July 1, 2016
When starting a new project, you’re often faced with a broad and unstructured problem.
Perhaps you’re creating a new service to reach a new customer segment. Maybe you’re responsible for developing a new product that responds to emerging market trends.
Regardless of the problem you’re working on, creating new value for your organization requires you to see the opportunity more clearly.
When a problem could be approached in many different ways, how do you start?
Here is a structured approach you can take to frame problems in a clear way:
- Dissect the problem into discrete components.
- Identify dimensions from the components and data about them.
- Use the dimensions to generate different approaches and ideas.
Use this approach to set yourself and your team up for success.
To better understand this approach, let’s take a real-world example. Imagine you are responsible for coming up with new offerings to help people who have a common cold.
Step 1: Dissect the problem.
Start by digging deep into the problem context itself.
Think about the user’s experience in the context of the problem: someone who just came down with a bad cold in the middle of winter.
How do they feel? What questions do they have? What do they do? What items do they use? With whom do they interact? Where do they go? How do these change over time?
For each of these areas, write down dozens of answers, each on a separate sticky note. Post all of them on a board. Dig into as many different components of their situation as you can.
Thinking through questions like this generates a nice set of data about the unstructured problem. Buying medicine, coughing, eating soup, drinking tea, visiting the doctor, trouble sleeping, congestion — these are all discrete pieces of the experience of being sick.
At this point, we’ve identified a variety of data points. But the data points aren’t inherently useful in and of themselves. They’re too numerous and unstructured.
The real impact comes from identifying patterns and relationships in this data.
Step 2: Identify patterns.
Identify patterns by grouping data points that relate to one another. This is called an Affinity Sort. You’re going to group like items together into clusters.
There are no hard-and-fast rules on the number of clusters or type of relationships you should use. Let your intuition and quick discussions with the team help make meaning and group ideas. Don’t get bogged down in debate.
For our person who has come down with a cold:
Do some events happen at the same time in your experience?
- Scratchy throat
- Sneezing and sniffling
- Aches and general tiredness
Are there objects that you use to combat your cold?
- Hot towel
- Extra blankets and sweaters
- Thermometer to monitor temperature
Combining these data points creates distinct categories. Give each group a name that describes what unites them. These categories provide potential dimensions of the problem.
Do your best to make your groups mutually exclusive. The form the language takes can be a clue. For example, activities (such as sniffling and sneezing) often end in "ing," while emotions (such as frustration or anger) will tend to be nouns.
For a pattern to be meaningful, it needs to be specific enough so that it’s unique from other groups. But it shouldn’t be so specific that there’s only one or two data points in the group.
If there is a group with just a couple of items, see if you can brainstorm several more to fill it out. That will help determine whether it deserves to be on its own.
The number and scale of groups you find in any given problem will vary. Especially enthusiastic teams will generate a lot of data and categories.
As a start, here are some common dimensions we tend to find valuable on projects
Values: categories of things that people actually care about
- comfort, speed, and cost of receiving care
Triggers: points in an experience that spark a user to take a particular action
- coughing fit or fever? take medicine? walking by the store? buy medicine?
Native categories: the natural way that people categorize information
- cheap stuff, from doctor, for the bathroom, for nighttime
Modes: the goal or mindset that a consumer holds while performing a given behavior
- manage my symptoms, prevent cold from worsening, avoid cold altogether
Pain points: specific issues a consumer encounters when trying to accomplish their goals
- difficult to open packaging, sore nose, drowsiness
Activities: the actions performed during an experience
- blow nose, go to bed early, disinfect surfaces
Use cases: the way in which different types of people use an offering
- help fall asleep, stop coughing, relieve congestion
Identifying dimensions helps us frame the problem in a new way. Each provides a different angle to focus on and assess what is happening. Then, in response to what we find, we can imagine new solutions or parts of a solution to help.
Step 3: Focus in to generate and evaluate ideas.
The dimensions of a problem provide a clear structure. Focusing in on specific dimensions is a great way to generate new ideas.
If we’re a pharmaceutical company, customer pain points open up new sources of potential value. Difficult to open packaging may lead us to question and reframe how we deliver our medication.
Large pills that are hard to portion would lead us to rethink size and portion-ability.
Sometimes changing the locus of problem-solving leads to new territory. If existing competition is all focused on “managing symptoms,” then a brief exploration of “avoiding a cold” might prove fruitful. Combine a simple medicine with an interesting menu plan and perhaps you’ve started to break barriers between food and drugs.
Creating and using a problem structure promotes clear thinking. It prevents you and your team from talking about everything at once. And it allows your critical and creative skills to be aimed at specific targets.
Dissect and structure your problem to work on it more productively
Most growth and innovation problems you face at work are not well-structured. Before you jump to possible solutions, broaden your understanding of what is happening. To do so, dissect your problem to create a lot of individual points of data.
Group that data in new ways to create categories and major dimensions of the problem. Finally, look at your problem through those dimensions to create focus for both analysis and synthesis.