Education FuturesIntroducing KnowMapped: A platform for mapping knowledge
KnowMapped began as an internal tool at Education Futures. We used it to make relationships visible during workshops, presentations, and research activities, especially when a conventional outline or slide deck could not capture the complexity of the ideas in front of us. If you have attended one of my keynotes or workshops, you may already have seen an early version in use. We sometimes built maps live with participants, and in other cases we fed workshop notes or collected data into the system to produce a quick visual representation for discussion.
What started as a practical internal solution became more useful over time. We used it to organize research, test relationships among ideas, visualize complex arguments, and turn large amounts of information into structures that were easier to inspect. As our own use expanded, so did the tool. We added better layouts, presentation controls, sharing options, import and export capabilities, and eventually AI-assisted map generation. At some point, the internal tool had become substantial enough that it made sense to develop it as a public platform.
That platform is now KnowMapped.
At its core, KnowMapped is built around knowledge maps rather than conventional mind maps. Both use visual structures to organize ideas, but they serve different purposes. A mind map typically begins with a central idea and branches outward in a hierarchy. A knowledge map can work without a central node and can represent relationships in many directions. One concept may connect to several others, relationships can cross between clusters, and the overall structure can resemble a network rather than a tree.
This distinction is useful when working with subjects that resist simple hierarchies. Research questions, policy issues, systems, theoretical frameworks, workshop discussions, and complex documents often contain overlapping causes, dependencies, tensions, and feedback loops. A knowledge map can make those connections visible without forcing them into a single branching structure.
| Approach | Typical structure | Best suited for | How KnowMapped differs |
|---|---|---|---|
| Mind maps | A central topic with branches and sub-branches | Brainstorming, outlining, organizing ideas around a primary theme | KnowMapped does not require a central node or a fixed hierarchy. Concepts can connect across the map in multiple directions. |
| Flowcharts and diagrams | Processes, sequences, or manually arranged visual structures | Workflows, procedures, and defined systems | KnowMapped focuses on concepts and explicit relationships, allowing the same knowledge structure to be reorganized through different layouts without rebuilding the underlying map. |
| Digital whiteboards | Free-form visual canvases | Collaboration, sketching, brainstorming, and visual facilitation | KnowMapped gives concepts and relationships an underlying structure, making the map more than a collection of objects placed on a canvas. |
| AI-generated diagrams | A visualization produced from a prompt or source | Rapid first drafts and automated summaries | AI generation is optional in KnowMapped. Generated maps remain fully editable, and users can also build maps from scratch without using AI. |
| KnowMapped | A network of concepts and defined relationships | Research, learning, systems thinking, knowledge organization, workshops, and communicating complex ideas | The map remains a portable, editable knowledge structure that can be analyzed, redesigned, shared, embedded, and exported. |
AI-assisted generation is built into KnowMapped, but it is optional. A user can upload a source document and ask the system to identify concepts and relationships, producing a first version of a map that can then be reviewed and revised. The generated structure is not treated as authoritative. Users can change concepts, remove or add nodes, rewrite relationships, reorganize the map, and refine the visual presentation.
The same platform can also be used without AI. Users can start from an empty map and build the entire structure manually. In many cases, that process is itself useful because deciding which concepts belong in the map and how they relate to one another is part of the analytical work. We therefore designed AI generation as one possible entry point rather than as the defining feature of the platform.
| Capability | What it provides |
|---|---|
| Manual knowledge mapping | Build maps from scratch by creating concepts and defining relationships among them. |
| AI-assisted generation | Generate a first-pass knowledge map from source material, then inspect, revise, expand, or reorganize it. |
| Flexible map structures | Create hierarchical, networked, clustered, or non-centralized maps without forcing ideas into a single branching structure. |
| Multiple layouts and visual systems | Change layout, typography, themes, density, and presentation while preserving the underlying concepts and relationships. |
| Portable data | Export maps in multiple formats and use the portable .know format to retain the underlying map structure. |
| Sharing and publishing | Keep maps private, share them with others, publish them publicly, or embed them in websites and other digital environments. |
| Developer integrations | Use provided embed code and publishing integrations, including Publii and WordPress support. |
| Free and Professional access | Use a generous free tier or unlock additional Professional capabilities through a one-time purchase rather than a recurring subscription. |
| Optional AI credits | Use credits only when generating maps with AI. Generation costs are shown before the map is created. |
| MCP support | Forthcoming support for Professional users will allow compatible AI systems to interact with KnowMapped through defined permissions. |
Portability has also been a priority. KnowMapped supports several export formats and includes its own portable .know format so users can move their work without losing the underlying structure. We wanted to avoid a model in which people invest time building knowledge inside a system and then discover that leaving the platform is difficult. Maps should remain useful outside KnowMapped, and users should retain practical control over the work they create.
That same principle extends to publishing and integration. Maps can remain private, be shared directly, or be embedded in other sites and applications. We provide embed code for controlled use, and integrations are available for publishing environments such as Publii, with WordPress support also part of the platform. This makes it possible to use a map as part of an article, course, workshop site, research project, or presentation rather than treating the map as an isolated artifact.
The visual system has also become an important part of the project. Knowledge maps can become difficult to read as they grow, so KnowMapped includes multiple layouts, themes, typography options, density controls, and presentation settings. The underlying concepts and relationships remain the same while the visual representation can change according to the purpose of the map. A working map used for analysis can later be refined for teaching, presentation, or publication.
We have also tried to keep the pricing model straightforward. KnowMapped has a generous free tier that allows users to create and work with maps without making an immediate purchase. For users who need the expanded feature set, Professional access is available as a one-time unlock rather than a recurring subscription. AI generation uses credits because those operations have direct processing costs, and the cost is shown before generation begins. Users who prefer to build maps manually can use the platform without relying on AI generation at all.
The business model reflects the same emphasis on user control. KnowMapped does not rely on advertising or the sale of user data. We prefer to charge directly for the capabilities that incur real costs and keep the rest of the platform useful without introducing unnecessary forms of lock-in.
One of the next major additions already in testing is support for the Model Context Protocol (MCP) for Professional users. MCP allows compatible AI systems to interact with KnowMapped through defined permissions. This could make it possible, for example, to turn a conversation into a map, locate an existing map related to a topic, or add new concepts and relationships from a research session without manually moving information between applications.
That development follows the same principle that has guided the project from the beginning. KnowMapped is intended to help people work with relationships among ideas, whether those relationships are built manually, generated from a source document, developed collaboratively, or extended through other tools. The map remains the working structure, and the user remains responsible for shaping it.
What began as an internal tool for workshops and research has grown into a platform we now use across a much wider range of activities. We are releasing it because we think others may find the same value in being able to see, question, and reorganize how ideas connect.
You can explore KnowMapped at https://knowmapped.com.
