[Texto original en inglés. En los próximos días incluiré la traducción.]
Three pieces of (legal) advice for digital implementation
As part of a course at the Harvard Kennedy School, I have been asked to share what I believe to be three pieces of advice for a legislator in order to avoid digital implementation nightmares later on. As a side note, I must add that the course has turned out to be a great opportunity to reflect on how to bridge my interests in both policy and digital media as well as a perfect excuse to start blogging again after three-month hiatus (
also, the professor will be reading this).
Given what I have seen in two very different settings, such as Peru and the U.S., I think that there are three broad lessons to keep in mind: Identify what requires flexibility, have a good sense on the context in which digital implementation takes place, and ensure that there is a reasonable timetable for the project.
a. Identify what requires flexibility (and what does’t)
Legislators and policy designers like to control everything and to that end they are prone to provide implementers with a detailed roadmap of what to do, step by step. The problem is, of course, that there is a reason why they are not the ones in charge of implementation–they do not necessarily know all the nuances involved in it. Therefore, there is value in giving implementers, specially those in the digital space, son room to maneuver and come with solutions of their own. This requires the legislator to identify what aspects of the policy admit flexibility–and which aspects don’t.
Peru offers a good example, as the PPK government has tried to step up its efforts to formalize small businesses using digital tools and simplified mechanisms for declaring taxes online. What things require flexibility? Well, those in charge of website design should probably have a lot of space to try different user interfaces, for example. What should not get a pass from the government is the system’s support that cross-checks identification using the databases from the Tax Revenue Agency, SUNAT.
b. Have a good sense of the context
It is very different to implement a policy that is popular than to work on one that is not. When a particular policy is unpopular, even the minor flaws of the rollout are magnified by its opponents, which increases the stakes for implementers. Legislators need to be aware of that and balance the political imperative of rapid implementation with the equally important need to get things right the first time. Just look at Obamacare as an example in the U.S.–a policy that was resisted by many was declared by many to be failing even though it is not because that is the impression that the rollout gave to the general public.
c. Ensure a reasonable timetable
When it comes to websites and applications, my personal experience is that the first impression it creates is very important in order to engage the user over the ling term. This means that the rollout is critical, and it needs to be done right more than it needs to be done quickly. There is an inherent tradeoff here that is important to accept: If we want the rollout to be successful, we need more time to get it right. Obamacare in the United States comes once again as an example of a situation in which expediency was prioritized over quality, to disastrous initial results. Legislators are only end-users of digital content, so it is easy for them to underestimate how much work it takes for websites and applications to work properly. As a rule of thumb, perhaps, any legislator should add at least 20% of extra time to whatever timetable they initially consider to be necessary to get implementation right.