A step-by-step guide to context charting with common tools

In an era where artificial intelligence significantly influences how we gather, analyze, and employ information, a new method emerges, reshaping the landscape of research, note-taking, and content creation: context charting. This method is particularly advantageous for content writers, researchers, and entrepreneurs who seek to maintain an edge in their domains by synthesizing ever-expanding information streams into coherent narratives or actionable insights. Context charting offers a structured approach to visually represent and explore the complex web of ideas, facts, information sources, and their interconnections, much like a map guides a journey through unfamiliar terrain.

The basic elements of context charting

Whiteboard

The central point of context charting is the whiteboard, where you gather and organize all items in a context map. An important rule about the whiteboard is that there are no rules except for one - every item should have a type assigned andthe types should be distinguished by colors, shapes or icons. The structure of the whiteboard is up to you - whatever works, use it - link items, put them in lists, use colors... Or just keep everything in random chaos if that suits you.

Item types

An item can have many forms: a single word, phrase, sentence, or whole paragraph. It can be a link or reference to an external article, an image, video, or other file attached to the map.

Regardless of its form, whenever you add an item to the map, you should assign it a role it will have in the map's context.

Item types are essential not just for you but also for anyone else who will work with the context map. That someone will often be an AI assistant for whom item types are fundamental.

The exact set of item types may depend on the use case. Here is a set designed for content creation, which can, however, fit most other use cases:

• Topic: something you want to talk about in the content. It can also be a fact or an idea.
• Keyword: a word or phrase you want to mention in the content.
• Question: something you want to answer.
• Heading: a part of the content structure you want to create.
• Article: any external source you want to cite or use - it can be a web page, social media post, video, PDF...
• Note: (or stickie) metadata, notes and comments about the items on your map. Everything that is not supposed to be a direct part of the context and the resulting content.

Search & AI sources

Even though we all already use web search and AI assistants whenever possible, they are an explicit key component of context charting because you can't create a proper context without checking existing information.

There are two ways or directions of using external sources for context charting: broadening the scope and deepening the understanding. You can use web search, AI assistants, or specialized tools like keyword research tools to find related topics and keywords and include those in your context map if it makes it more comprehensive.

Then, you can research existing content (or consult AI) to learn about topics in the context you know less about or find supporting evidence for your claims coming from your expertise.

A step-by-step guide to context charting

Let’s show context charting on a simple example. Let’s say I am a Lord of the Rings fan and I am interested in Gandalf and want to write an article about him on my blog.

My toolbox

I’ll be using Miro for the whiteboard, Copilot as AI and classic Google for search. In Miro, I set up a “palette” for the item types as follows:

Step 1: Define your focus question and goal

The initial step in context charting is clearly articulating the focus question you aim to explore. This question serves as the foundation of your map, guiding the collection and organization of related ideas, topics, and sources. It should be based on the overall goal you are aiming at.

In our example, the focus question will be “What was Gandalf's backstory and motivation?” I chose it because I know all about what Gandalf did in the movie, but I know little of where he came from or why he did all that. My example goal is writing a blog post for my fellow Tolkien fans.

Step 2: Dump all you know - related to the question

I begin by gathering topics, keywords and questions related to my focus question onto the whiteboard. This brainstorming phase captures the initial breadth of your knowledge.

Why?

It is important to ground any new knowledge in the context of what we already know. By linking new concepts or facts to something we know we encourage building interconnected knowledge structure in our brain, where we actually understand the new things instead of just memorizing some facts or definitions. Speaking of memorizing, linking new facts to what we know ensures we remember it much longer than disconnected facts we just memorized. This is all coming from the theory of meaningful learning by David Ausubel.

Leaving the hardcore science stuff aside, I am sure we can all agree that if I am to write an article about some events and characters from Gandalf’s past, I need to put them in the context of what my audience knows, introduce them properly. Doing the brain dump in the beginning helps gather items to relate to. And, and that’s equally important, it gives me some starting points on what to research, where to dig deeper.

When to stop?

Even with a simpler question, the brain dump could go on forever. We can always come up with more questions or related topics. After all, everything is related to everything else, somehow. The important thing to keep in mind is that we are gathering something to start with, not everything. Here are some pointers:

It does not have to be perfect

That is true about the whole context charting process - the important thing is the process and what happens in your head during it, not the actual result that you see on the whiteboard. Although, of course, the whiteboard can be useful for sharing and collaboration, that’s not the main point. The main point is that it helps you think, backtrack and forward your research, and structure your output - which can be an article or a video which you create based on the whiteboard.

‍3 is too few, 30 is too much

If you want to hear some numbers, we can learn from concept mapping theory, which says, based on tens of years of research, that you shouldn’t have more than 25 items on the whiteboard - because than it becomes cluttered and hard to read.

Depends on the goal

‍As with everything, it depends on what is the end goal. In my case, it’s just a short blog post. If it were a PhD thesis, I would spend much more time here of course.

My dump

Here is my dump of Gandalf related stuff. My knowledge is based on the two movie trilogies, so that’s what I put on the whiteboard:

Step 3: Expand with suggestions and search

One of the reasons I was not trying to make the initial dump complete is that I am going to use external help now in the second step, which is enriching your map by using external sources to surface additional topics, keywords, and questions. This is crucial for expanding the horizon of your research and ensuring that your map covers the whole context. The basic idea is to retrieve suggestions of related items that might fit your context map, browse them, and assess if they actually fit your map. If they fit, add them with a proper item type set. The only other important thing is to stop expanding your context map at some point. Your focus question helps with that - if your context map is big enough to answer the question with respect to your audience or use case, you are good to go.

The suggestion sources you use depend on your use case, but there are some general ones that fit any use case.

The obvious is using your favorite GPT AI chat tool. You can give them your focus question and ask for related topics or keywords.

Another obvious source is Google. Although it's designed more for answering questions than looking for related topics, you can still use its People also Ask feature. And, of course, you can find related topics by actually reading the content found by Google, but that's time-consuming.

A more old-school but still valid source is Wikipedia. Thanks to it's comprehensive linking, it allows you to start from the core topic related to your question and checking all the outgoing links that might be relevant in your context.

Then, there are specialized tools, especially keyword research tools like SemRUSH, AHREFs, or SpyFu. Most of them allow searching for related keywords, and some even search for generalized related topics.

In my case, I am using Copilot. I will give it my initial dump as the context and ask it to suggest what to add - using the item types.

So first, I ask for related topics:

Notice I enclose the context in the <context> tags to ensure Copilot understands it properly, and I also structure it according the item types - that helps it understand as well.

Now I use my intuition, my focus question and the knowledge of my audience to decide what to add to the whiteboard. But still keeping in mind the process is the essential thing here, so I am not overthinking the decisions. I can always decide to delete or move items aside on the whiteboard and, e.g., leave them for another blog post.

I continue in a similar way with questions and keywords:

Of course, if I were concerned with SEO, I would use some keyword research tool and spend much more time on it. In my case, keywords are just a good tool to capture important characters, places and events.

After going through all the suggestions, here is how my context map looks like:

Notice that I did not care much about the layout yet, I just put things I considered closely related close to each other.

Organize items whenever needed

Organize the items on your whiteboard by clustering closely related ideas. This step is pivotal in uncovering hidden relationships and starting to bring structure to the map. Aim to identify and name groups of related items. That will help you identify patterns and help you with further understanding. If an important relationship spans across distant groups, represent it with a link.

You will probably organize your context map iteratively throughout the context expansion and research phases. Every time you feel the map became too cluttered, spend some time organizing it.

However, remember that the amount of organization and structuring depends on your thinking style - you don't need to organize much if you don't like or need it.

There is only one actually important rule for organizing items: keep more useful items separate from the less useful ones. As you build your context map, you will usually gather more items than needed and you will need to decide which of them you will use in your current content piece (or current learning session, current project or whatever your use case is) and which of them you leave for later. A simple way is to drag the "in use" items to the right side of the whiteboard and leave the rest on the left side.

Here is my map after I did some organizing:

You see I did not move anything aside yet, because I don’t have that many items here. I just grouped the Hobbit and LOTR items together. Another group that emerged is the Valar and Istari stuff which I don’t know much about yet.

Search for sources

At this stage, integrate sources—articles, blogs, podcasts, videos, etc.—into your context map. Each source should be represented as a unique item linked to the related topics or questions it elaborates upon. Include the search queries used to find these sources to maintain the research path. It is important to be able to back-track and prevent redundant searches.

There are two main goals of finding sources - to gain an understanding of the topics you need to answer the focus question and to support your own claims.

The context map suggests you what to search for

You can look at the items on your context map as a set of possible search queries. Start with the most intriguing ones. If you find a resource that answers or covers multiple items, reorganize your map accordingly. It is simple but extremely powerful feature of context map - it can guide and track your research at the same time.

Extract Key Messages

Optionally, as you integrate each source, extract key messages pertinent to your focus question and add these as separate items on your map. This distillation of insights ensures that the map captures information and crystallizes significant findings that directly inform your original focus.

In my case, the main goal of the research is to learn about the things I don’t know much about yet. So I can start with the questions and simply google them:

I’ll link each article I find to the item I used as the search query, and add a citation of facts that I find useful for my focus question:

After doing the same with another question, I actually have enough material for my focus question:

More Arrangement and Linkage

Finally, refine the arrangement of items on your map. Reevaluate the connections and ensure that closely related items are grouped effectively while less relevant items are kept floating around the central frame. Use links to underscore relationships across groups, emphasizing interconnectedness and thematic resonance between disparate elements.

By maintaining this network of relationships, context charting enables a more nuanced understanding of the material. It builds a framework where both qualitative insights and quantitative data are equally valued, bridging the gap between them in a coherent visualization.

Here is my context map after some linking:

Notice that these few links are enough to cover the most important thing - that the key thing for Gandalf’s motivation is fighting Sauron and he started that as a member of a wizard order.

Step 5: reducing the map to fit your goal

The process of building a context map is the goal itself because as you work on the map, you organize your thoughts, learn, and find sources. But there are many possible next steps.

Sharing and collaboration

The obvious is sharing your map for comments or collaboration. You can also use your context map to build an outline or a whole content plan. Or you can feed it to AI and ask it to do it (or anything else) for you.

Feeding the map to AI assistants

The context map can serve as a context for AI LLM models, provided you create the map using a tool that supports easy export or integration with GPT chatbots. The beauty of this is the AI does not care much about how well-organized your context map is. It will make sense of it on its own and then perform your task being perfectly informed, with the right context. The task can be content creation or just outlining content, creating a summary of the most important facts, a list of FAQs, or building a study guide...

Reducing my example map to fit my goal

In my case, the next step is drafting a blog post. To do that, I create a copy of my context map in Miro, so that I don’t lose any information. The original map can serve for aditional articles in the future. In the new map, I select those items that I want to use in my article. I drag them aside and order them roughly to form a preliminary flow of thoughts.

Now I can use this map to create a prompt for Copilot:

Notice that I don’t include the articles nor the citations into the prompt. I’ll only use them to fact check the Copilot’s output and to edit the article to align it with my voice and my audience. I could of course include some more instructions in the prompt about the desired length, tone, goal, audience etc. But let’s keep it simple for the sake of this example.

Final thoughts

We went through all the key context charting steps, following an example of writing a blog post. You can follow the same process to learn something new, building a study guide, to develop and map a new content strategy, or to brainstorm a new product for your company. Similarly as with mind mapping, the method is universal and the possibilities are endless.

I have shown the example using common tools. However, there is an obvious option for more automation and combining the tools together. Consider using specialized context charting tools like Contextminds that makes the whole process much more seamless and faster.