1. Start with the solution
Problem definition is for wimps. Pick a neat solution (preferably imported from an entirely different sector or from a friendly think tank) and claim that it can solve everything for which you are responsible. The more problems you can apply it to irrespective of context, the better.
2. Don’t bother improving existing services - ‘innovate’ instead
Avoid focusing on the steady improvement of existing services, especially if it means funding them properly. You’re only going to be a minister for one or two years, so there’s no time to hang about (especially if you’re ambitious, see also point 10.) Aim for massive institutional reorganisation and system-wide changes, the more complex and disruptive the better. As these trundle on, try to distract people further with at least one new half-baked idea every week.
3. Use evidence sparingly and only to justify your policy
Ideally your policy should be led by polls and focus groups, or better still reliant on received wisdom in the Westminster bubble. Try to keep as much of the analysis and advice that has informed your policy secret (this should be commissioned from self-interested consultancies and big providers), and resist calls to publish it. This will help to generate suspicion and opposition, which could prove useful later on (see also point 10). If you do have to provide evidence, make sure you carefully select only that which supports your presuppositions and be sure to rubbish any conflicting data (see also point 1).
4. Don’t explain what you’re really doing or what you hope to achieve
Even better, load the same policy with multiple and conflicting objectives.
5. Be overly-optimistic
Ignore changes in the economy. Make your policy reliant on best case scenarios and ignore worst cases (if the latter are ever presented to you, never acknowledge or publish them).
6. Don’t bother consulting early or sincerely
You’ve already announced your policy and invested political capital in it. So try to avoid people who might know what they’re talking about, especially those who work at the frontline. They might challenge some of your most cherished assumptions (as point 1). This will also serve to undermine confidence and trust in the political system generally - so further reducing people’s engagement and ensuring your successors can introduce even more disastrous policies in the future.
Be sure to push your legislation through Parliament as quickly as possible. Proper debate and scrutiny could inadvertently improve your policy. Rely on statutory instruments and codes of practice to patch-up the policy later if you have to (these have the added benefit of receiving less scrutiny than legislation). Only then should you establish a few pilots - but don’t let the results from these affect your timetable or plans in any way.
8. Ensure that your policy is dependent on a complex, costly, never-done-before IT system
Believe what the big IT providers tell you. This can increase your chances of massive policy failure exponentially.
9. Under-fund your policy
And if you can, assume that ‘competition’ will drive down costs (see also point 5).
10. Ignore the opposition
Co-opt a couple of dominant providers to suggest support for your policy, and ensure they are too big to fail. There may be times when you start to doubt what you’re doing, but remember however that criticism proves that you are a conviction politician, the more the better. Label those who have reasonable doubts as ‘trouble makers’ or ‘ignorant’. This will allow you to ignore or dismiss the obvious problems with your policy for as long as possible. This stance will also play well to your ideological allies in the commentariat. It may even boost your ‘leadership credentials’.