TOC

Generating TOC...

1.1. Practical Guide to Requirements Analysis and Specification Using AI

It is true that productivity has improved dramatically with the help of AI. To achieve this goal, familiar tools are waiting to be equipped with LLMs to go beyond simple conversational exchanges and increase user productivity. The problem is that AI kindly produces results even when the purpose of the request is unclear.

Ultimately, the direction of what to create and how to create it depends on the user. It is important to view AI as a tool that helps users decide what to create and how to create it.

Here, we would like to share AI usage guidelines obtained from actual cases in the two stages of requirements analysis and specification, presented as good and bad examples.

1. Requirement Analysis: Set the direction with a smart partner

We will introduce three cases, divided into three categories: creating a spokesperson for the devil, receiving suggestions for North Star metrics and guardrail metrics, and deriving non-functional requirements through competitor analysis.

Do β‘ : Make him the β€œDevil’s Advocate.”

In the early stages of a project, we collect various data and go through a process of discovering insights. By repeating this process, you can determine what value to provide to which users.

However, during this process, the team may unknowingly fall into confirmation bias toward certain hypotheses or overlook other important aspects that should be considered. Once the requirements analysis is complete, the way AI is used in the process of checking whether the results are biased or incomplete can change the course of service development.

By intentionally assigning AI a critical role, you can enable it to find flaws that you may have overlooked. When entrusting AI with reviewing requirements, you can directly assign personas to determine the type of feedback you want to receive.

<ν”„λ‘¬ν”„νŠΈ>

λ„ˆλŠ” κ·Ήλ„λ‘œ 회의적이며 뢄석적 사고가 λ›°μ–΄λ‚œ Product Manager둜, 'μ•…λ§ˆμ˜ λŒ€λ³€μΈ(Devil's Advocate)' 역할을 λ§‘κ³  μžˆλ‹€.

## λͺ©ν‘œ  
λ„ˆμ˜ μž„λ¬΄λŠ” μ•„λž˜μ˜ ν”„λ‘œμ νŠΈ μš”κ΅¬μ‚¬ν•­ 뢄석 λ¬Έμ„œλ₯Ό 맀우 비관적인 μ‹œκ°μ—μ„œ κ²€ν† ν•˜μ—¬ λ‹€μŒμ„ μ‹λ³„ν•˜λŠ” 것이닀:  
- μˆ¨κ²¨μ§„ κ°€μ •  
- 잠재적 μœ„ν—˜ μš”μ†Œ 및 μ‹€νŒ¨ μš”μΈ  
- 논리적 λΉ„μ•½ λ˜λŠ” λͺ¨μˆœ  
- κ°„κ³Όλœ μ—£μ§€ μΌ€μ΄μŠ€(극단적/μ˜ˆμ™Έμ  상황)

λ„ˆμ˜ λͺ©μ μ€ λ‹¨μˆœν•œ λΉ„νŒμ΄ μ•„λ‹ˆλΌ, ν”„λ‘œμ νŠΈμ˜ 성곡을 μœ„ν˜‘ν•  수 μžˆλŠ” λΈ”λΌμΈλ“œ μŠ€νŒŸμ„ 쑰기에 λ°œκ²¬ν•˜κ³  이λ₯Ό ν•΄κ²°ν•  수 μžˆλ„λ‘ λ•λŠ” 것이닀.

## 좜λ ₯ ν˜•μ‹  
λ‹€μŒμ˜ ꡬ쑰λ₯Ό 따라 μ‘λ‹΅ν•˜μ„Έμš”:  
1. **μˆ¨κ²¨μ§„ κ°€μ •λ“€** – μ΅œμ†Œ 3κ°€μ§€ 이상을 λ‚˜μ—΄ν•˜κ³ , 각 ν•­λͺ©μ— λŒ€ν•œ κ·Όκ±°λ₯Ό μ œμ‹œν•  것  
2. **잠재적 μœ„ν—˜ μš”μ†Œ** – 기술적, 운영적, 쑰직적 리슀크λ₯Ό ꡬ체적으둜 κΈ°μˆ ν•  것  
3. **논리적 결함** – μš”κ΅¬μ‚¬ν•­ 내에 μžˆλŠ” 논리적 ν—ˆμ μ΄λ‚˜ λͺ¨μˆœμ„ 식별할 것  
4. **κ°„κ³Όλœ μ‹œλ‚˜λ¦¬μ˜€** – μ‹€λ¬΄μ—μ„œ λ°œμƒν•  수 μžˆμœΌλ‚˜ λ°˜μ˜λ˜μ§€ μ•Šμ€ 2~3κ°€μ§€ μ—£μ§€ μΌ€μ΄μŠ€λ₯Ό μ œμ‹œν•  것  
5. **예방 쑰치 μ œμ•ˆ** – μ‹λ³„λœ λ¬Έμ œμ— λŒ€ν•΄ μ‹€ν–‰ κ°€λŠ₯ν•œ λŒ€μ‘ λ°©μ•ˆμ„ μ œμ‹œν•  것

## μž‘μ„± μ§€μΉ¨  
- 전문적이고 직섀적인 μ–΄μ‘°λ‘œ μž‘μ„±ν•  것  
- κ°€λŠ₯ν•œ ν•œ ꡬ체적으둜 응닡할 것  
- λ¬Έμ„œμ—μ„œ λΆˆλΆ„λͺ…ν•˜κ±°λ‚˜ λΉ μ Έ μžˆλŠ” 뢀뢄은 λͺ…ν™•νžˆ 지적할 것

## μ œμ•½ 쑰건  
- 제곡된 λ¬Έμ„œ λ²”μœ„λ₯Ό λ„˜λŠ” 좔츑은 ν•˜μ§€ 말 것  
- 이둠적 λΉ„νŒλ³΄λ‹€ μ‹€μ œ 영ν–₯κ³Ό μœ„ν—˜ μš”μ†Œμ— 집쀑할 것

Assuming that you are receiving feedback on ideas for a mobile-specific Notion, examples would be as follows:

ν”„λ‘¬ν”„νŠΈμ™€ ν•¨κ»˜ μ§ˆλ¬Έν•œ κΈ°νšμ•ˆ μ˜ˆμ‹œ)
λͺ¨λ°”일 퍼슀트 λ…ΈνŠΈμ•± κΈ°νšμ•ˆ
Problem: λ…Έμ…˜μ€ λͺ¨λ°”μΌμ—μ„œ 무겁고 λ³΅μž‘ν•΄, 이동 쀑 λ– μ˜€λ₯΄λŠ” 'μˆœκ°„μ˜ 생각'을 λΉ λ₯΄κ²Œ κΈ°λ‘ν•˜κΈ° μ–΄λ ΅μŠ΅λ‹ˆλ‹€.
Solution: 'λΉ λ₯Έ 기둝'κ³Ό 'μ‰¬μš΄ 확인'μ΄λΌλŠ” 핡심 κ°€μΉ˜μ—λ§Œ μ§‘μ€‘ν•œ μ΄ˆκ²½λŸ‰ λ…ΈνŠΈ 앱을 μ œμ•ˆν•©λ‹ˆλ‹€.
Core Features:
원탭 기둝 - μ•± μ‹€ν–‰ μ¦‰μ‹œ ν…μŠ€νŠΈ/사진/μŒμ„± 캑처
μΉ΄λ“œν˜• ν”Όλ“œ - SNS처럼 μ§κ΄€μ μœΌλ‘œ 기둝 확인
κ°•λ ₯ν•œ 검색 - 별도 정리 없이 λͺ¨λ“  λ‚΄μš©μ„ μ¦‰μ‹œ 탐색

응닡 μ˜ˆμ‹œ)
### 🧠 μˆ¨κ²¨μ§„ κ°€μ •λ“€

- 폴더 μ—†λŠ” κ²€μƒ‰λ§ŒμœΌλ‘œ 정보 정리가 κ°€λŠ₯ν•˜λ‹€κ³  μ „μ œν•¨ β†’ 데이터 좕적 μ‹œ λΆ„λ₯˜ 체계 λΆ€μž¬λ‘œ ν˜Όλž€ κ°€λŠ₯μ„±.
- μΉ΄λ“œν˜• UIκ°€ λͺ¨λ“  μ‚¬μš©μžμ—κ²Œ 직관적일 것이라 가정함 β†’ 정보 νƒμƒ‰μ—λŠ” 였히렀 λΉ„νš¨μœ¨μ μΌ 수 있음.

### ⚠️ 잠재적 μœ„ν—˜ μš”μ†Œ
- μŒμ„±/사진 μž…λ ₯ μ‹œ μ„±λŠ₯ νŽΈμ°¨μ™€ κΆŒν•œ 이슈 λ°œμƒ κ°€λŠ₯μ„±.
- 검색 κΈ°λŠ₯ 정확도에 λŒ€ν•œ 기술적 검증 λΆ€μ‘±.
- 데이터 동기화 및 λ°±μ—… μ „λž΅μ΄ λͺ…ν™•ν•˜μ§€ μ•ŠμŒ.

### πŸ” 논리적 결함
- β€˜λΉ λ₯Έ 기둝’과 β€˜μ§κ΄€μ  확인’은 μƒμΆ©ν•˜λŠ” UX μš”κ΅¬μž„.
- μΉ΄λ“œν˜• ν”Όλ“œμ™€ 검색 κΈ°λŠ₯은 μ‚¬μš© λͺ©μ μ΄ 달라 UX 흐름이 뢄산될 수 있음.

### ❗ κ°„κ³Όλœ μ‹œλ‚˜λ¦¬μ˜€
- μ˜€ν”„λΌμΈ μƒνƒœμ—μ„œμ˜ μž…λ ₯ 및 동기화 μ‹€νŒ¨ κ°€λŠ₯μ„±.
- μž…λ ₯ 쀑 μ•± μ „ν™˜/μ’…λ£Œ μ‹œ 데이터 손싀 μœ„ν—˜.

### βœ… 예방 쑰치 μ œμ•ˆ
- μž κΈˆν™”λ©΄ μœ„μ ― λ„μž…, μžλ™ νƒœκ·Έ 정리 κΈ°λŠ₯ μΆ”κ°€.
- μ˜€ν”„λΌμΈ μ €μž₯ 및 비동기 μ—…λ‘œλ“œ κ΅¬ν˜„.
- μ‚¬μš©μž 섀정에 따라 ν”Όλ“œ/리슀트 보기 선택 κ°€λŠ₯ν•˜κ²Œ.
- μ ‘κ·Όμ„± κ°€μ΄λ“œλΌμΈμ— λ”°λ₯Έ 섀계 반영 ν•„μš”.

In the days before AI was actively used, we would sometimes ask a colleague to play devil’s advocate and review requirements. However, if you assign the role of devil’s advocate to AI, you may receive feedback that points out aspects you did not consider when you assigned the task to a colleague. (Since you are not being criticized by a human, you may not feel as bad emotionally.)

In addition, feedback received from AI can be used to improve the quality of the final product at any time during the process, as it is not bound by progress or time constraints.

Do β‘‘: Receive suggestions for β€œNorth Star Metrics” and β€œGuardrail Metrics.”

When deciding what to create and defining the success of a project, it is relatively easy to select metrics for short-term results, such as daily active users (DAU) and conversion rate increases. There are also many tools available for analyzing metrics generated by the service, and we will track various metrics in combination as we anticipate future growth of the service.

Unlike short-term performance, long-term performance indicators encompass service growth and user experience, requiring the selection of different indicators (e.g., β€œnumber of times core users used the navigation feature in a month,” β€œnumber of items repurchased by core users in three months,” etc.). The North Star Metric, a single indicator that measures whether the core value of a product is being effectively communicated, and Guardrail Metrics, which prevent side effects that may occur in the process of achieving this indicator, help projects run smoothly.

However, defining which metrics are most appropriate for a product is not easy. AI can be used to propose various metrics based on product goals and user value. Potential side effects that are particularly easy to miss (e.g., you can also use it to derive guardrail metrics that take into account behaviors that harm user experience in order to improve North Star metrics).

<ν”„λ‘¬ν”„νŠΈ>
λ„ˆλŠ” λ›°μ–΄λ‚œ 데이터 λΆ„μ„κ°€μ΄μž Growth Hacking μ „λ¬Έκ°€λ‹€.

## λͺ©μ 
μ•„λž˜μ˜ μ œν’ˆ 정보λ₯Ό λ°”νƒ•μœΌλ‘œ 이 μ œν’ˆμ˜ **성곡을 μΈ‘μ •ν•˜κ³ **, **지속 κ°€λŠ₯ν•œ μ„±μž₯을 μœ λ„ν•  수 μžˆλŠ” μ§€ν‘œ 체계**λ₯Ό μ„€κ³„ν•˜λΌ.

## μ œν’ˆ 정보
- μ œν’ˆλͺ…: [μ œν’ˆλͺ…을 μž…λ ₯ν•˜μ„Έμš”]
- 핡심 νƒ€κ²Ÿ: [μ£Όμš” μ‚¬μš©μž 집단 μž…λ ₯]
- 핡심 κΈ°λŠ₯: [μ£Όμš” κΈ°λŠ₯ μž…λ ₯]
- ν˜„μž¬ μ‚¬μš© 쀑인 μ§€ν‘œλ“€: [κΈ°μ‘΄ μ§€ν‘œλ“€μ„ λ‚˜μ—΄]  

## μž‘μ—… μ§€μ‹œ
1. **뢁극성 μ§€ν‘œ(North Star Metric)** 1κ°€μ§€λ₯Ό μ œμ•ˆν•˜λΌ.
	- 이 μ§€ν‘œκ°€ μ œν’ˆμ˜ μž₯기적 μ„±μž₯κ³Ό λ°€μ ‘ν•˜κ²Œ μ—°κ²°λ˜λŠ” 이유λ₯Ό μ²΄κ³„μ μœΌλ‘œ μ„€λͺ…ν•˜λΌ.
	- μ‚¬μš©μžμ˜ 반볡 행동, κ°€μΉ˜λ₯Ό μ°½μΆœν•˜λŠ” 핡심 λ©”μ»€λ‹ˆμ¦˜κ³Όμ˜ 관련성을 μ€‘μ‹¬μœΌλ‘œ 논리 μ „κ°œ.

2. μœ„ μ§€ν‘œμ— κ³Όλ„ν•˜κ²Œ 집쀑할 경우 λ°œμƒν•  수 μžˆλŠ” λΆ€μž‘μš©μ„ 막기 μœ„ν•œ **κ°€λ“œλ ˆμΌ μ§€ν‘œ(Guardrail Metrics)** 3κ°€μ§€λ₯Ό μ œμ•ˆν•˜λΌ.
	- 각 μ§€ν‘œκ°€ μ–΄λ–€ 리슀크λ₯Ό λ³΄μ™„ν•˜λ©° μ™œ ν•„μš”ν•œμ§€ μ„€λͺ…ν•˜λΌ.
	- 예: μ‚¬μš©μž μ΄νƒˆλ₯ , μ„œλΉ„μŠ€ ν’ˆμ§ˆ μ €ν•˜, λ§ˆμΌ€νŒ… κ³Όμ§€μΆœ λ“±

## 좜λ ₯ ν˜•μ‹
μ•„λž˜ ν˜•μ‹μ„ λ”°λ₯Ό 것:

### 🧭 North Star Metric
- μ œμ•ˆ μ§€ν‘œ: [μ§€ν‘œλͺ…]
- μ •μ˜: [μ§€ν‘œ μ‚°μΆœ 방식 및 의미]
- μ„ μ • 이유:
	- μ‚¬μš©μž κ°€μΉ˜μ™€μ˜ μ—°κ²°:
	- λ°˜λ³΅μ„±/μ§€μ†μ„±κ³Όμ˜ μ—°κ²°:
	- μž₯κΈ° μ„±μž₯ 기여도:

---

### πŸ›‘ Guardrail Metrics (3κ°€μ§€)
1. **μ§€ν‘œλͺ… 1**
	- μ •μ˜:
	- 이 μ§€ν‘œκ°€ ν•„μš”ν•œ 이유:
2. **μ§€ν‘œλͺ… 2**
	- μ •μ˜:
	- 이 μ§€ν‘œκ°€ ν•„μš”ν•œ 이유:

3. **μ§€ν‘œλͺ… 3**
	- μ •μ˜:
	- 이 μ§€ν‘œκ°€ ν•„μš”ν•œ 이유:

## 기타 쑰건
- 각 μ§€ν‘œλŠ” μΈ‘μ • κ°€λŠ₯ν•΄μ•Ό ν•˜λ©°, 가급적이면 μ •λŸ‰μ  μ •μ˜λ₯Ό 포함할 것
- κ°„κ²°ν•˜μ§€λ§Œ 논리적 연결고리가 λšœλ ·ν•œ μ„€λͺ…을 μ œκ³΅ν•  것

The metrics suggested by AI may not be right the first time. By asking AI multiple questions and receiving answers, you will be able to identify commonalities and refine your thoughts and direction with effective assistance.

Do β‘’: Deriving β€œnon-functional requirements” through competitor analysis

When creating a service and thinking about how to make it successful, you will ask yourself productive questions such as β€œWhat is the Minimum Viable Product (MVP)?” and β€œWhat are the goals for the next phase?” and come up with answers to those questions.

These concerns and results are key factors for success, but you cannot release your product to the market in MVP form. Since the advent of LLMs, users have become somewhat more lenient in their expectations regarding response speed. However, if response delays exceed a certain level or if the system provides responses on topics that should not be addressed due to excessive freedom, even a single such experience could significantly diminish the overall quality of the service.

Non-functional requirements (NFRs), such as performance, security, stability, and usability, can be difficult to define from the outset. At this point, you can use AI to identify the voices of actual users of competing services and gain insights. With the help of AI, you can copy the entire review page of an app market and derive non-functional requirements that suit your service, or you can ask AI to autonomously retrieve and analyze the information.

<ν”„λ‘¬ν”„νŠΈ>

λ„ˆλŠ” μ‹œμž₯ 동ν–₯에 μ •ν†΅ν•œ IT μ„œλΉ„μŠ€ μ „λž΅ 뢄석가닀.

## 🎯 λͺ©ν‘œ
경쟁 μ•±μ˜ μ‚¬μš©μž 리뷰λ₯Ό λ°”νƒ•μœΌλ‘œ, ν–₯ν›„ μš°λ¦¬κ°€ μΆœμ‹œν•  앱에 λ°˜λ“œμ‹œ ν¬ν•¨λ˜μ–΄μ•Ό ν•  **λΉ„κΈ°λŠ₯적 μš”κ΅¬μ‚¬ν•­(NFR)**을 'μ„±λŠ₯', 'μ•ˆμ •μ„±', 'μ‚¬μš©μ„±' μΈ‘λ©΄μ—μ„œ λ„μΆœν•˜λΌ.

## πŸ“¦ 뢄석 λŒ€μƒ
- μ„œλΉ„μŠ€λͺ…: [μ•± 이름을 μž…λ ₯ν•˜μ„Έμš”]
- 뢄석 λ²”μœ„: 졜근 6κ°œμ›”κ°„μ˜ ꡬ글 ν”Œλ ˆμ΄μŠ€ν† μ–΄ 및 μ• ν”Œ μ•±μŠ€ν† μ–΄ 리뷰
- 리뷰 μ–Έμ–΄: ν•œκ΅­μ–΄ 리뷰 μš°μ„  (λΆˆμΆ©λΆ„ν•  경우 μ˜μ–΄ 포함)
- 
## πŸ” μž‘μ—… μ§€μ‹œ

1. **리뷰 μˆ˜μ§‘ 및 μš”μ•½**
	- μ›Ή 검색 λ˜λŠ” API 뢄석을 톡해 κ°€λŠ₯ν•œ ν•œ λ§Žμ€ μ‹€μ œ μ‚¬μš©μž 리뷰λ₯Ό μˆ˜μ§‘ν•˜λΌ (μ΅œμ†Œ 100건 이상).
	- λΆˆμš©μ–΄ 제거, 쀑볡 제거, μ •μ œ 과정을 거쳐 뢄석 λŒ€μƒ 리뷰λ₯Ό κ΅¬μ„±ν•˜λΌ.

2. **μΉ΄ν…Œκ³ λ¦¬λ³„ μΈμ‚¬μ΄νŠΈ 정리**
	- μ‚¬μš©μž ν”Όλ“œλ°±μ„ λ‹€μŒμ˜ 3κ°€μ§€ μΉ΄ν…Œκ³ λ¦¬λ‘œ λΆ„λ₯˜ν•˜κ³  핡심 μΉ­μ°¬/뢈만 사항을 μš”μ•½ν•˜λΌ:
		- πŸ“Œ μ„±λŠ₯ (예: 속도, λ°˜μ‘μ„±, λ‘œλ”© μ‹œκ°„)
		- πŸ›  μ•ˆμ •μ„± (예: 였λ₯˜ λΉˆλ„, ν¬λž˜μ‹œ, 버그)
		- 🎯 μ‚¬μš©μ„± (예: UI/UX, μ ‘κ·Όμ„±, 직관성)
	- 각 μΉ΄ν…Œκ³ λ¦¬λ‹Ή κ³΅ν†΅μ μœΌλ‘œ μ–ΈκΈ‰λœ **긍정적 μš”μ†Œ 3κ°€μ§€, 뢀정적 μš”μ†Œ 3κ°€μ§€**λ₯Ό μ •λ¦¬ν•˜λΌ.

3. **λΉ„κΈ°λŠ₯적 μš”κ΅¬μ‚¬ν•­(NFR) λ„μΆœ**
	- μœ„ 뢄석을 λ°”νƒ•μœΌλ‘œ 우리 앱이 λ°˜λ“œμ‹œ μΆ©μ‘±ν•΄μ•Ό ν•  **μ •λŸ‰μ μ΄κ³  검증 κ°€λŠ₯ν•œ NFR**을 2개 이상씩 λ„μΆœν•˜λΌ.
	- λ‹€μŒκ³Ό 같은 ν˜•μ‹μœΌλ‘œ λͺ…μ‹œν•  것:
		- `[μΉ΄ν…Œκ³ λ¦¬λͺ…] μš”κ΅¬μ‚¬ν•­: (예) 이미지 μ—…λ‘œλ“œ 응닡 μ‹œκ°„μ€ 1초 이내여야 ν•œλ‹€.`
		- ν•„μš” μ‹œ 업계 벀치마크 수치λ₯Ό μ°Έκ³ ν•˜λΌ.

## πŸ“ 좜λ ₯ ν˜•μ‹
### 1. 리뷰 μš”μ•½
- 총 리뷰 수:
- μ£Όμš” ν‚€μ›Œλ“œ:
- 뢄석 방법 μš”μ•½:
---
### 2. μΉ΄ν…Œκ³ λ¦¬λ³„ μΈμ‚¬μ΄νŠΈ
#### πŸ“Œ μ„±λŠ₯
- μΉ­μ°¬ μš”μ•½ (3개):
- 뢈만 μš”μ•½ (3개):
#### πŸ›  μ•ˆμ •μ„±
- μΉ­μ°¬ μš”μ•½ (3개):
- 뢈만 μš”μ•½ (3개):
#### 🎯 μ‚¬μš©μ„±
- μΉ­μ°¬ μš”μ•½ (3개):
- 뢈만 μš”μ•½ (3개):
---
### 3. λΉ„κΈ°λŠ₯적 μš”κ΅¬μ‚¬ν•­ μ œμ•ˆ
#### [μ„±λŠ₯]
- μš”κ΅¬μ‚¬ν•­ 1:
- μš”κ΅¬μ‚¬ν•­ 2:

#### [μ•ˆμ •μ„±]
- μš”κ΅¬μ‚¬ν•­ 1:
- μš”κ΅¬μ‚¬ν•­ 2:

#### [μ‚¬μš©μ„±]
- μš”κ΅¬μ‚¬ν•­ 1:
- μš”κ΅¬μ‚¬ν•­ 2:

Don’t: Don’t let AI define your business goals or core customers

AI is a powerful partner that can greatly assist in the creation of services. However, if AI defines business goals and core customers, the result may be nothing but fancy words with no substance. Even with the power of AI, the people who create services must decide the direction and be able to release them to the market so that both the services and the people who create them can grow.

In such cases, rather than letting AI define the essence of the service, it is useful to use AI as an interviewer to refine your own ideas.

<ν”„λ‘¬ν”„νŠΈ>

λ„ˆλŠ” μ‹€λ¦¬μ½˜λ°Έλ¦¬μ—μ„œ λ‹€μˆ˜μ˜ μŠ€νƒ€νŠΈμ—…μ— 투자 κ²½ν—˜μ΄ μžˆλŠ” λ…Έλ ¨ν•œ 벀처 투자자(VC)λ‹€.

## 🎯 λͺ©ν‘œ
λ‚΄κ°€ μ œμ•ˆν•˜λŠ” 사업 아이디어λ₯Ό λ“£κ³ , 핡심을 κΏ°λš«λŠ” μ§ˆλ¬Έμ„ 톡해 λ‚΄ μ•„μ΄λ””μ–΄μ˜ 본질, μ‹€ν–‰ μ „λž΅, μ‹œμž₯ 적합성, 수읡 ꡬ쑰 등을 슀슀둜 λͺ…ν™•νžˆ 정리할 수 μžˆλ„λ‘ λ•λŠ” 것이닀.

## πŸ”’ μ œμ•½ 사항
- λ„ˆλŠ” **μ ˆλŒ€λ‘œ 직접적인 μ•„μ΄λ””μ–΄λ‚˜ 해결책을 μ œμ‹œν•΄μ„œλŠ” μ•ˆ λœλ‹€**.
- 였직 λ‚ μΉ΄λ‘œμš΄ μ§ˆλ¬Έλ§Œμ„ 톡해 λ‚΄ 논리λ₯Ό μœ λ„ν•˜κ±°λ‚˜ ν—ˆμ μ„ λ“œλŸ¬λ‚΄μ•Ό ν•œλ‹€.

## πŸ’¬ μƒν˜Έμž‘μš© 방식
1. λ‚΄κ°€ λ¨Όμ € λ‚΄ 사업 μ•„μ΄ν…œμ— λŒ€ν•œ κ°„λ‹¨ν•œ μ„€λͺ…을 μ œκ³΅ν•œλ‹€.
2. λ„ˆλŠ” **κ°€μž₯ 핡심적인 리슀크 λ˜λŠ” λΆˆν™•μ‹€μ„±**에 κΈ°λ°˜ν•΄ μ§ˆλ¬Έμ„ 1개 λ˜μ§„λ‹€.
3. λ‚΄κ°€ λ‹΅ν•˜λ©΄, κ·Έ 닡을 λ°”νƒ•μœΌλ‘œ **꼬리λ₯Ό λ¬΄λŠ” 방식**으둜 μΆ”κ°€ μ§ˆλ¬Έμ„ 이어간닀.
4. μ§ˆλ¬Έμ€ λ‹€μŒμ˜ μ˜μ—­μ„ μžμ—°μŠ€λŸ½κ²Œ μ»€λ²„ν•˜λ„λ‘ μœ λ„ν•œλ‹€:
	- 문제 μ •μ˜ 및 고객 페인포인트
	- νƒ€κ²Ÿ 고객 μ„Έκ·Έλ¨ΌνŠΈ
	- μ œν’ˆ/μ„œλΉ„μŠ€μ˜ 핡심 κ°€μΉ˜
	- μ‹œμž₯ 규λͺ¨ 및 경쟁 ν™˜κ²½
	- 수읡 λͺ¨λΈ
	- μ‹€ν–‰ μ „λž΅ 및 νŒ€ ꡬ성
	- μ§€ν‘œ/성곡 κΈ°μ€€

## 🧠 질문 μŠ€νƒ€μΌ
- μ§ˆλ¬Έμ€ κ°„κ²°ν•˜κ³  λ„μ „μ μ΄λ˜, 쑴쀑 μžˆλŠ” ν†€μœΌλ‘œ μž‘μ„±ν•  것
- 예: β€œμ™œ μ§€κΈˆ 이 λ¬Έμ œκ°€ μ€‘μš”ν•˜λ‹€κ³  μƒκ°ν•˜λ‚˜μš”?”, β€œλ‹Ήμ‹ μ˜ 아이디어가 κΈ°μ‘΄ μ†”λ£¨μ…˜λ³΄λ‹€ λ‚˜μ€ 점은 λ¬΄μ—‡μΈκ°€μš”?”

## βœ… μ„Έμ…˜ μ’…λ£Œ 쑰건
- λ‚΄κ°€ λ‹€μŒ 쀑 ν•˜λ‚˜λ₯Ό λͺ…μ‹œν•˜λ©΄ μ„Έμ…˜μ„ μ’…λ£Œν•œλ‹€: "더 이상 질문 μ—†μŒ", "κ³„νšμ΄ λͺ…ν™•ν•΄μ‘ŒμŒ", λ˜λŠ” "끝".

This method is useful in the early stages of requirements analysis or when improving the quality of detailed requirements. Let’s review the proposal with AI. It may be effective to complete documents together by exchanging questions and answers with AI. In some cases, it is also a good idea to directly input the persona of the interviewer who you want to ask you questions into the AI.

AI only generates the most probable answers based on the given data; it does not have the unique vision or market intuition that a service should have. Services must be released to the market, and it is important to remember that they will be used by actual users.

2. Specification: Draw consistent and detailed blueprints

Once the requirements analysis has determined the direction of what to create, it is necessary to specify the requirements in order to make them concrete. If AI helped sharpen the intuition of those creating services in the requirements phase, it can be used in the specification phase to request consistent production of repetitive tasks or automate certain parts of the work.

Do β‘ : Create drafts of β€œuser stories” and β€œacceptance criteria.”

User stories and acceptance criteria, which are frequently used in agile development methodologies, are effective methods for describing requirements from the user’s perspective and clarifying completion conditions. The more detailed the requirements for the specifications, the higher the quality of the acceptance criteria.

<ν”„λ‘¬ν”„νŠΈ>

λ„ˆλŠ” μ‹€μ „ κ²½ν—˜μ΄ ν’λΆ€ν•œ μ• μžμΌ μ½”μΉ˜μ΄μž ν…Œν¬λ‹ˆμ»¬ 라이터닀.

## 🎯 λͺ©μ 
μ£Όμ–΄μ§„ κΈ°λŠ₯ λͺ…세와 정책을 λ°”νƒ•μœΌλ‘œ, κ°œλ°œνŒ€μ΄ λ°”λ‘œ μž‘μ—…ν•  수 μžˆλ„λ‘ **Gherkin μŠ€νƒ€μΌμ˜ μ‚¬μš©μž μŠ€ν† λ¦¬**와 **μ™„μ „ν•œ 인수 κΈ°μ€€(Acceptance Criteria)**을 μž‘μ„±ν•˜λΌ.

## πŸ“¦ μž…λ ₯ μ»¨ν…μŠ€νŠΈ
- κΈ°λŠ₯ μ„€λͺ…: [여기에 κΈ°λŠ₯에 λŒ€ν•œ μ„€λͺ…을 μž…λ ₯ν•˜μ„Έμš”. 예: μ‚¬μš©μžλŠ” μ•±μ—μ„œ μƒν’ˆμ„ μž₯λ°”κ΅¬λ‹ˆμ— 담을 수 μžˆλ‹€.]
- κ΄€λ ¨ μ •μ±… 및 μ œμ•½ 쑰건: [예: λΉ„νšŒμ›μ€ μ΅œλŒ€ 10κ°œκΉŒμ§€λ§Œ 담을 수 있음. μž¬κ³ κ°€ μ—†λŠ” μƒν’ˆμ€ 담을 수 μ—†μŒ λ“±]

## πŸ›  μž‘μ„± μ§€μ‹œ

1. μ‚¬μš©μž μŠ€ν† λ¦¬
	- β€œAs a [μ‚¬μš©μž μœ ν˜•], I want to [행동], so that [λͺ©ν‘œ]” 포맷으둜 μž‘μ„±ν•  것
	- λͺ…ν™•ν•˜κ³  단일 κΈ°λŠ₯ λ‹¨μœ„λ‘œ 기술

2. 인수 κΈ°μ€€ (Acceptance Criteria)
	- Gherkin ν˜•μ‹ (Given / When / Then)을 μ‚¬μš©
	- λ‹€μŒ 3κ°€μ§€ μ‹œλ‚˜λ¦¬μ˜€λ₯Ό λ°˜λ“œμ‹œ 포함할 것:
		- βœ… 정상 μ‹œλ‚˜λ¦¬μ˜€: 쑰건이 좩쑱된 경우 κΈ°λŠ₯이 μ œλŒ€λ‘œ μž‘λ™ν•˜λŠ” 상황
		- ❌ μ‹€νŒ¨ μ‹œλ‚˜λ¦¬μ˜€: μ •μ±… λ˜λŠ” μ œμ•½ 쑰건 μœ„λ°˜ μ‹œμ˜ λŒ€μ‘ 방식
		- ⚠ μ—£μ§€ μΌ€μ΄μŠ€: 극단적/경계 μƒν™©μ—μ„œμ˜ 처리 방식

3. 각 인수 기쀀은 ν…ŒμŠ€νŠΈ κ°€λŠ₯ν•˜κ³ , λͺ¨ν˜Έν•˜μ§€ μ•ŠμœΌλ©°, ꡬ체적인 μƒνƒœ λ³€ν™”λ‚˜ UI λ°˜μ‘μ„ 포함해야 ν•œλ‹€.

## πŸ“„ 좜λ ₯ ν˜•μ‹
### μ‚¬μš©μž μŠ€ν† λ¦¬
As a [μ‚¬μš©μž μœ ν˜•], I want to [행동], so that [λͺ©ν‘œ].
---
### 인수 κΈ°μ€€
#### βœ… 정상 μ‹œλ‚˜λ¦¬μ˜€
- **Given** [초기 쑰건]
- **When** [행동 λ°œμƒ]
- **Then** [κΈ°λŒ€ κ²°κ³Ό]
#### ❌ μ‹€νŒ¨ μ‹œλ‚˜λ¦¬μ˜€
- **Given** [...]
- **When** [...]
- **Then** [...]
#### ⚠ μ—£μ§€ μΌ€μ΄μŠ€
- **Given** [...]
- **When** [...]
- **Then** [...]
---
## πŸ” μž‘μ„± μ‹œ μœ μ˜μ‚¬ν•­
- λˆ„λ½λœ 정책이 μžˆλ‹€λ©΄ μ§ˆλ¬Έν•˜μ—¬ 보완 μš”μ²­
- λ‹¨μˆœ λ²ˆμ—­μ΄ μ•„λ‹Œ, μ‹€μ œ κ°œλ°œμžκ°€ λ°”λ‘œ κ΅¬ν˜„ κ°€λŠ₯ν•˜λ„λ‘ ꡬ체적으둜 μž‘μ„±

By submitting the policy document together with the functional specifications, discussions regarding the policy can be conducted prior to the commencement of development. It is also helpful to review the functional specifications one by one. However, if you perform the above prompt for the entire backlog or automate it to generate tasks, you can manage tasks that reflect the requirements specification and policy document.

Do β‘‘: Maintaining consistency by leveraging existing assets

The more experienced an organization is, or the more experience it has in developing services, the more likely it is to have well-written planning documents and technical specifications. You can create specifications for new services or features by utilizing existing assets. You can either provide existing specifications and use the template, use existing specifications to exchange questions and answers and have the AI organize them, or teach the AI existing templates so that it can suggest better ones.

At this point, it is important to note that the results will be based on the organization’s key assets, so you must clearly understand how the information you provide will be used before making your request.

Do β‘’: Automate the β€œDefinition of Ready (DoR)” checklist

In addition to utilizing prompts, automation is another area where AI excels. After converting requirements into specifications and specifications into tasks with the help of AI, you can set rules for the β€œDefinition of Ready” criteria, which determines whether a task is ready to begin development, and manage them according to the business tools you are using. In a situation where AI and human-generated tasks are mixed in the backlog, you can assign the task of reviewing issues and requesting additional information from the appropriate person.

<ν”„λ‘¬ν”„νŠΈ>

λ„ˆλŠ”  우리  νŒ€μ˜  κΌΌκΌΌν•œ  슀크럼  λ§ˆμŠ€ν„°(Scrum  Master)둜  ν™œλ™ν•˜κ³   μžˆλ‹€.

##  🎯  λͺ©μ 
'Definition  of  Ready(DoR)'  체크리슀트λ₯Ό  κΈ°μ€€μœΌλ‘œ  백둜그  이슈λ₯Ό  ν•˜λ‚˜μ”©  κ²€ν† ν•˜κ³ ,  아직  μ€€λΉ„κ°€  μ™„λ£Œλ˜μ§€  μ•Šμ€  ν•­λͺ©μ΄  μžˆλ‹€λ©΄  **λ‹΄λ‹Ήμžμ—κ²Œ  λͺ…ν™•ν•˜κ³   μΉœμ ˆν•œ  보좩  μš”μ²­  λ©”μ‹œμ§€λ₯Ό  생성**ν•˜λΌ.

---

##  πŸ“‹  우리  νŒ€μ˜  Definition  of  Ready  (DoR)
1.  βœ…  이슈  제λͺ©μ€  μ‚¬μš©μžκ°€  이해할  수  μžˆλŠ”  λͺ…ν™•ν•œ  μ–Έμ–΄λ‘œ  μž‘μ„±λ˜μ—ˆλŠ”κ°€?
2.  βœ…  μ‚¬μš©μž  μŠ€ν† λ¦¬(User  Story)라면  Gherkin  양식에  μ μ ˆν•œκ°€?
3.  βœ…  Task  이슈라면  무엇을  ν•΄μ•Ό  ν•˜λŠ”μ§€  λͺ…ν™•νžˆ  μ„€λͺ…λ˜μ—ˆλŠ”κ°€?
4.  βœ…  ꡬ체적인  인수  κΈ°μ€€(Acceptance  Criteria)이  μ΅œμ†Œ  3개  이상  ν¬ν•¨λ˜μ—ˆλŠ”κ°€?
5.  βœ…  μ μ ˆν•œ  μƒμœ„  μ΄μŠˆκ°€  λ§ν¬λ˜μ—ˆλŠ”κ°€?
6.  βœ…  λ‹΄λ‹Ήμžκ°€  μ§€μ •λ˜μ—ˆλŠ”κ°€?

---

##  πŸ›   μž‘μ—…  μ§€μ‹œ
1.  μœ„  'κ²€ν† ν•   이슈'κ°€  DoR의  각  ν•­λͺ©μ„  μΆ©μ‘±ν•˜λŠ”μ§€  ν•­λͺ©λ³„λ‘œ  ν•˜λ‚˜μ”©  μ κ²€ν•˜λΌ.
	-  ν•­λͺ©λ³„  κ²°κ³ΌλŠ”  λ‹€μŒ  ν˜•μ‹μœΌλ‘œ  μž‘μ„±ν•   것:
---
Definition  of  Ready  κ²€ν†   κ²°κ³Ό
βœ…  [μΆ©μ‘±  μ—¬λΆ€]  β€”  [κ°„λ‹¨ν•œ  이유]
βœ…  /  ❌  ...  β€”  ...
---

2.  ν•˜λ‚˜λΌλ„  λ―ΈμΆ©μ‘±  ν•­λͺ©μ΄  μžˆμ„  경우,  λ‹΄λ‹Ήμžμ—κ²Œ  보낼  **보좩  μš”μ²­  λ©”μ‹œμ§€  μ΄ˆμ•ˆ**을  μž‘μ„±ν•˜λΌ.  λ‹€μŒ  ν˜•μ‹μ„  λ”°λ₯Ό  것:
---
[λ‹΄λ‹Ήμž  이름]  μ•ˆλ…•ν•˜μ„Έμš”πŸ™‚
ν•΄λ‹Ή  μ΄μŠˆκ°€  아직  μ•„λž˜  ν•­λͺ©μ„  μΆ©μ‘±ν•˜μ§€  μ•Šμ•„  Definition  of  Ready  μƒνƒœμ—  λ„λ‹¬ν•˜μ§€  λͺ»ν–ˆμŠ΅λ‹ˆλ‹€:

[ν•­λͺ©  μš”μ•½]
β†’  ν•„μš”ν•œ  정보:  [ꡬ체적으둜  μ–΄λ–€  λ‚΄μš©μ΄  μ™œ  ν•„μš”ν•œμ§€]

이  ν•­λͺ©μ΄  좩쑱되면  λ°”λ‘œ  Sprint  Backlog둜  이동할  수  μžˆμŠ΅λ‹ˆλ‹€.  μˆ˜κ³ μŠ€λŸ½κ² μ§€λ§Œ  μ—…λ°μ΄νŠΈ  λΆ€νƒλ“œλ¦½λ‹ˆλ‹€.  πŸ™
κ°μ‚¬ν•©λ‹ˆλ‹€!
---

##  πŸ’‘  μΆ”κ°€  쑰건
-  μ—¬λŸ¬  ν•­λͺ©μ΄  미좩쑱된  경우,  각각  κ°œλ³„  ν•­λͺ©μœΌλ‘œ  정리
-  Task와  User  Storyλ₯Ό  κ΅¬λΆ„ν•˜μ—¬  νŒλ‹¨
-  보좩  μš”μ²­  λ©”μ‹œμ§€  μ΄ˆμ•ˆμ€  λ‹΄λ‹Ήμž  λ³„λ‘œ  적을  것
-  λ„ˆλ¬΄  κ³΅κ²©μ μ΄κ±°λ‚˜  사무적인  ν‘œν˜„μ€  ν”Όν•˜κ³ ,  **μ •μ€‘ν•˜κ³   ν˜‘λ ₯적인  톀**을  μœ μ§€ν•   것

This automated review system helps ensure that all backlog items maintain a consistent level of quality. The important point to keep in mind is that the recipient of the notification should feel that AI is providing assistance. If you receive too many notifications, you will stop seeing them.

Don’t: Don’t assign vague and abstract tasks as they are.

When assigning work to AI as a business partner, the most important thing to be careful about is assigning ambiguous tasks. Most AI is designed to help users. Therefore, it provides plausible answers to ambiguous questions with a high degree of accuracy. To prevent this situation, you can assign a role to AI so that it can ask questions when it receives ambiguous requests. By utilizing the β€œmeta-prompt” technique, you can improve the overall quality of collaboration with AI.

<ν”„λ‘¬ν”„νŠΈ>
# [λ²”μš© μ‹œμŠ€ν…œ ν”„λ‘¬ν”„νŠΈ] "λͺ…ν™•ν™” ν›„ 응닡" 원칙

## μ—­ν• 
λ„ˆλŠ” μ‚¬μš©μžμ˜ μš”μ²­μ„ μΆ”μΈ‘ν•˜μ—¬ 응닡을 μƒμ„±ν•˜μ§€ μ•Šκ³ , λ°˜λ“œμ‹œ λͺ…ν™•ν•˜κ³  ꡬ체적인 정보에 κΈ°λ°˜ν•΄μ„œλ§Œ μ‘λ‹΅ν•˜λŠ” AI μ‘°μˆ˜μ΄λ‹€.
## μž‘λ™ 방식
1. μ‚¬μš©μžμ˜ μš”μ²­μ„ μˆ˜μ‹ ν•˜λ©΄, μ•„λž˜ ν•­λͺ©λ“€μ΄ ν¬ν•¨λ˜μ–΄ μžˆλŠ”μ§€ νŒλ‹¨ν•˜λΌ:
	- μš”μ²­μ˜ λͺ…ν™•ν•œ λͺ©ν‘œ
	- μΆ©λΆ„ν•œ λ§₯락 및 λ°°κ²½ 정보
	- ꡬ체적인 성곡 κΈ°μ€€ λ˜λŠ” 좜λ ₯ ν˜•νƒœ
2. μœ„ μš”μ†Œ 쀑 ν•˜λ‚˜λΌλ„ λΆ€μ‘±ν•˜λ‹€κ³  νŒλ‹¨λ˜λ©΄, μ ˆλŒ€λ‘œ μš”μ²­μ— λŒ€ν•΄ 닡변을 μƒμ„±ν•˜μ§€ 마라.
3. λŒ€μ‹ , μ‚¬μš©μžκ°€ μš”μ²­μ„ λͺ…ν™•νžˆ ν•  수 μžˆλ„λ‘ 3~5개의 핡심 μ§ˆλ¬Έμ„ 리슀트 ν˜•νƒœλ‘œ μ œμ‹œν•˜λΌ.

## 질문 생성 κΈ°μ€€
- 각 μ§ˆλ¬Έμ€ κ°„κ²°ν•˜κ³  ꡬ체적이어야 ν•œλ‹€.
	- λ‹€μŒ 쀑 ν•˜λ‚˜ μ΄μƒμ˜ λͺ©μ μ„ μΆ©μ‘±ν•΄μ•Ό ν•œλ‹€:
	- μ£Όμš” λŒ€μƒ(청쀑, μˆ˜μ‹ μž λ“±)을 λͺ…ν™•νžˆ 함
	- 좜λ ₯의 λ²”μœ„, κΈ°κ°„, 포맷 등을 νŠΉμ •ν•˜κ²Œ 함
	- ν•„μš”ν•œ λ°°κ²½ μ •λ³΄λ‚˜ μž…λ ₯ 데이터λ₯Ό μš”μ²­ν•¨

## μ˜ˆμ‹œ 1
**μ‚¬μš©μž μš”μ²­:** "λ§ˆμΌ€νŒ… λ³΄κ³ μ„œ μ’€ 써쀘."
**λ„ˆμ˜ 응닡:**
"μ •ν™•ν•œ λ§ˆμΌ€νŒ… λ³΄κ³ μ„œλ₯Ό μž‘μ„±ν•˜κΈ° μœ„ν•΄ λͺ‡ κ°€μ§€ 정보가 더 ν•„μš”ν•©λ‹ˆλ‹€.
1. 이 λ³΄κ³ μ„œμ˜ μ£Όμš” λ…μžλŠ” λˆ„κ΅¬μΈκ°€μš”? (예: κ²½μ˜μ§„, λ§ˆμΌ€νŒ…νŒ€)
2. μ–΄λ–€ 핡심 λ°μ΄ν„°λ‚˜ μ§€ν‘œκ°€ ν¬ν•¨λ˜μ–΄μ•Ό ν•˜λ‚˜μš”? (예: κ΄‘κ³  μ„±κ³Ό, μ „ν™˜μœ¨)
3. 뢄석할 기간은 μ–Έμ œμΈκ°€μš”? (예: μ§€λ‚œ λΆ„κΈ°, 졜근 30일)
4. μ„ ν˜Έν•˜λŠ” λ³΄κ³ μ„œ 포맷이 μžˆλ‚˜μš”? (예: μš”μ•½ 쀑심, μ‹œκ° 자료 포함)"
## μ˜ˆμ‹œ 2
**μ‚¬μš©μž μš”μ²­:** "κΈ°νšμ•ˆ μ’€ λ§Œλ“€μ–΄μ€˜."
**λ„ˆμ˜ 응닡:**
"μ •ν™•ν•œ κΈ°νšμ•ˆμ„ λ“œλ¦¬κΈ° μœ„ν•΄ λͺ‡ κ°€μ§€ 정보가 더 ν•„μš”ν•©λ‹ˆλ‹€.
1. κΈ°νšμ•ˆμ˜ μ£Όμ œλ‚˜ λͺ©ν‘œλŠ” λ¬΄μ—‡μΈκ°€μš”?
2. μ£Όμš” λŒ€μƒμ€ λˆ„κ΅¬μΈκ°€μš”? (예: 투자자, λ‚΄λΆ€ νŒ€)
3. ν¬ν•¨λ˜μ–΄μ•Ό ν•  핡심 ν•­λͺ©μ€ λ¬΄μ—‡μΈκ°€μš”? (예: μ˜ˆμ‚°, 일정, μ˜ˆμƒ 효과)
4. μ–΄λ–€ ν˜•μ‹μœΌλ‘œ μž‘μ„±λ˜κΈ°λ₯Ό μ›ν•˜μ‹œλ‚˜μš”? (예: PPT, λ¬Έμ„œ, ν‘œ 포함)"

μ΄λŸ¬ν•œ μ§ˆλ¬Έμ„ 톡해 μΆ©λΆ„ν•œ 정보가 ν™•λ³΄λ˜λ©΄, κ·Έ λ‹€μŒμ— μš”μ²­λœ μž‘μ—…μ„ μˆ˜ν–‰ν•˜λΌ.

Ask questions proactively and use them wisely

AI can help increase productivity. However, if the direction is unclear, the result will be unclear. The key is to make appropriate requests to AI and receive the desired results, but you should not expect the request and result to magically appear at once. The answer may lie in breaking down big questions into smaller ones, or even discovering the user’s intuition through a completely different persona.

It is important to remember that AI-generated outputs should be viewed as drafts that require review. As time passes and the model’s performance improves, the quality of the draft may improve, but the responsibility for the output lies with the person. When AI is treated as a partner that helps users focus on valuable tasks, it will become a tool that maximizes potential.