How can architects use place-specific data on design outcomes to have a meaningful impact in future projects? Rory Bergin of HTA Design proposes a change of approach
‘If you can’t measure it, you cannot manage it’ has been the mantra of management gurus, data scientists and engineers for generations, with good reason. The premise is that if you collect data on a problem over time, aggregate it and analyse it for patterns, it should be possible to change the behaviour of people or systems to optimise the outcomes.
This thinking hasn’t had as much of an impact on design, in part because designers aren’t generally taught to use data. The question of whether there are ways in which we can use data to inform us in our efforts to design better buildings, social spaces and the networks that connect them, is not generally being asked in architecture education or practice. The type of data that is being used to drive design in London is often used in negative ways to direct design away from something; the sound monitoring of traffic driving orientation of buildings, the temperature data from remote weather stations driving microclimate design, or the demographic data on population growth and immigration driving local or national policy.
Little of this data is specific to the place we are designing for, its people, its character, use patterns and history. The arrival of cheap sensors and the Internet of Things (IoT) is meant to enable an explosion of potential in the management of our physical lives. Innovations like parking spaces that tell you when they are empty, fridges that order the milk for you, and other small but useful incremental changes that make our lives easier and reduce what is called ‘friction’, a short-hand term for the bureaucracy of daily life. But so far, these technologies have not fed into the activity of design.
The potential prize from data collection is enormous. If we could get to a position where we have an evidence base to inform design that is specific to the place we are designing for, it would have a big impact on design behaviour and outcomes. We would move from a world where design is based on the personal ideas and ambitions of designers, planners, and other stakeholders, often based on assumptions or limited personal experience, to a world where design activity could be supported by an up-to-date and relevant evidence base. In my experience the use of data to support design in cities hasn’t been specific enough to the place to be effective. To make data about a place meaningful, the data needs to be rich and specific to the place. Data collection is currently sporadic and patchy and often collected in situations where there is a problem, like crime data in a poorly designed neighbourhood. Data is rarely collected about happy people.
When it comes to the use of data for planning policy, data is collected for policy assessment or development, but the lengthy timeframes for planning policy decisions often means that the societal drive behind policy will have changed before the policy is implemented.
Currently, place-specific data is often collected by app providers through our phones as a way of selling services to us. The ranking of restaurants and pubs, the photographs of tourists, the local bus timetable. But the generic nature of many of our apps, which have become the primary means of data collection, means that app users have access to information collected for a specific purpose but not enough access to a wider pool of data created by individuals. Where place-specific data exists, it tends to be held within a specific app, like restaurant rating tools, or exercise apps, whereas to be useful, we need this to be more widely available in a way that we can analyse it and derive meaning from it.
If we knew that 10,000 people use a park every weekend to go running, and 500 used it to take their dog for a walk, which use case would we prioritise when we design the movement routes in the park?
At HTA we have begun the task of capturing data on our projects and monitoring this over time. It’s a slow but valuable effort that will take time to bear fruit, and perhaps that is part of the problem with design data. We are all guilty of having a short attention span related to the design life of buildings and the task of data collection is a long-term one more related to the place. When buildings are completed, everyone related to a project moves on and there is little emphasis on collecting data or feedback.
Because there is very little activity to collect data on completed projects, the profession runs the risk of designing and constructing buildings that have the same mistakes in them as the last generation’s version. Where we should aim to get to, collectively, is an evidence base on the successes of our schemes, as well as feedback on the things that didn’t work so well. There are good reasons to collect positive and negative feedback, as positive data tells us what is working and can be used to provide an evidence base to reinforce reasons for repeating something.
External pressures like the Building Safety Act and the Golden Thread will help to bring in a better culture of data creation, storage and management on design projects, and this will eventually feed into the systems of clients who manage portfolios of buildings. Designers are increasingly being asked to provide data to enable investors to make comparisons between assets in different countries and under different jurisdictions. In an international market, data may become the currency that enables designers to compare themselves to the competition overseas.
There are many sound environmental reasons for the collection of data on buildings, as too often decisions are made based on aspirations rather than evidence. In my career we have gone through several cycles of silver bullets that would solve the environmental crisis, from biomass boilers, to Combined Heat and Power systems and finally arriving at heat pumps. Perhaps if there had been some better data at the beginning of all of this, we might have got to the answer quicker?
The growth of data around embodied carbon in construction is another case in point, and there are now a number of tools available for the industry to use based on large datasets, enabling evidence-based decision making. But is the data based on enough information, is there a good evidence base?
We also have the arrival of AI to complicate the situation. For it to participate in the effort of design it needs training data, ideally data covering the performance of buildings that work well for their occupants, owners, society and the environment. That way, anyone who trains AI on design data that is available will do so knowing that they aren’t going to repeat the mistakes of the past. The alternative is quite frightening, there is a real prospect of untrained users using AI-based tools to design buildings cheaply that repeat the mistakes of the past. Would you like a McMansion? Of course – here it is, or perhaps you would like a version of a French 18th century chateau? No problem.
Lets try and work together to provide the data that the future generation of designers need to ensure that the future is utopian – rather than dystopian.
Rory Bergin is partner, sustainable futures at HTA Design