Problem Definition: Public Entrepreneurship | Barcelona - YouTube

Hi, my name is Beth Noveck.

I’m a professor at the NYU Tandon School of Engineering and the Director of the Governance Lab.

As we discussed in our intro video on public entrepreneurship, the ability to adopt new

technological innovations for governing, especially to use technology to collaborate with others,

demands, first and foremost, the ability to define the problem one is trying to solve.

Although it seems obvious that we need to define the problem, most of us – if we are

honest -- aren’t very rigorous in defining the problem and identifying its root causes,

preferring, instead, to jump right to the solution and start lobbying for its implementation.

Who doesn’t want to launch a new website or spearhead a new policy or create a new

financing mechanism?

But when we fail to define the problem adequately we end up pursuing solutions that, no matter

how innovative, don’t work.

Let me give you an example: Recently, I talked to a government official who was developing

a program to reduce school absences in his city by sending text messages to parents to

remind them that school attendance was mandatory.

He was excited about “behavioral insights” or so-called nudges and wanted to apply what

he had learned about leveraging human psychology to solve the problem.

But the solution assumed that the problem was that parents were failing to supervise

children and, to quote him, “taking them to Disneyland on vacation during the school

year” when there was no evidence to support this definition of the problem.

In fact, the children who were not going to school, first, were concentrated in one part

of the city.

Second, they did not necessarily have parents who were present in the home, and, third,

their failure to go to school had more to do with the economic need to work than going

on holiday.

Thus, the project was doomed to fail.

Becoming an effective public entrepreneur is not simply about knowing how to use big

data or the technologies of collective intelligence or social media but starts, above all with

the hard work of defining the problem.

Thus, our goal in this short introduction is to:

1) explain why problem definition matters; and

2) introduce a process for defining the problem

Why is good problem definition essential?

At GovLab, we know both from our research on open innovation and crowdsourcing and from

our own work coaching thousands of public entrepreneurs and graduate students that the

skill with which a problem is defined is the single most important factor in finding workable

solutions.

The world faces enormous challenges from global warming to economic inequality and we need

more people to work in the public interest on “stuff that matters” and there is value

to articulating ambitious and specific problems to solve that will improve the world.

While issues are compelling, they are too broad to be actionable.

It is not possible to achieve measurable progress without a clear understanding of the problem

to be solved and its root causes.

Only stating the issue – global warming, lack of trust in government, congestion -- leads

to paralysis.

Articulating a specific problem enables action.

The most powerful driver of innovation is “explainability,” namely being clear about

precise goals rather than jumping to solutions.

The more accurate the explanation of assumptions, the more rapid the process of innovation.

Done right, problem definition enables:

Crystallizing opportunities where progress can be made within larger, seemingly intractable

problem areas Developing solutions that are narrowly tailored

to addressing a problem; More informed brainstorming about solutions

that are feasible and relevant; Accurate analysis of a theory of change;

Better mapping of the problems that various actors are working on to avoid duplication

of efforts; More compelling job descriptions that attract

better candidates because they describe problems to be solved rather than vague titles;

Above all, by defining the problem, we make collaboration possible by explaining to others

what we are trying to accomplish.

Whether one seeks help from a colleague, a university professor or a member of the public,

asking others to help depends upon doing more than surfacing an issue or arguing for a solution.

Collaboration depends upon precise and compelling problem definition,

What is Problem Definition?

A problem definition is a clear, specific, and most importantly, actionable description

of a real-world challenge.

A good problem definition is not the same thing as a broad issue area (such as the spread

of Zika), nor is it a solution to be implemented (such as an online community, website, new

profit, grant, or data analytics tool).

Although Zika and other pandemics might be a very compelling issue, they are not, in

and of themselves, an actionable problem.

Although relevant solutions might end up taking advantage of new tools or require the creation

of nonprofit, describing the solution does not tell people what the thing is to be solved

and how they can help.

Moreover, solutions are only impactful insofar as they directly address the root causes of

an issue, which is impossible to assess without a problem definition.

For example, the spread of Zika is causedby mosquitos, which lay eggs in standing water,

which accumulate due to the failure to pick up the trash.

Thus asking for “help with Zika” yields few results.

But a clearly defined statement about accumulating trash and an identification of why the trash

isn’t getting picked up creates a clear pathway for others to participate and to do

so in ways that are useful.

And, because the problem is both clear and important, there is an incentive to collaborate.

Problem definition has a long history of experimentation and scholarship in entrepreneurship and touches

on related methods including design thinking, user-centered design and systems thinking.

But its importance in the world of government, where we tend to focus on and reward solutions

rather than problem identification, is just beginning to be recognized.

How to Define a Problem

We will spend more much time on this in the workshop but

In brief, problem definition is the process of articulating: 1) a specific and actionable

problem, 2) the root causes of the problem, 3) and finding those causes by using data,

research, and collaboration to support that understanding.

Problem definition is a multi-stage process that always requires multiple drafts to learn:

How Do I Do It?

STEP 1: From issue to problem.

The first draft is designed to get past a vague statement of the issue to articulate

an actionable problem that lends itself to concrete solutions.

Write down the issue you are trying to solve in a few lines.

(Zika, global warming, corruption, crime, traffic, school truancy)

Now write down one or more statements of the problem that underlie the issue in one paragraph

per problem.

(Too many people driving into the city each morning for work is leading to congestion

or kids are not showing up for school or small businesses overburdened with paperwork)

STEP 2:Articulating root causes The second draft endeavors to add a description

of root causes to the definition of the problem and, invariably, leads to redefining the problem.

Linking cause to effect requires stripping back and laying bare assumptions.

The better one can explain the problem, the better one can design a solution that addresses

those causes.

Why do you think the problem is happening?

Can you make a list of the root causes of the problem?

List all the potential causes

Again, let’s take an example.

The issue is school truancy and absenteeism.

The initial statement of the problem focuses on kids not coming to school but do we know

why they are not coming.

What do we know about the causes?

Are their parents taking them to Disneyland during the school year to get cheaper rates

or are their parents absent?

Or is coming to school less safe than staying home?

Do they lack a way to get to school?

Do they have to work instead of coming to school?

Develop a hypothesis for the likely root cause.

Write down why is this taking place.

Repeat 3-5 times until you arrive at a series of possible root causes.

Now ask yourself, did I describe the problem at its most granular level.

Can I break the problem down further into more specific root causes and write up 1 page

about the problem.

We have too many people driving into the city because people cannot afford to live in the

city because real estate prices are too high because…...

Now ask yourself, why has the problem not been solved already.

Does answering this question reveal anything new about the problem definition.

Did you get stuck at some point?

Is that because you’ve arrived at the most granular and specific definition of the problem

you can articulate or because you have gaps in your knowledge that you need to research.

If so, write down the unanswered questions as you’ll use the next steps to answer those.

Determine the analysis required to test the hypothesis

STEP 3 Identifying those most affected.

The public entrepreneur needs to understand the people most affected by the problem, their

needs and incentives in order to describe the problem fully

Write down who are those most affected by the problem?

Of course, at some level everyone is always affected by everything but try to avoid generic

statements like “the public.”

in favor of a more granular description of those immediately impacted.

For the truant children, who are they?

Are the children evenly distributed or are they clustered?

What types of children do not come?

Is absenteeism most prevalent in specific schools?

When do they stay home?

At specific times of year?

In the case of public problems, those affected might be both institutional and individual

actors, such as the bureaucrat providing the service and the citizen receiving it.

Are there institutional actors who are implicated in addition to individuals?

Relevant constituencies could include:

· People directly affected by the problem · Organizations working to solve the problem

· Government entities responsible for policy in the area

· Community or social groups

If you could ask them 10 questions to test your assumptions about the problem, what would

those questions be?

Write them down.

STEP 4 Building an evidence base : The fourth draft comes after some research (with affected

people and communities, with experts, at the library and using data) and enables further

sharpening of one’s understanding.

Public entrepreneurs must develop rapid result research skills, especially a knowledge of

how to use data to evaluate the problem they are tackling.

Once you’ve developed an initial problem statement to refine:

Before we close, a note on common pitfalls

Perhaps the most common challenge and a constant temptation is to articulate a solution rather

than a particular problem.

For example, “the problem is that we aren’t using predictive analytics to spot the most

frequent instances of a disease” does not define a problem.

Rather, explain the challenge of knowing where outbreaks occur due to the tendency among

the population to avoid hospitals for specific economic or cultural reasons.

Support with data and explain why it matters.

Challenges also arise when the identified problem is too broad to be actionable.

For example, an initiative aimed at stopping buses from idling while passengers board is

more actionable than an initiative broadly aimed at stopping climate change.

Even if a problem is defined in a sufficiently granular way, a failure to articulate assumptions

and root causes and to identify who is already working on the problem can introduce new challenges.

When problem-solvers fail to map the system in which a problem exists and map the constituencies

working on an affected by the problem, there is the potential for solutions to exacerbate

other, related problems or lead to the duplication of efforts already in place in government,

civil society or the private sector.

Finally, it is difficult to know if a solution works if the problem is not accompanied by

success metrics.

Especially in a government context where budgets are tight and resources only go toward efforts

that can demonstrate success, ensure that the problem you are seeking to address has

a baseline against which to measure progress, and a potential source of data to collect

and analyze after implementation of a new solution.

But, above all, defining the problem is the best way to arrive at a definition of what

success will be if we solve it.

Thank you