Time Controls

Custom Range:
├── Start: January 1, 2024
└── End: January 31, 2024

Include: All events from Jan 1 00:00 to Jan 31 23:59

Rolling 30 days:
├── Always shows the last 30 days
├── Updates automatically each day
└── Great for dashboards

✅ Good combinations:
├── Last 24 hours → Hourly granularity
├── Last 7 days → Daily granularity
├── Last 30 days → Daily or weekly
├── Last 90 days → Weekly granularity
├── Last 365 days → Monthly granularity

❌ Poor combinations:
├── Last 365 days → Hourly (too many data points)
├── Last 24 hours → Monthly (not enough data)

Last 7 days vs Previous 7 days:
├── Current: 1,234 swaps
├── Previous: 1,100 swaps
└── Change: +12.2%

January 2024 vs January 2023:
├── Current: 45,000 users
├── Last year: 28,000 users
└── Change: +60.7% YoY

Compare:
├── Period A: Dec 1-31, 2023 (holiday season)
├── Period B: Nov 1-30, 2023 (pre-holiday)
└── See: Holiday impact on metrics

Session Definition:
├── Session starts: First event from user
├── Session continues: Events within 30 min of last event
├── Session ends: 30 min of inactivity
└── New session: Activity after 30 min gap

User Activity Timeline:
├── 10:00 - page_view (Session 1 starts)
├── 10:05 - button_click (Session 1 continues)
├── 10:15 - swap_completed (Session 1 continues)
├── 10:45 - (30 min idle)
├── 11:00 - page_view (Session 2 starts)
└── 11:30 - session ends

Result: 2 sessions

Filter: timestamp last 30 days
Filter: timestamp between "2024-01-01" and "2024-01-31"
Filter: timestamp since "2024-01-01"

Filter: first_seen last 7 days
→ Only new users from this week

Filter: last_seen before_the_last 30 days
→ Users who haven't been active in a month

Default Property: timezone
Example values: "America/New_York", "Europe/London", "Asia/Tokyo"

Swaps by day of week:
├── Monday: 15%
├── Tuesday: 16%
├── Wednesday: 17%
├── Thursday: 16%
├── Friday: 18%
├── Saturday: 10%
└── Sunday: 8%

Insight: Weekday trading dominates

Swaps by hour (UTC):
├── 00-04: Low activity
├── 04-08: Rising (Asia)
├── 08-12: Peak (Europe)
├── 12-16: High (US East opens)
├── 16-20: Peak (US full)
└── 20-00: Declining

Insight: Schedule campaigns for peak hours

Monthly active users:
├── Jan: 45K
├── Feb: 48K
├── Mar: 52K
├── Apr: 58K
├── May: 55K
├── Jun: 50K
├── Jul: 48K
├── Aug: 51K
├── Sep: 60K
├── Oct: 65K
├── Nov: 70K
└── Dec: 62K

Insight: Q4 is your peak season

User clicks "Swap" at 10:00:00 (event_time)
Transaction confirms at 10:00:45 (block_time)

Events use event_time
Transactions use block_time

Dashboard: Weekly Overview
Global Time: Last 7 days

All charts on dashboard use "Last 7 days"
Override individual charts if needed

Dashboard Time: Last 7 days

Chart 1: Uses dashboard time (7 days)
Chart 2: Uses dashboard time (7 days)
Chart 3: Custom time (Last 30 days) ← Override

Question: What happened today?
Time: Today, hourly granularity

Question: How are we trending this month?
Time: Last 30 days, daily granularity

Question: Year-over-year growth?
Time: Last 365 days, monthly granularity

✅ Good comparisons:
├── Week-over-week (similar periods)
├── Month-over-month (business rhythm)
├── Year-over-year (seasonal adjustment)

❌ Poor comparisons:
├── Monday vs Sunday (different behavior)
├── Holiday vs normal week (outliers)
├── Launch week vs normal (unusual activity)

If analyzing "Today":
├── Events appear in real-time
├── Transactions may have confirmation lag
├── Some aggregations update hourly

Recommendation: Use "Yesterday" for complete daily data

Daily view: Volatile, hard to see trend
Weekly view: Smoother, clear trend visible