An ethical approach to service design in government?

[a six minute read]

Ethics Hero Image

Earlier this month I spoke at the Public Sector Network’s Service Innovation series in Adelaide. One of the workshop themes we discussed around our table was where the line between ‘creepy’ and ‘cool’ lies for digital experience. How much personalisation, from brands vying for our dollar, is too much? How comfortable were all of us with the increasing demand for our information by platforms like Facebook? That sort of thing.

The discussion got me thinking about the stakes for ethical service design in government. Specifically, what are the stakes for government agencies given their unique role as custodians of a special type of relationship – one where the other person generally doesn’t get a choice?

My hunch is that this sort of relationship makes the stakes much higher.

This is already being reflected in a rising level of distrust around privacy and data emerging in conversations like the one around the 2016 Census. Data is an obvious starting point but some of the ethical questions we should be asking may not be as overt as that.

Where do the ethical lines lie?

Government service design is generally driven by a desire to make things easier, improve access for citizens and ultimately make the experience we deliver a better one. As designers striving for these outcomes, how we manage data is actually just one of a whole gamut of ethical decisions that we need to consider. Others include:

  • Decisions about the problem we are trying to solve – how we define it, its impact and why it is deemed important.
  • Decisions about who we involve – the makeup of our teams, who we talk to, who we decide those people represent and who we design for.
  • Decisions about how we engage – the information we give participants in our design processes, the questions we ask, the environment we engage in, the influence we offer them, the types of research and engagement we do (and don’t do).
  • Decisions about how we identify people what details we collect, what categories or segments we put people into and how we choose to talk about them to others.
  • Decisions about how to interpret what they tell us – how objective we are, what assumptions we make, what bias we come in with and whether we choose to acknowledge it.
  • Decisions about how we use data – what we do after the people we have engaged with have left, where we keep their information, what other data we overlay it with and what we use it for.
  • Decisions about the solution we land on – how we define a ‘commonly good’ outcome, the options we exclude, the extent to which we design for the majority and the way we balance the rights of everyone with a stake in what we are designing.

Ethical checkpoints in practice

So how does this play out in practice and what are the risks of not getting it right?

Below are two fairly common examples of scenarios that frequently come up in service design projects. On the surface they seem routine – probe a bit further and a need for ethical decision making comes into play. What is right here and what is not?

Scenario 1: Identifying a customer as belonging to a specific segment to give them a more personalised service (for example, designing a support service for low income earners to manage a payment). In this scenario…

  • Is it right to design a service without including them?
  • Is it right to profile people for research purposes?
  • Is it right to profile (or make value judgements) using data they don’t know we have?
  • Is it right to invite them into a workshop and not tell them why they have been selected?
  • Is it right to permanently flag their record in our system to say they belong to the segment we created?
  • Is it right to do this without them knowing?
  • Is it right if we get it wrong and place the person in a group they shouldn’t be part of and the service we deliver is better (or worse) because of it?

Scenario 2: Using personal information to contact a person proactively about an issue that is important to them (for example, using a mobile phone number to send a reminder that something is due). In this scenario…

  • Is it right to use a personal mobile number to let someone know something important?
  • Is it still right to use this number if you obtained it without them knowing?
  • Is it right if we have assumed the information is important to the person when really it isn’t?
  • Is it right to do it if there are only some potential benefits for the person, but significant benefits for us?
  • Is it right to use that number if our message potentially puts them at risk?

These examples show that even routine service design problems include a large number of decisions about what is ok and not ok. Because we’re not designing services that people can usually choose, these decisions also throw up large risk (to reputation, to trust, to relationships, to service adoption) if we get them wrong.

How then do we decide what is right?

Walking the ethical line: personal rules of engagement

Thinking about a more conscious practice of ethics in the design work I do, I started jotting down some of rules I’ve made for myself that I use to sense-check what I am doing. My manifesto for ethical design practice would go something a bit like this:

  1. Make a commitment to not make judgments about what people need without asking them first.
  2. Be upfront about the level of participation we are offering before we start (and try as offer as much as possible).
  3. Make sure we encourage diversity and inclusion across both design teams and the people those teams are  engaging with (and be honest about who we have left out and what this means).
  4. Be honest about the questions we don’t ask and the answers we haven’t sought.
  5. Admit when our sample is too small to be representative.
  6. Acknowledge our bias.
  7. Be transparent about how we collect, store and use information.
  8. Build alternatives into our solution so that people can opt out at any stage.
  9. Admit when we have been unable to solve the problem (or when we have made the problem worse) and try something else.
  10. Be honest when the outcome is really a win for us (and not for them).
  11. Continually ask ourselves if we would be comfortable is this were us.
  12. Expand the boundaries of policy and risk management to do not just what is allowed, but also what is right.

How are you tackling ethical questions as part of your service design practice? What’s your own manifesto and what rules do you live by to help you make the right decisions?