Core Primitive
Templates for common output types let you start producing immediately.
The most expensive moment in production
The blank page is not a creative challenge. It is an operational bottleneck.
You sit down to write a project update. You know what happened this week. You know the audience. You know the deadline. And yet, for the first seven minutes, you produce nothing. You are not thinking about the content. You are thinking about the container — what sections to include, what order to use, how long each section should be, whether to open with the headline or the context, whether to use bullet points or paragraphs.
These are structural decisions, not content decisions. They consume willpower, burn working memory, and produce nothing visible. By the time you have settled on a format, you have spent your sharpest cognitive minutes on scaffolding that will be invisible to the reader. The actual content — the part that creates value — gets produced by a slightly more tired, slightly more depleted version of you.
This is startup friction. It is the cognitive cost of beginning, and it is the single largest hidden tax on knowledge-worker output.
What templates actually do
A template is a pre-made structural decision. It answers the questions that arise before you start writing content — what sections exist, what order they follow, what kind of information goes where — so that when you open the document, you face prompts instead of blankness.
The primitive here is direct: templates for common output types let you start producing immediately.
Not "more quickly." Immediately. A well-designed template reduces the time between opening a document and writing your first sentence of real content to near zero. There is no structural negotiation. There is no format anxiety. There is only the content, waiting for you to fill in the structure that already exists.
This is not about rigid formatting. It is about removing unnecessary decisions from the moment of production so that your cognitive resources go where they produce value — into the substance, the argument, the insight, the work itself.
The science behind startup friction
The case for templates is not just practical. It rests on well-established cognitive science that explains why the blank page is so reliably paralyzing.
Decision fatigue. Roy Baumeister's research on ego depletion demonstrated that decision-making draws from a limited pool of cognitive resources. Each decision — no matter how small — depletes the pool. The format decisions you make at the start of every output feel trivial, but they are decisions nonetheless, and they cost exactly the same willpower as decisions about content. By the time you have decided on your section structure, your font size, your heading hierarchy, and your opening approach, you have made a dozen decisions that produced zero output. Baumeister's work suggests that this depletion is cumulative across the day. If you produce five outputs and negotiate structure from scratch each time, you have spent sixty decisions on scaffolding before the important decisions even begin.
Friction and behavior design. BJ Fogg, in "Tiny Habits," identifies friction as the primary barrier to desired behavior. The harder it is to start something, the less likely you are to do it — and the relationship is nonlinear. A small reduction in startup friction produces a disproportionately large increase in action. Fogg's framework applies precisely: every structural question you must answer before producing content is a friction point. Templates eliminate those friction points wholesale. The behavior (producing output) becomes easier, so it happens more reliably.
Choice architecture. Richard Thaler and Cass Sunstein, in "Nudge," describe how the structure of a decision environment shapes outcomes. Defaults matter enormously. When the default is a blank page, the default behavior is hesitation. When the default is a pre-structured template, the default behavior is filling in content. You have not changed the person. You have changed the architecture. The output follows.
Working memory load. Baddeley's model of working memory shows that the central executive can only manage a limited number of active tasks. When you are simultaneously deciding structure and generating content, both tasks compete for the same executive resources. Templates offload the structural decisions entirely, freeing the central executive to focus on a single task: producing the content. The result is not just faster output — it is higher quality content, because your full cognitive capacity is directed at the work that matters.
Templates are not constraints — they are launch pads
The most common objection to templates is that they constrain creativity. This is backwards. Templates liberate creativity by removing the least creative part of production.
Consider the sonnet. Fourteen lines. Iambic pentameter. A specific rhyme scheme. These are severe constraints. And yet Shakespeare produced 154 sonnets that remain among the most celebrated creative works in the English language. The form did not limit his creativity. It channeled it. He did not have to decide how long each poem should be, or what rhythm to use, or how to structure the argument. Those decisions were already made. His creative energy went entirely into the language, the imagery, the argument — the parts that matter.
The same principle applies to knowledge work. When you use a template for a weekly status report, you are not being uncreative. You are being strategic about where to deploy creativity. The structure of a status report is not where your insight lives. The content is. The template frees you to put all of your cognitive effort into the content.
The distinction is between constraining templates and liberating templates. A constraining template dictates content — it tells you what to think. A liberating template provides structure — it tells you where to put your thinking. Every template you build should be the second kind.
Templates in practice: four domains
Templates are not a writing trick. They operate in every domain where recurring output types exist.
Professional communication. Amazon's six-page memo template is perhaps the most famous organizational output template in modern business. Jeff Bezos banned PowerPoint in favor of structured narrative memos with a specific format: context, tenets, the detailed proposal, and FAQ. The template does not tell anyone what to think about their proposal. It tells them where to put each element of their thinking. The result is that Amazon's leadership meetings begin with thirty minutes of silent reading, and the quality of proposals — by Amazon's own assessment — is dramatically higher than what the same people produced in slide format.
The military's BLUF (Bottom Line Up Front) format is another canonical example. Military communications follow a strict structure: the conclusion comes first, followed by supporting information in order of priority. This is not bureaucratic rigidity. It is a template designed for the operational reality that the reader may stop reading at any point. The most important information must be first. BLUF turns a structural decision (what goes first?) into a default, which means every military communication starts with the most important sentence rather than the most natural one.
Software engineering. The entire concept of scaffolding and boilerplate in software development is a template system. Tools like Create React App, Yeoman, and cookiecutter exist because programmers discovered the same truth that writers know: starting from a blank file is slower and more error-prone than starting from a proven structure. A React application needs a specific directory structure, specific configuration files, specific dependency declarations. Building these from scratch every time is not principled engineering — it is wasted effort on solved problems. The scaffolding tool generates the structure. The developer fills in the logic.
Professional services. Law firms run on templates. Contract templates, brief templates, memo templates, discovery request templates. This is not laziness. It is quality control. A contract template embeds the structural lessons of hundreds of prior contracts — which clauses are necessary, what order they should appear in, what language has survived judicial scrutiny. A lawyer who starts from a blank page on every contract is not more creative than one who starts from a template. She is more likely to omit a critical clause.
Medical documentation follows the same pattern. The SOAP note (Subjective, Objective, Assessment, Plan) is a template that every physician uses for patient encounters. It ensures no category of information is overlooked. The physician's expertise is in the content — the diagnosis, the treatment plan. The template handles the structure so that expertise is never lost to a formatting decision.
Personal knowledge work. If you produce weekly reports, monthly reviews, meeting notes, project proposals, or any other recurring output, each one should have a template. Not a rigid form. A structural default — the sections that always appear, the prompts that always apply, the format that has proven effective. You should not be reinventing the wheel every Monday when you write your weekly update. You should be filling in the template you built three months ago, the one that captures everything your audience needs in the order they need it.
How to build a personal template library
Building templates is not a weekend project. It is an extraction process that happens over weeks as you notice patterns in your own output.
Step 1: Identify your recurring output types. Look at the last thirty days. What did you produce more than twice? Status reports. Client emails. Code reviews. Meeting summaries. Design briefs. Whatever appears repeatedly is a candidate for a template.
Step 2: Extract the structure from your best examples. Take your three best instances of each recurring output type. Read them side by side. Notice what sections appear in all three. Notice what order they follow. Notice what kind of content goes in each section. The common structure across your best examples is your template.
Step 3: Write the template with prompts, not blanks. A good template does not just provide empty sections. It provides prompts — brief questions or instructions that tell your future self what to put in each section. Instead of "Context:" followed by nothing, write "Context: What does the reader need to know to understand why this matters? One to three sentences." The prompt reduces friction even further by making each section a question to answer rather than a space to fill.
Step 4: Store templates where you start working. A template buried in a folder you never open is not a template. It is an artifact. Templates must be accessible at the moment of creation — when you open a new document, they should be within one or two clicks. Text expansion tools, document management systems, pinned folders, or a dedicated template notebook all work. What matters is that the distance between "I need to produce this output" and "I am looking at the template" is as close to zero as possible.
Step 5: Revise templates after use. Every template is a hypothesis about the optimal structure for a given output type. Some hypotheses are wrong. After using a template three or four times, review it. Did you consistently skip a section? Remove it. Did you consistently add a section that was not in the template? Add it. Did the order feel wrong? Reorder. A template that evolves with your practice is a tool. A template that never changes is a fossil.
The spectrum of template rigidity
Not all templates should be equally structured. The right level of rigidity depends on the output type.
High rigidity — recurring operational outputs. Weekly status reports, invoice summaries, bug reports, meeting minutes. These outputs serve a consistent purpose for a consistent audience. The template should be nearly complete — section headings, formatting, boilerplate text, everything except the variable content. You should be able to produce these outputs on autopilot, because the cognitive load should be near zero.
Medium rigidity — recurring analytical outputs. Project proposals, design reviews, strategic memos. These serve a consistent purpose but require variable structure depending on the specific content. The template should provide a default section structure and prompts, but you should expect to add, remove, or reorder sections for any given instance.
Low rigidity — creative or exploratory outputs. Blog posts, thought pieces, brainstorming documents, exploratory analyses. These outputs vary significantly in structure from one instance to the next. The template should provide a minimal starting point — perhaps three questions to answer, or a simple opening/middle/closing framework — but should not prescribe the structure in detail. Here, the template's job is simply to defeat the blank page, not to define the architecture.
The mistake is applying high-rigidity templates to creative outputs (which produces formulaic work) or low-rigidity templates to operational outputs (which wastes decisions on outputs that should be automatic).
Connection to the creation-editing separation
First drafts are for content final drafts are for quality established that creation and editing are separate phases with different cognitive requirements. Templates supercharge the creation phase specifically.
When you separate creation from editing, the creation pass has one job: get the content out of your head and onto the page as fast as possible. A template accelerates this by removing every non-content decision from the creation pass. You do not decide on structure during creation — the template decided for you. You do not worry about format — the template handled it. You do not negotiate section order — the template already optimized it based on past experience.
The result is that your creation pass becomes purely about content. You open the template, you see the prompts, and you start filling them in. The cognitive mode is generation — fast, associative, uncritical — exactly as First drafts are for content final drafts are for quality prescribed. The template ensures that even in this uncritical, fast-moving mode, the output will have coherent structure, because the structure was designed at a different time by a different cognitive mode (your planning mind, not your creating mind).
This is a genuine multiplier. The creation-editing separation already accelerates production. Adding templates to the creation phase accelerates it again. The compounding is real: two-pass production with templates is significantly faster than two-pass production without templates, which is already faster than single-pass production.
The Third Brain: AI-generated templates on demand
Your personal template library will cover your recurring output types. But what about the non-recurring ones — the output you have never produced before, or the format you use once a year?
This is where AI transforms the template game. You can generate a high-quality starting template for any output type in seconds.
The prompt pattern is simple: "I need to write [output type] for [audience]. The purpose is [goal]. Create a template with section headings and one-sentence prompts for each section." The AI produces a structural starting point — not content, not a draft, but the skeleton that your creation pass will fill.
Example: "I need to write a post-mortem analysis for a failed product launch. The audience is my executive team. The purpose is to identify root causes and prevent recurrence. Create a template." Within ten seconds, you have a structured template with sections for timeline, root cause analysis, contributing factors, what worked, what failed, recommendations, and next steps. You did not spend twenty minutes deciding what sections a post-mortem should have. You spent ten seconds, and now you are producing content.
You can also use AI to improve existing templates. Paste a template you have been using and ask: "What sections am I missing for this output type? What reordering would improve the reader's experience?" The AI serves as a template editor — analyzing structure, not content — in exactly the role where it excels.
One critical rule: use AI to generate templates, not to fill them. The template is the structure. The content is yours. If you use AI to generate both, the output is AI's work wearing your template's clothes. The point of a template is to free your cognitive resources for the content — the part where your expertise, your judgment, and your thinking produce irreplaceable value.
Templates as institutional memory
There is a deeper function of templates that goes beyond individual productivity.
When you build a template for a recurring output, you are encoding your accumulated experience with that output type into a reusable structure. The template captures what you have learned about what works — which sections matter, what order serves the audience, what information is always needed, what is usually missing. This is institutional memory at the individual level.
When you share templates with a team, the effect multiplies. A project proposal template that encodes the lessons of fifty prior proposals prevents every new proposal from making the same structural mistakes. A code review template that includes a security checklist ensures that security considerations are never accidentally omitted. The template carries the institution's learning forward, even as individuals rotate in and out.
This is why the best teams are obsessive about their templates. It is not bureaucracy. It is knowledge preservation. Every good template is a distillation of past experience into future advantage.
The bridge to minimum viable output
You now have a system for starting fast. The creation-editing separation (First drafts are for content final drafts are for quality) tells you how to produce. Templates tell you where to begin. But there is still a question that templates do not answer: how much is enough?
A template gives you a structure. It does not tell you how thoroughly to fill it. You could spend two hours on a status report template or twenty minutes. Both follow the same structure. Both use the same sections. But they differ by an order of magnitude in the time invested.
The minimum viable output will address this directly. The minimum viable output is the simplest version of your work that delivers the value the audience needs. Templates set the structure. Minimum viable output sets the depth. Together, they answer two of the most expensive questions in production: "What should this look like?" (the template answers this) and "When is it done enough?" (the minimum viable output answers this).
But those are two different lessons, because they require two different modes of thinking. You have the structure. Now let us figure out the threshold.
Sources:
- Baumeister, R. F., Vohs, K. D., & Tice, D. M. (2007). "The Strength Model of Self-Control." Current Directions in Psychological Science, 16(6), 351-355.
- Fogg, B. J. (2020). Tiny Habits: The Small Changes That Change Everything. Houghton Mifflin Harcourt.
- Thaler, R. H., & Sunstein, C. R. (2008). Nudge: Improving Decisions About Health, Wealth, and Happiness. Yale University Press.
- Baddeley, A. D. (2000). "The episodic buffer: a new component of working memory?" Trends in Cognitive Sciences, 4(11), 417-423.
- Bezos, J. (2018). "2017 Letter to Shareholders." Amazon.com, Inc.
- Fogg, B. J. (2009). "A Behavior Model for Persuasive Design." Proceedings of the 4th International Conference on Persuasive Technology.
- Baumeister, R. F., & Tierney, J. (2011). Willpower: Rediscovering the Greatest Human Strength. Penguin Press.
Frequently Asked Questions