Contents
A template can save me from forgetting the source. It can also give me twelve empty headings to feel guilty about. I would judge it by the note it helps me write, not by the amount of structure it displays before I have had a thought. The best starting point is a task small enough to demonstrate.
For this article, that task is a constructed project note about choosing a format for a local reading group's handout. There is no real client or measured outcome. I will build a template that keeps the question and its supporting material visible, then show when I would change it.
Choose a note type with a clear boundary
I would begin with “project decision note,” rather than “all-purpose knowledge template.” The first has a recognizable job: explain a choice, its basis and its next check. A reading log, meeting record and source summary have different jobs. Combining them can make every note carry fields that belong to someone else's task.
For the imaginary handout, I need the audience, the choice under consideration and the evidence that would affect it. I do not yet need a rating system, a permanent identifier and a dozen metadata fields. I can add structure when a repeated problem gives it a reason to exist.
Keep the prompts close to the work
My opening prompt would ask what decision is being made. The second would ask what is currently known. A third would separate uncertainty from a conclusion. That sequence reflects the thinking I want to preserve. It also makes an unfinished note intelligible: the uncertainty has a place without being disguised as a finished answer.
A prompt should invite a response in ordinary language. “What would change this decision?” is more useful to this note than “confidence score,” unless a project already has a meaningful scoring method. An attractive number can feel precise while leaving the basis of the judgment unexplained.
Make the first version small
Here is the complete starting template I would use. It is intentionally plain text, so it can be copied into a document, a Markdown note or a paper notebook. It does not depend on a third-party plugin or a particular paid plan.
For the demonstration project, I might write that the handout needs to be readable at a shared table, that a sample layout is still needed and that the reading group must choose between two formats. I would not fill the source section with invented user feedback merely to make every heading look complete.
# Decision or question ## What this is for Audience and next use: ## What I know Source or observation, with a route back: ## What remains uncertain What would change the decision? ## My current view A reasoned choice, with its limits: ## Next step Action, owner if relevant, and stopping point:
Use the tool without letting it decide the content
Obsidian's core Templates plugin inserts predefined text and supports title, date and time variables. That can reduce repeated typing. It does not establish whether the note has a meaningful question or whether a source supports the answer. I would keep those judgments in the writing process.
If I used properties, I would add only fields I expect to filter or retrieve. The current documentation describes properties as structured data stored at the beginning of a note. A project name and review status might earn a place. A mandatory field I ignore every time is evidence that the template needs attention.
Try an awkward note on purpose
A useful test is a decision with no external source yet. The handout note might begin with a preference and a question for the group. I would record those as such. The template should allow “not checked yet” without encouraging me to dress an assumption up as evidence. Blank space can be honest information.
Another test is a note with no next action. Perhaps the group chose a format and the record simply needs to remain understandable. I would write “No current action; retain as decision context” rather than manufacture a recurring task. A template should support closure as well as activity.
Revise the scaffolding without rewriting the past
After a few notes, I would look for repeated friction. If I keep putting the audience in the title because I forget the audience prompt, that suggests a change. If a field never helps me retrieve or interpret anything, I would remove it from the next template rather than expand it to look more complete.
I would leave older notes alone unless there is a reason to migrate them. A new template is not proof that previous thinking was invalid. When a change affects meaning—such as a status field acquiring a new definition—I would record the change so that an old note does not silently become a new kind of statement.
Fill the template once before improving its design
I would use the fictional reading-group handout to fill every prompt in plain language. The question is which format suits the next discussion. The audience is the people using the handout at a shared table. What I know may include the group's stated purpose and the source material to be discussed. What remains uncertain may include preferred text size or whether participants want space for notes. These are questions for the group, not findings I can invent to make a demonstration look complete.
My current view might be to prepare two short sample layouts for review. That is a proposed next step, not an approval to produce a final handout. I would make the distinction explicit in the note. A good template helps me express a provisional position without making it sound weak or disguising it as settled fact. The useful outcome is a reader who understands why I would try the samples and which response would affect the final choice.
I would then look for a field that tempted me to repeat myself. If “What this is for” and “Decision or question” have become identical paragraphs, I might make one a single-line audience prompt. If “Next step” contains a complete restatement of the uncertainty section, I could shorten it to the specific action. Filling the template shows where structure creates value and where it merely asks the writer to say the same thing in a different place.
Only after that inspection would I adjust formatting. A visually balanced empty template can conceal a poor sequence of questions. In contrast, an awkward-looking filled example can reveal a useful design change. I would keep the filled demonstration beside the reusable text while testing the revision. That makes it possible to compare what the new prompts produce, rather than judge the template solely by how attractive the headings look before there is anything under them.
Try a second example that refuses to fit
The second test would be an ordinary reading note rather than a decision. Imagine I have read an essay about how public libraries describe their collections. I want to remember a distinction and a source location, but have no project or action in mind. If the template forces me to supply an owner, deadline and recommendation, it is asking for information that the note does not need. I would allow the note's purpose to determine its structure.
For this reading example, I might retain only the question, source pointer and my response. The reusable project template can remain unchanged for project work. This is a useful reason to have a small family of templates rather than one universal form. The family should stay small enough that selecting a starting point is easier than writing the note. If choosing between templates becomes another difficult classification task, I would simplify the options before adding more.
A third awkward case is a closed decision. The reading group chose the handout format, no further action is due and the note needs to explain the choice. I would replace a live next-action prompt with a closure line identifying the agreed result and its conditions. The source section remains relevant because it helps interpret the decision later. An empty task field should not imply that the record is unfinished when the appropriate state is reference.
These tests separate reusable prompts from mandatory requirements. A template can suggest a useful question without insisting every note answer it. I would include a short instruction such as “omit sections that do not serve this note” if that helps the intended reader. The instruction should be visible where people use the template, not buried in a separate manual. A little permission to adapt can prevent a simple tool from becoming a form that rewards invented completeness.
Decide which information deserves a property
Properties are useful when I expect to search, sort or filter a consistent kind of information. For the handout project, a project identifier or review state may qualify. A nuanced explanation of why the layout matters belongs in prose. I would not put the entire argument into a property just because structured data feels easier to organize. The distinction is about the future use of the field, not its apparent importance to the writer.
I would test a proposed property with a retrieval question. “Show the notes awaiting a group decision” gives a review-state field a clear job. “Capture everything about the note” does not. A field with no foreseeable use adds a maintenance obligation: someone must choose its value, remember its definition and update it when the work changes. I would make that obligation worthwhile before adding the field to every new note.
The plain-text version should remain understandable without the application's special view. If a property contains an abbreviated state, I would explain it somewhere stable and use the same meaning across the collection. A date should say whether it refers to creation, a meeting or a decision. Different meanings compressed into one field can produce a beautifully sorted list that answers the wrong question. I would rather have fewer fields whose interpretation survives export and handover.
I would also consider the reader's privacy context. A demonstration template does not need real names, private meeting details or actual account identifiers. When sharing a filled example, I would replace sensitive content with clearly illustrative material and check links as well as visible text. A reusable template can travel far beyond the project that created it. Its example should teach the structure without accidentally carrying information that belonged only in the original working environment.
Maintain the template as a small shared tool
If several people use the template, I would ask what repeatedly causes confusion. An answer such as “we do not know whether this field is an opinion or a confirmed decision” points to a meaningful design problem. A preference about heading color may matter less to the note's function. I would prioritize changes that clarify the work and make a brief explanation of any new field meaning available to the group.
I would avoid silently changing a property's meaning while retaining its name. Suppose “reviewed” originally means that a writer checked a source, but later means that the group approved the final document. Those are different states. Reusing the word can make older notes appear to have received an approval that never happened. I would choose distinct labels or document a careful migration, rather than rely on everyone remembering when the convention changed.
A useful revision test is to take an old note and a new note and ask the same retrieval question of both. Can I tell what decision each records, which source supports it and whether a next action remains? They need not share identical headings to be understandable. Consistency should help interpretation; it should not force a wholesale rewrite of valid historical records just to make the folder look uniform.
I would finish template maintenance with a small release note: what changed, why and who needs to use the new version. A personal collection may need only one sentence. A shared project may need a short conversation. In either case, the template should return to being an aid to writing rather than a permanent redesign project. I would keep it when it saves a meaningful decision and change it when actual use shows that the current prompts no longer fit.
Save a filled example beside the reusable prompts
I would retain the demonstration decision note as an example of how the short template can be used, with its fictional project clearly identified. The reusable text and the filled example have different roles. One provides prompts; the other shows how evidence, interpretation and an open question can coexist without every heading receiving the same amount of writing.
A new reader should be able to adapt the example without copying its conclusions. I would point out which fields reflect the particular handout decision and which questions may help another project. The example should not become a hidden standard that forces every note to look equally complete. A meeting record or a reference note can remain shorter when the additional structure serves no actual use.
The last check is whether the template makes a useful note easier to write and return to. If filling it requires inventing deadlines, assigning unaccepted responsibility or disguising a preference as evidence, I would change the prompts. A small reusable tool should make the work more honest and legible. It earns continued use when the next reader can identify the question, understand the current view and find the material needed to inspect it.
Frequently Asked Questions
Q: Should every note use the same template?
A: No. A shared starting structure can help within one note type, but a meeting record and a source summary serve different tasks. I would keep a small set of templates with clear purposes rather than force everything into one form.
Q: Do I need an Obsidian plugin?
A: You can use the plain-text template in many tools. The documented core Templates plugin is one way to insert it in Obsidian. The workflow here does not require additional community plugins.
Q: Should I automatically add a date?
A: Add it when the date helps interpret the note. Creation, meeting and decision dates can mean different things. Give a date a clear label instead of assuming every timestamp answers the same question.
Q: What if most fields stay empty?
A: Check whether the template fits the task. Some blanks honestly mark unknowns, but repeated irrelevant fields may indicate an overbuilt design. Remove fields that do not help the next reader.
Q: Can I share this template with a team?
A: Yes, but agree on field meanings first, especially statuses and responsibility. Use fictional or sanitized examples when demonstrating it, and follow your organization’s rules for confidential information.
One thing to take forward
I would keep the smallest template that preserves the question, the basis of a decision and its open edge. The template supplies a place to think; it should not pretend that filling every space is the same as reaching a sound conclusion.
Sources & further reading
Something needs correcting? Write to the editors with the article link.



