# Habit 3: Put First Things First → Prioritize Depth over Speed

#### What It Means in LLM Context

Covey’s original Habit 3 is about **execution aligned to priorities**, not urgencies. It’s the act of living from your center, not reacting to the clock or noise.

In LLM usage, this means:

Resist the urge to “just get the answer.”\
Instead, prioritize **depth, structure, and signal**, even if it takes longer.

Speed is deceptive. Shallow prompting feels fast but creates cognitive debt downstream.

***

#### Common “Urgency-Driven” Patterns

| Pattern           | LLM Behavior                            | Consequence                      |
| ----------------- | --------------------------------------- | -------------------------------- |
| One-shotting      | “Just give me the code”                 | Low adaptability, opaque bugs    |
| Skimming          | “Summarize this PDF” without saying why | Irrelevant focus, loss of nuance |
| Iteration fatigue | Keep regenerating instead of refining   | No improvement loop              |
| Prompt gimmicks   | “Act like Steve Jobs and explain X”     | Theatrical output, minimal rigor |

***

#### Depth-First Prompting = Quality Thinking

Depth doesn’t mean verbosity. It means:

* **Staying in the question** longer
* Building **layers** (context → insight → reframing)
* Slowing down enough to see what matters

Covey would call this **Quadrant II** behavior — high value, non-urgent, often neglected.

***

#### Depth-Oriented Prompting Moves

| Move               | Example                                                                     |
| ------------------ | --------------------------------------------------------------------------- |
| Start with terrain | “Before we answer, what’s the landscape of views on this?”                  |
| Build layers       | “Let’s first define the problem, then explore causes, then test solutions.” |
| Stress test        | “What would critics say about this answer? Where might it fail?”            |
| Stay longer        | “Don’t rush to summarize. Stay with the ambiguity a bit longer.”            |
| Reflect, revise    | “Here’s what I learned. Now let’s rebuild the prompt with this insight.”    |

***

#### Case Study: Resume Writing

**Fast-first prompt:**

“Rewrite this resume to sound better.”

**Depth-first sequence:**

1. “What themes are emerging from this person’s past roles?”
2. “What kind of story does this resume tell? linear? fragmented? aspirational?”
3. “Now rewrite it to highlight those themes, while keeping the voice consistent.”

Result: a *crafted* document, not just a *rewritten* one.

***

#### Self-Check

* Am I optimizing for **speed** or **structure**?
* Would I ask a **human expert** this same question, or would I prepare them first?
* Did I pause long enough to identify **what matters most** before engaging the model?

***

#### A Prompting Pyramid

|  <p>Strategic Framing</p><p>(purpose + depth + voice)</p>  |
| :--------------------------------------------------------: |
| <p>Multi-Layer Scaffolding</p><p>(map → define → test)</p> |
| <p>Raw Content Request</p><p>(summarize, reword, list)</p> |

&#x20;

Most users operate at the **bottom**.\
But impact lives in the **top two layers**.

***

#### Downstream Risk of Shallow Use

| Shallow Prompting      | Later Consequence              |
| ---------------------- | ------------------------------ |
| Generic ideas          | Feels “meh” on delivery        |
| Vague code             | Debugging takes longer         |
| Unchallenged arguments | Rejected by audience           |
| Copied structure       | Accidental plagiarism          |
| Fast summary           | Decision-making based on noise |

***

#### Integration Cue

*Before every prompt, ask: “If I slow down here, will I save time or error later?”*

True LLM mastery is **long-game prompting.**\
It looks slower. But it moves further, with fewer rebuilds.

***

#### Metaphor

Think of fast prompting like painting on a fogged mirror.\
The shapes appear instantly, but they vanish when the heat changes.

Depth-first prompting is like **etching into glass.**\
Slower, but permanent. And transferable.

***


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://praxistutor.gitbook.io/llm/reframing-the-7-habits-for-reflective-llm-use/section-2-the-seven-habits-of-reflective-llm-use/habit-3-put-first-things-first-prioritize-depth-over-speed.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.