Feb 13, 2020
This post first appeared on the One Team Gov blog.
We’re repeatedly guilty of attempting ‘decisive action’ based on little evidence and with strong incentives not to change strategy if we get it wrong. And getting it wrong is a likely thing if you’re dealing with huge complexity or profoundly difficult problems that no-one else can fix. Which, let’s face it, is in government’s job description.
We know from experience that even simple, transactional services are hard to get right but government does a lot more than simple transactional services and no amount of iterating the delivery will fix the problem if the underlying policy needs a rethink.
So when designing new legislation and services we have to acknowledge a simple truth: human beings are very poor at making choices about circumstances they haven’t experienced first-hand. For better or worse, government has to do this all the time, usually under time-pressure, often with very few data points and always in the public eye. The reality is that new public services — especially big ones — are likely not going to work in practice how they were imagined in an office in Whitehall.
How do we change the game so that government is allowed to design for change when voters and the media demand certainty?
A hypothetical example
Let’s say there is a burning issue for 18 year olds where we need to promote better personal choices and that we want to intervene at least six months before a critical moment. Government might decide that passing legislation to create a free, personalised advisory service is the right lever and that surveying people after they made their choices is the best accountability measure.
Just to be clear, what this really means is delivering personalised advice to hundreds of thousands of people, all living different lives, all over the country. Under those circumstances “right first time” is a noble ambition but a lousy strategy.
What if it turns that for many people the timing is wrong: giving them advice six months before they make their choice is no good, it actually has to be less than six weeks? If we’re completely committed to ploughing ahead with version 1.0 of the policy, government will continue to spend taxpayers money not solving the problem. We need the ability to adapt based on real-world feedback from the service.
Staying the course vs. falling forwards
In politics changing course on an issue is always seen as a failure and an unwelcome opportunity for the opposition to poke fun at PMQ’s. In the most extreme case it can end up with the resignation of a minister or affect an election outcome. By contrast, in the private sector the speed with which an organisation can change strategy based on new evidence is a strong indicator of management competence. While Government fetishises ‘right first time’ the private sector reveres the ability to adapt to the unknown. So why the difference?
In the private sector there’s a maxim that you should always have a business plan but you should never expect it to survive contact with the real world. The most successful businesses are always ‘falling forwards’ and hardly ever assume that something will be right when it’s first launched — they rely on information from live services to continuously improve and if necessary substantially rethink.
I think government is often backed into an entrenched position because ministers are advised — or incentivised — to rush to a solution rather than asking a better question. Why? Because it looks more like progress. So we dive for big ideas and monolithic executions where there’s no opportunity to learn and no room for manoeuvre. Factor in an electorate that’s extremely intolerant of anything that looks like a u-turn and it’s no wonder that so much policy is cast in aspic before we ever really know if it’s a good idea.
So what do we do about it?
It’s tempting to think that the only way to tackle this set of problems is to “fix all of government” or “change how voters think”. Obviously we need something more actionable than that…
- Create a virtual policy environment to test policy ideas out prior to submission
- Create better language/tools for private offices to shape better ministerial announcements
- Embed user-centred design (UCD) skills and toolsets into policy teams so they can more quickly interrogate ideas and substantiate advice with compelling evidence e.g. video research
- Anticipate the timing of advice and subs to ministers: ask the hard questions early so you don’t get caught on the back foot
- Create toolkits to help comms teams message the value of learning and experimentation to citizens and media
- Pick exemplars that are lower risk/politically less sensitive
I’d love to hear your suggestions too.