Trial and error is the most successful problem solving technique

So said Tim Harford (known to many as the presenter of Radio 4’s More or Less as well as a distinguished economics writer) last night at the Royal Society of Arts. Harford was kicking-off a week long series of talks to promote his new book, Adapt: Why Success Always Starts With Failure and was making the point that the modern world is too complex for us to hope to solve many problems by simply getting the best brains to think up the one answer. A far better route to take is to experiment with many different solutions and see what works.

Tim Harford applied that both to business – a business culture that sees numerous different start-ups tried out is more likely to succeed – and to politics – piloting and testing a wide variety of policies is better than relying on the one magic solution. Especially as the one magic solution often turns out to be not so magical and not that much of a solution.

Yet in politics there is often a deep reluctance to test. Sometimes it’s seen as a matter of principle – the policy should be the same everywhere with no postcode lotteries, thank you very much. Sometimes it’s seen as a measure of political leadership – good leaders are strong leaders are people who know their minds are people who don’t need to test.

He put up a big photo of George W Bush and John Kerry from the 2004 US Presidential election and pointed out that the election could be summarised as the one who said they were sure about everything beat the one who said they weren’t so sure and sometimes changed their mind.

The immediate political implication in the UK, though Tim Harford didn’t explore this point, is that having a diversity of approaches to supplying public services can harness the benefits of experimentation and learning. That means relaxing about the role of the likes of not-for-profit firms, mutuals and charities in experimenting with different ways of providing services, rather than seeking to have it all run through the one local council direct provision route. It also means giving up the idea that all should be ordered from Whitehall towards some mythical uniform nirvana (which never ends up uniform, of course).

For Liberal Democrats it means kicking the pernicious habit of saying how much we like diversity and local choices but so often finding a reason as to why on this particular occasion it’s just not quite right, yet. It also requires a willingness to defend variation rather than bemoan postcode lotteries. As I wrote before:

There is no inherent reason why the population of a small Scottish island should have the same health priorities as that of a densely populated London borough. There is no inherent reason why a swathe of rural Wales should have the same transport priorities as a chunk of urban Midlands.

The problem is not variation. The problem is when variation happens for the wrong reasons. And so the answer is not trying to impose uniformity, it’s about greater local decision making – so the variation is for the right reasons, not the wrong ones.

It’s the lottery part of postcode lotteries which is the problem, not the postcode part.

(And of course remember that the logic of the benefits of testing apply to political campaign techniques too. When was the last time your local party tested out different versions of mailings to promote postal votes for example? Or different formats of appeal letters? Or different ordering of questions on residents surveys? I’ve done all three and know my work is much the better for it. So if you’re not in the experimenting club, do give it a whirl.)

Read more by or more about , or .
This entry was posted in Op-eds.


  • The trouble is with the local things for local people argument is that it often fails to account for economies of scale. Take, for instance, hospital sizes in scotland. There are huge hospital complexes in small Scottish cities, such as Dundee or Aberdeen, because they have to cater for huge geographical areas with varying needs from the run down areas of the big cities to affluent country folk. Ceding control to the immediate locality of the hospital would change the standard of care for patients in more rural areas as their priorities would be second to the needs of the cities. It is for reasons such as this that there is no desire in Scotland to decentralise state provisions.

    Local control means prioritising the selfish needs of one population over another thanks to a lack of strategic oversight and central planning. The idea that the virtues of selfishness will lead to improvement is a philosophy whose application in the real world has proven unhelpful.

    That said there is nothing wrong with considering the use of evidence and experimentation to guide decision making in government. I look forward to Lib Dem Voice robustly challenging government policies that are at odds with the available evidence. Education, Health, Benefits, Transport, Energy, Immigration….. there’s lots to start with.

  • Matthew Huntbach 7th Jun '11 - 8:38am

    Yes, trial and error is a good way of solving problems. We need the freedom to have a diversity of solutions to be tried so that we can see which work and then build on those.

    This may seem like a prelude to the usual call for “opening up services to competition”, but I suggest the reverse. People will experiment with different ways to solve problems, using innovations which may or may not work and we don’t know until we try them, when they are confident they will keep their jobs and when their surroundings are stable. When people fear that one false step means they are out, they will “keep their heads down”, they will do things in exactly the way they are always done and in the way everyone else does them, because they know that works and they won’t lose their job if they have not singled themselves out by doing something different and finding it doesn’t work.

    I have seen this work in the public sector in many forms – the introduction of a more competitive attitude and tighter control on budgets means it isn’t worth anyone’s while to take the risk and innovate. To me this is a large part of the explanation as to why when money has been poured into pubic services and we have had an opening up of competition in many places, perceived quality has fallen rather than risen. Competition driven by fear of failure may work well for the very wealthy, who if they lose never really lose because they have their wealth and contacts to fall on and bring themselves back, but for those where losing really can mean sinking into a life of utter degradation, fear of failure is numbing. Those who rule us and those who dominate political discourse cannot see this because they mostly come from backgrounds of extreme wealth and privilege, so they just do not have the life experience to see how the market competition they urge on us has the opposite effect to that they suppose.

    I don’t see that much difference in the private sector. The old adage “no one ever got sacked for buying IBM”, though now outdated, sums it up well – the safe option if you want to keep your job is to do what is standard, what everyone else does. It does seem to me that in many sectors competition leads to sameness, not diversity, because the safe option is to go for what everyone else is going for.

    The public sector in this country when there was much less control (both financial and regulatory) and much more stability had a great record for innovation. Sure, some of the things that got tried went wrong, but at least that means we know they don’t work. What is needed in this country is a restoration of pride and dignity in work, and of confidence that comes from stability. At the moment, fear is all pervading, and the still fashionable marketisation lines are based on making that fear worse under the misassumption that will drive people to success. It hasn’t and it won’t. Indeed, the fear that is pervading now seems to be driving the economy down as no-one wants to take on any big spending if they don’t know whether they’ll have a job next year.

  • Malcolm Todd 7th Jun '11 - 3:07pm

    I’m just glad to see that having tried it, you’ve decided the upside-down Comments policy was an error! 🙂

  • Dinti Batstone 7th Jun '11 - 5:56pm

    Hear, hear. I’ve tried to push this mindset with initiatives like targeting EU Voters and job-shares for parliamentarians.

    Innovation always involves a degree of risk, but we can learn a lot through trial and error, so fear of failure is not a good excuse for sticking with status quo.

    The late Andrew Reeves (RIP) understood this. Initially extremely sceptical about targeting EU Voters, he came round to the idea as London Campaigns Officer in the run up to Mayoral and Euro elections and later told me he adapted the initiative for Scotland. There have been many very moving tributes to Andrew over the last few days, so this may seem like a trivial one by comparison, but I will always remember him for being big enough to change his mind and try something new. My heart goes out to Roger and all his family and friends.

  • Ed Shepherd 8th Jun '11 - 3:24pm

    Matthew Huntback’s comments on this article are very insightful. Thank you.

Post a Comment

Lib Dem Voice welcomes comments from everyone but we ask you to be polite, to be on topic and to be who you say you are. You can read our comments policy in full here. Please respect it and all readers of the site.

To have your photo next to your comment please signup your email address with Gravatar.

Your email is never published. Required fields are marked *

Please complete the name of this site, Liberal Democrat ...?


Recent Comments

  • Alex Macfie
    Remember that ordinary voters are typically not partisan ideologues. In the Kennedy years we were able to appeal to younger voters with a radical agenda but wit...
  • Robin Stafford
    For those interested in health and the NHS, Roy Lilley's blog is worth a look. A strong supporter of the NHS thought not uncritical. https://myemail-api.cons...
  • Peter Martin
    @ Simon R, Yes well done! :-) It was a somewhat tongue-in-cheek comment. I usually have difficulty in persuading contributors to this site to think in ter...
  • Simon R
    @Peter Martin haven't you yourself have (quite correctly) often pointed out the need to think in terms of resources rather than money? ;-) There are differi...
  • theakes
    Over the past fortnight the United States Opinion polls seem to be turning, some have Biden in the overall lead, just! State by State unclear....