The document doesn't give information on the definitions of the seven categories that are represented in the table. It might be interesting to find out, how the broad bundles of practices have been made -- why is the toilet flush not subsumed under body care, why is car cleaning in the same category as garden watering? The categorizations made might come from the urge to find broad enough categories to make graphics and tables readable (opposed to a fine-grained detailed table with hundreds of different practices that use water). But they might also emerge from contraints in measurment, if the spended water is measured with counters intstalled at different water valves in the house. That would fit very plausible to the categorizations as for example watering a garden and cleaning a car is most probably both done with a water valve outside, if existing and accessible. It would be interesting, wether queries have been made to configure the list of practices that are mentioned and how they are clustered or wether they result from mere considerations of what valve is most propably used for what.
The category that evades my ways of making sense of the data and imagining their crafting-process is that of "small business" in the last row. If it would be different water valves, this would not make sense, unless this number just points to workshop-watervalves which would not include tea-cooking in the kitchen of creative workers in home office for example.
