shows how important it is to actually measure water use rather than make assumptions about it and then design to match those assumptions. The default measurement for WC use in homes is five flushes a day and in offices and schools it is three. Both WC use at home and in the office is based on research and the findings do not appear controversial. But the school ‘fact’? Well that is another matter entirely. ech2o has recently completed and published research that shows that this is an entirely flawed way to calculate the reduction in water usage in a school once WC upgrades are introduced.4

The report analyses data from 457 primary school pupils and 152 secondary school pupils. Less than 25% of the pupils we surveyed use the toilets three times a day, and average use is just 1.3 visits a day. If we consider secondary school pupils separately the discrepancy between the received wisdom of three uses a day and the reality is even starker. Of the 152 secondary school pupils we asked, 65% of them do not use the toilets at all at school, and the average use is just 0.6 times per day.5As girls at secondary school use the toilets even less than the boys at 0.5 times a day on average, water savings from upgrading WCs or retrofitting flush reduction measures or using rainwater to flush the toilets (as opposed to urinals) are over calculated by a factor of six.

So as design professionals what can you do?

The mains reasons school pupils cite for not using the toilets is that they are ‘dirty’ or not safe. Designing individual toilets that are accessed directly from corridors prevents bullying and whilst not getting over the ‘toilets are dirty’ argument does make WC use more likely. If upgrading bathrooms in social housing put in showers (sadly missing from the Decent Homes standard), and don’t specify power showers, when designing bathroom makeovers. Understand the different environmental impacts between cold water and hot water and mains water and rainwater. Specify solar thermal for high hot water users. Use the AECB Water Standards to ensure both a water and carbon efficient building.6And, when discussing with clients how much water a particular solution will save, ensure you are using robust figures of water use to start with.

First published in Green Building Magazine, Autumn 2013


  1. www.homesandcommunities.co.uk/download-doc/6473/11054
  2. What is truer is that, in most studies, more people have a five minute shower than a ten minute shower or a two minute shower or a thirty minute shower. But that is very different to average shower time.
  3. www.energysavingtrust.org.uk/About-us/The-Foundation/At-Home-with-Water
  4. Reducing WC flush volumes in schools – are we overestimating the water savings? Available to download atwww.ech2o.co.uk/reports_wc_schools.shtml
  5. Even in primary school, 18% of children never use the toilets while at school and average use is 1.6 times a day.
  6. www.aecb.net/publications/publication-categories/aecb-water-standards/

Back to water column


Getting the facts right – why empirical data matters