6.3 KiB
Closer Question Rewrite Plan v3
See also: QUESTION_CONTENT_GUIDE.md | QUESTION_SCHEMA.md | QUESTION_QUALITY_CHECKLIST.md
Product Goal
Closer is a relationship app, not a survey.
The experience should feel like a couples game that naturally creates conversations, laughter, memories, flirting, and meaningful moments.
If a question feels like homework, rewrite it.
Repo Schema
Use these type names exactly:
- written
- single_choice
- multi_choice
- scale
- this_or_that
Do not rename them unless the app code is updated first.
Standard Question Mix per 250
| Type | Count |
|---|---|
| multi_choice | 140 |
| single_choice | 50 |
| scale | 35 |
| this_or_that | 15 |
| written | 10 |
76 percent of questions must be choice based.
Written questions should stay rare.
Free and Premium Split
Per standard 250 question pack:
- 75 free
- 175 premium
Suggested free split:
- multi_choice: 42
- single_choice: 15
- scale: 11
- this_or_that: 5
- written: 2
Suggested premium split:
- multi_choice: 98
- single_choice: 35
- scale: 24
- this_or_that: 10
- written: 8
Free should feel useful, fun, and complete.
Premium should feel deeper, richer, and more varied.
Special Pack Exception
Some packs are product-specific and may override the standard 250 question mix.
A special pack must document:
- pack id
- file name
- expected count
- allowed question types
- free count
- premium count
- required tags
- reason for the exception
Special packs still must pass tone, duplicate, option, and schema review.
Daily Single Choice Weekday Pack
Pack id:
daily_single_choice_weekly_v1
Current compatibility file, if the app still expects it:
daily_fun_multiple_choice_v3.json
Recommended file name after code cleanup:
daily_single_choice_weekly_v1.json
Expected counts:
- total questions: 500
- free questions: 75
- premium questions: 425
- type count: 500 single_choice
These counts are a special daily pack exception. Do not use the standard 250 question mix for this file.
Rules:
- every question must be single_choice
- every question must have 4 to 6 options
- every question must have exactly one weekday tag
- every option must answer the prompt
- no therapy worksheet tone
- no wellness survey tone
- no repeated option lists
Weekday tags:
- daily_monday_mood_check
- daily_tuesday_tiny_win
- daily_wednesday_real_one
- daily_thursday_laugh
- daily_friday_flirty
- daily_saturday_side_quest
- daily_sunday_slow_burn
Weekday lineup:
| Day | Theme | Purpose |
|---|---|---|
| Monday | Mood Check Monday | Pick what kind of day the relationship needs |
| Tuesday | Tiny Win Tuesday | Pick one small thing to do together |
| Wednesday | Real One Wednesday | Pick the most honest answer |
| Thursday | Laugh It Off Thursday | Pick something silly or low pressure |
| Friday | Flirty Friday | Pick something playful, sweet, or date like |
| Saturday | Side Quest Saturday | Pick a tiny adventure or date idea |
| Sunday | Slow Burn Sunday | Pick a warm intimacy or closeness answer |
Daily Pack Writing Workflow
Do not write 500 daily questions in one pass.
Use this process:
- Write 20 to 30 sample questions for one weekday.
- Review tone before scaling.
- Remove therapy coded words.
- Confirm every option answers the prompt.
- Check opening variety.
- Check exact option list reuse.
- Only then write the next batch.
- After all weekdays are drafted, read 30 random questions aloud.
- Reject the pack if it sounds repetitive, clinical, generic, or AI written.
Emotional Mix
Every standard category should roughly include:
- 35 percent playful
- 25 percent everyday relationship
- 20 percent meaningful conversation
- 10 percent future dreams and planning
- 10 percent deeper vulnerability
Never group heavy questions together.
Every category must include lighter moments.
Daily packs should lean more playful, practical, and low pressure than standard category packs.
Rewrite Rules
For every category:
- rewrite every question from scratch
- keep category ids unless renaming is planned with app code
- use conversational language
- avoid therapy wording
- avoid survey wording
- avoid repetitive templates
- use 4 to 6 options for choice questions
- keep written questions rare
- add fun to every category
- make every category feel different
- reject questions that feel generated
Rewrite Order
- fun
- date_night
- quality_time
- communication
- gratitude
- future
- home_life
- values
- stress
- money
- boundaries
- conflict
- conflict_repair
- trust
- rebuilding_trust
- difficult_conversations
- emotional_intimacy
- physical_intimacy
- couple_intimacy
- sex_and_desire
- sexual_preferences
- daily_single_choice_weekly_v1
Serious Category Rule
Serious categories still need warmth, humor, and relief.
Do not make conflict, trust, boundaries, or intimacy feel like couples counseling homework.
Use soft, safe, specific wording.
Validation Rules for Standard Packs
Every standard rewritten pack must pass:
- valid JSON
- exactly 250 questions
- exactly 75 free
- exactly 175 premium
- exactly 140 multi_choice
- exactly 50 single_choice
- exactly 35 scale
- exactly 15 this_or_that
- exactly 10 written
- at least 76 percent choice based
- valid depth values only: light, medium, deep
- no numeric depth values
- no malformed keys
- no duplicate ids
- no duplicate question text
- every choice question has options
- every scale question has scale configuration
- every written question has answer configuration
- no placeholder text
- no therapy worksheet tone
- no obvious AI repetition
Validation Rules for Special Daily Packs
Every daily single choice weekday pack must pass:
- valid JSON
- metadata counts match actual counts
- exactly 500 questions unless a new documented app decision changes the count
- exactly 75 free questions
- exactly 425 premium questions
- every question is single_choice
- every choice question has 4 to 6 options
- every question has exactly one weekday tag
- no duplicate ids
- no duplicate question text
- no exact repeated option lists
- no placeholder text
- no therapy worksheet tone
- no wellness survey tone
- no obvious AI repetition
Final Rule
The schema only proves the file can load.
It does not prove the questions are good.
Quality review is mandatory before commit.