top of page

Designing Frameworks with Clarity

A Step-by-Step Guide to Crafting Effective Frameworks


In two previous posts, I presented a taxonomy for decision-making & problem-solving frameworks, consisting of four patterns, and how the shape of our thoughts impacts the design of frameworks. 


Now, we will delve into the how-to of creating a framework. To reveal the underlying structure of information (the organizing principle) and arrange it into a visual representation, follow a process with three stages¹:

  1. Purpose

  2. Conceptual Model

  3. Visual Form


1. Purpose

Think of this first stage as the solution requirements for your framework. It’s important to be clear about why the framework is being built and who will use it:

  • Goal: What is the goal of the framework? What are you hoping to achieve? What real-world problem or opportunity should this framework address? 

  • Desired end result: What specific outcome should the framework help achieve? What should users be able to do with it? What does success look like? Think of this as a tangible, concrete outcome.

  • Users: Who is this framework intended for? Who is affected by this outcome? 

  • Cognitive activity: What type of cognition is required to achieve the desired end result? 


For example, the purpose of the Ansoff matrix:

  • Goal: Decide which growth option will work best for your organization

  • Desired end result: Identify the most suitable growth option

  • Users: Consultants, strategists

  • Cognitive activity: Decide


2. Conceptual Model

With the purpose defined, we can move to identifying and organizing the key concepts that will make up a framework. This structural modeling of content has the goal to detect, extract, and simplify underlying patterns. It can be broken down into a three-step, iterative cycle: 

  • Collect

  • Zoom in and out

  • Organize. 



Collect

The goal is to identify the existing key concepts in the domain of interest. This information comes in all kinds of types: explicit information from books and documents, heuristics that aren’t documented but communicated within a culture, and personal experience, to name a few. This is essentially a big dump of information; similar to a brainstorming. The output of this work is a content library. 


Your subject domain is determined by the purpose you defined previously. As you can imagine, it’s easy to drown in information overload during this phase. Without expertise in a domain, where do you even start? The key skill here is the art of questioning. I developed a template that helps me familiarize with new topics. 



You can use GenAI, too! Use it to brainstorm key concepts for your framework. Describe your domain of interest and ask for help in identifying relevant concepts, heuristics, and personal experiences that could contribute to your framework. For example: 

‘I’m a [role] and want to design a framework focused on [domain]. The framework should help [users] to [desired outcomes]. Can you suggest key concepts and heuristics that should be included?’

Zoom in and out 

Too often, we look at frameworks on an abstract level, and this can cause us to lose sight of practical experiences. Or we go the other way: we focus only on real life experiences and miss the larger dynamics at play. The key is learning to zoom in and out between these different perspectives to get a more holistic view.


Laying out all this messy information visually helps us more easily navigate between the different levels and elements that will make up our framework. I like to map my conceptual models visually:


  1. Start with the goal (stage 1) of your framework in the center: What must the framework accomplish?

  2. Then, based on your research, radiate outwards with essential elements and practices. Nestle sub-concepts that serve upper-level concepts and build a tree diagram, using hierarchical and/or procedural deconstruction.

  3. As I engage in this activity, I zoom in and out between elements and add new ones: What is this made of? Why does this matter? Why else? What purpose does this element serve? What’s my personal bias? Where can I intervene? What is a practical example of this?



This stage of framework design requires a good dose of critical thinking. It may not be the best idea to ask GenAI to assist you to do the actual leg work of zooming in and out. Rather, use it to augment your critical thinking with reflection prompts to help you consider your assumptions, biases, and overlooked areas. For example: 

‘Here is a screenshot of my concept map on [topic, goal]. Zoom in on cluster [area]. What are some alternative perspectives or considerations? How might different stakeholders view this cluster?’ 
‘Cluster [area] is overly complex. Please help me think of an analogy or metaphor to explain this to [audience].’ 
‘Cluster [area] and [area] seem quite similar. How do they differ in purpose or audience?’ 

Organize 

As you zoomed in and out between the different layers and elements, you enriched them with new ideas. Now, you can discover new structures among these elements and transform your mind map into a conceptual model.


To add meaning to the mess, consider the two key ingredients of each framework: building blocks and relationships:


  1. Building blocks: The essential components or elements that provide structure to the framework. They will become the steps, categories, levels, functions, or entities or your framework. Use affinity mapping to group elements into clusters based on shared attributes. Play with the emerging material, use colors to create categories, and give each cluster a theme. These clusters become the initial building blocks or your framework.

  2. Relationships: The connections, interactions, or dependencies between the building blocks within the framework. These connections can be hierarchical, parallel, reciprocal, or other types of associations that define how the building blocks work together to achieve the desired outcomes. 



Seek GenAI’s assistance in organizing your framework’s building blocks and relationships. Share the elements and potential clusters you’ve identified and ask for advice on structuring them into a coherent model. For example: 

‘What patterns do you observe in my concept map? What names would you give those patterns?’ 
‘Are the categories and their elements MECE? Elaborate your reasoning.’ 
‘What hidden or non-obvious relationships do you observe that might affect the framework’s effectiveness?’ 

3. Visual Form 

In this last stage, we transform the initial concept map into a visual form, so we can interact and engage with it according to its purpose.


From a variety of possible visualizations to choose from, which one will facilitate the cognitive tasks at hand? This is where we go back to stage 1, “Purpose” and review the cognitive activity required from users to achieve the desired outcome. The choice of cognitive activity influences how the framework is formed.


For example, in the Ansoff matrix, the cognitive activity is to “compare” and “decide” between different growth options . This informs the organizing principle: the underlying structure information should take to promote the cognitive activity. We can then choose between different forms to visualize the organizing principle. To compare two or more options, a 2x2 matrix is a great starting point.


Consult GenAI on choosing the best visual form for your framework. For example: 

‘Based on the building blocks and elements I identified, what visual form would best support the cognitive activity of [xzy]?’
‘How does this visual form facilitate the cognitive activities required by the framework’s purpose?’
‘Please make suggestions what additional tactics might be required to facilitate communication and collaboration among users of the framework.’

Notes on the Design Process

In practice, all design is a non-linear process. Learning Designers go back and forth from one aspect to another. It is often natural to begin with existing ideas instead of a blank template. It doesn’t matter where you enter the design process, it only matters that you end with a coherent framework.


For example, you might be interested in exploring some results you observed. You look at the result as though it was the outcome of some unknown framework. You could then speculate about the framework that might have produced such a result. You then develop the actual framework, using the principles presented in this article. And finally, you stress-test the framework using real-world data and critical thinking to deduce other results.


Said differently, it is more important to test against the stages I discussed today, than to think of the process as a step-by-step process in which you don’t need to ever look back. 


[1] There are great collections of frameworks, such as The Decision Book or The Great Mental Models. Yet, little has been written about how to develop these frameworks. I found the following two resources most useful:

 

If you read this far, thank you for your attention. This post is part of a 5-part series that will also be able to download as an e-book. Join my mailing list to get notified when it becomes available.

bottom of page