Taming my Chaotic Mind with Tana
I began my productivity journey in 2004 whilst in college when I discovered David Allen’s Getting Things Done. Since then, I have remained fascinated by personal productivity systems and, more recently, personal knowledge management (PKM). As someone who is often balancing many projects and information flows, finding the best methodology and set of tools has been an unending quest. This article shares my journey so far and introducing a tool that I believe offers great potential. If you have struggled to get and stay organized, some of this may resonate with you.
Despite having explored these systems and tools for almost two decades, I have enjoyed little success in getting something to stick. Often I will implement a new system or tool, use it religiously for a few weeks, then realize I’ve let if fall by the wayside. Even on days when I commit to working in time blocks, I rarely make it a full day before become derailed or distracted by urgent items.
In 2019 I was officially diagnosed with ADHD. I never had suspected that I might have it because I was relatively successful in school and worked successfully in jobs that required focus. At the same time, I began finding areas where I struggled to stay organized and manage routine processes. As I began researching the symptoms of ADHD, I started to see a clear connection. Finally, at the encouragement of my wife, I went to the doctor and took an official assessment. I don’t think I am as impacted as strongly by ADHD as many but I can now see where it creates challenges for organization and discipline. This realization was quite freeing and provided a framework from which to explore solutions.
Finding the Right Tools
Part of my journey to productivity and knowledge management has been finding the best tools to support my work. I’ve tried a few analog tools such as a bullet journal but find that digital is still ideal. For managing tasks and projects, I’ve tried all the major options from simple task lists in Outlook to complex tools such as OmniFocus and Clickup. Often, I would end up with endless lists of tasks that become too overwhelming to manage. More recently I had used Akiflow which merges task lists from various sources and also integrates with your calendars. Akiflow’s design made it easy to capture tasks and plan out when to complete them through time blocking. However, Akiflow still required the conscious effort of adding and updating tasks which is where I struggled. Tasks often would arise during meetings so I would have to remember to review the notes and process action items.
In terms of knowledge management (PKM), I had long been a user of Evernote. Evernote was a great system for tracking lots of notes but as my notebooks grew, it became increasingly harder to manage the overall structure and make use of the information. A couple of years ago, I came across Obsidian. Obsidian uses locally-stored Markdown files to store information and offers a huge ecosystem of plugins. The open canvas was highly customizable and flexible. The customization become one of the main challenges because I was always finding recommendations of different plugins and setups. It was hard to establish a system that really worked out of the box. What Obsidian did really well, though, was linking of notes to one another. I had recently discovered the Zettelkasten method of knowledge management and Obsidian worked well to create a linked set of notes.
In using Obsidian, I discovered a friction point I hadn’t quite put my finger on. When I wanted to add a new piece of information to my notes, I found myself stopping to decide exactly where to store that information. More importantly, deciding what constituted a new note file versus an addition to an existing file. What resulted was a mashup of notes that were either very atomic (only one thought) and quite long with several layers of subheadings. While the system was more useful than Evernote, it still made it hard to both quickly capture and then quickly find information as I worked.
I had tried other popular products such as Roam Research but they didn’t click at the time. Roam uses a model some refer to as an infinite outline. Everything in one’s notes are just bullets and sub-bullets, creating a giant knowledge tree. This solved the question of what should be a note versus a sub-heading but I still wasn’t sure how best to implement the approach.
Introducing Tana
I became aware of a new entry into the PKM field around October of 2022. Tana, the “Everything OS”, is a new tool currently still in early access. There were several aspects of the platform that intrigued me and I finally got my early access in December. Since starting to build out my systems in Tana, I think I may have finally found the tool to meet all my needs.
Tana falls into the infinite outliner class of tools. All data in the system exists as “nodes” within a hierarchy in either your personal library or a shared workspace. Nodes are mostly plain text (no formatting beyond bold or italic) or files such as images, PDFs, or audio clips. Nodes can also have fields associated with them that act as metadata or attributes. For instance, a task node might have a status field and a due date field.
Supertags
One differentiating factor of Tana is the Supertag. Like many systems, you can add a tag such as “#task” or “#book”. But tags can have predefined fields and can be inherited from other tags. By applying tags, nodes now become intelligent objects. This enables Tana to work as a database but where the records are stored throughout the system versus in a tables of related items.
What the tagging approach does for me is allow me to store different types of information in context of my work. For instance, when in a meeting I can take notes as bullets. When an action item comes up, I simply tag that line as a #todo. Or I can take an item as a #decision, if I wanted to track a major decision. I can also mention a specific individual as a #person and tie together any references to that person in one place.
Search Nodes
Another powerful feature of Tana is that a node can be used to display other nodes based on a set query. So I can have a node that lists any outstanding tasks for specific project. Or a node that finds any books I’ve mentioned anywhere in the system. This solved my major pain point of discoverability. I no longer have to worry about how and where a note was stored. All I have to do is tag it and perhaps add a field or two. I know with a simple query I can find whatever I need.
The central workspace in Tana is the Day node. Each day is a dedicated node that you can customize however you wish. My day node includes my agenda for the day to quickly take meeting notes and a series of task search nodes to surface things I need to do. As I work during the day, I just add thoughts, action items, and anything else to my day node. They will live on that day node unless I move them elsewhere. However, I rarely need to move them because search nodes will find them when and wherever needed. Finally, groups of nodes can be displayed as a list, a set of cards, a table, tabs, or on a calendar. This allows you to organize information in many ways. The calendar view, a newer addition, now makes it easy to schedule out tasks or use timeboxing during a single day.
Evolving Features
Tana is very much in heavy development and releasing new features regularly. In the past couple of months they have introduced two powerful features: commands and AI integration.
Commands are defined actions that can be as simple as filling in a field or as complex as calling an external API, processing the data, and returning to the system. As an experiment, I built a command that connected to my work CRM to retrieve records needing review. It would then add the results as a list on my current day for further processing.
As with many products, Tana has added integration with OpenAI. However, Tana has taken it a step further and built AI into the core of the system versus just an add-on tool that can generate content. Tana can send the context of a node along with a prompt to OpenAI and then parse the result. It also uses AI to suggest relevant fields for a Supertag and retrieve data. For example, say I had a list of restaurants. I could have a field of “cuisine” on each one. Running one AI command would attempt to fill in that field for each entry. AI can also be used to transcribe and summarize audio notes. This just scratches the surface on what I’ve seen people build with AI and Tana.
There is so much more that Tana can do. But these are the features that I have found to be a game changer in my work. Using Tana, I now have a powerful system that can manage both my projects/tasks and my personal knowledge tree. I still have a long way to go in creating a lasting project and task system. Maintaining the system requires strong habits and intentional upkeep. But I’m hopeful that in Tana I have found a tool that will make it easy enough to quickly capture information, find it when I need it, and to maintain the information through regular reviews.
