Based on marketing data analysis, this report presents the strategy, initiatives, and roadmap to grow TOMOGO! bookings from the current 1.8/day to 10/day. It covers areas relevant not only to the Ads team but also to Product, Engineering, Creative, and Tour Design.
The initiatives outlined here cannot be executed by any single team alone. CRM implementation requires the Engineering team, creative production requires the Design team, messaging validation requires the Tour Design team, and integrating all of these requires cross-functional Product × Marketing decision-making. For each section's initiatives, please discuss role assignments within your teams, agree on priorities and owners, and then execute.
Creative Analysis & Production Guidelines (JP/EN) / Product Improvement Recommendations (JP/EN)
| # | Reason | Details |
|---|---|---|
| 1 | Data Accumulation → Better Analytics | Current sample sizes are too small for statistically significant analysis. More bookings enable reliable cohort analysis and A/B testing |
| 2 | Meta Ads Learning Data | Meta ad optimization requires a minimum of 50 CV events per week. We are far below this threshold, preventing the algorithm from learning effectively |
| 3 | Faster PDCA Cycles | Booking volume shows weekly trends clearly. Revenue is affected by price fluctuations, making it slower to measure initiative effectiveness |
Growing booking volume is not a "precursor" to revenue growth — it is the foundation. Only with sufficient data can we optimize unit economics and LTV.
The Install → Booking correlation is r=0.66-0.76 (Post CA flow change, 7-day lag). When Installs increase, Bookings follow approximately 2 weeks later.
Bring users in. There is an approximately 2-week lag before impact materializes.
Pipeline BuildingConvert the users we've acquired into bookings. Immediate impact once implemented.
Quick WinCartAdd (r=-0.82) and Purchase (r=-0.87) are negatively correlated with bookings. This is caused by a loop in the Tour Leader selection screen. These must never be used as CV objectives.
Increase Installs (Lever 1) and raise conversion rates for those users (Lever 2). Running both levers simultaneously is the path to 10 bookings/day. CRM delivers immediate impact; ads take effect after ~2 weeks.
75.7% (305/403) of First Bookings used no coupon. It is highly likely that users are simply unaware the coupon exists.
| Coupon | Distribution Channel | Usage Rate | Structural Issue |
|---|---|---|---|
| ebr20 | In-app chat (Admin → Guest) | 9.2% | Only visible if user opens the chat |
| WEL20 | Welcome email | 2.5% | Email volume dropped 82% after CA flow change |
Improvements: Display on profile screen, auto-apply in booking flow, "20% OFF first booking" on LP, Push reminder notification
MoEngage is currently used for re-engagement only: geo-triggers, browse abandonment, Spring Promo. Coupon distribution via Push has not been implemented.
| Initiative | CVR | Comparison | Monthly BK | Projected Uplift |
|---|---|---|---|---|
| Push Overall | 9.1% | 2.5x vs overall 3.6% | 5 | - |
| Cart Abandonment Push | 25% | Highest CVR | 4 | 40 eligible → +9 |
| First Coupon Notification Push | 14.3% | High CVR | 7 | 70 eligible → +9 |
Spring Campaign: delivered to 2,400-2,900 recipients, Open Rate 32%. Bookings increased +19-48% within 14 days of send.
After the 6th send, effectiveness drops sharply (pipeline exhaustion). Email is a booking "trigger," not a "pipeline." When new user inflow (Installs) stops, email effectiveness evaporates.
Welcome Email: Due to the CA flow change, weekly recipients dropped from 370 → 30, an 82% decrease. Restoring the CA flow is a prerequisite for email initiatives.
| Initiative | Monthly BK Uplift | Prerequisite |
|---|---|---|
| ebr20 UX Improvement (Profile / Booking Flow Display) | +5-8 | Engineering implementation |
| Cart Abandonment Push Expansion | +9 | Relax MoEngage trigger conditions |
| First Coupon Notification Push (New) | +9 | New MoEngage flow design |
| Email Delivery Restoration (CA Dependency Fix) | +3-5 | CA flow restoration or alternative |
| Total | +26-31/month |
| CV Objective Candidate | Correlation r | Weekly Volume | Verdict |
|---|---|---|---|
| ContentView (View Tour Detail Page) | r=0.62-0.71 | 54/week | Recommended |
| App Search | r=0.69 | Insufficient | Low Volume |
| CartAdd | r=-0.82 | - | NG |
| Purchase | r=-0.87 | - | NG |
Set ContentView (View Tour Detail Page) as the CV objective. It has a positive correlation + 54 events/week, providing sufficient volume for algorithm learning. CartAdd/Purchase must never be used as CV objectives due to their negative correlation.
| Acquisition Path | In-App Event/Install | Rating |
|---|---|---|
| Via WEB Conv | 568-760% | High Quality |
| Direct Install | 96-127% | Standard |
WEB-sourced Installs show ~7x higher in-app engagement than Direct Installs. Users who view tour information on the web before installing have clear booking intent.
ENGAGEMENT objective: 0 Purchases. While it has awareness value, budget allocation should be reduced from 44% → 10-15%.
| Parameter | Recommended Setting | Rationale |
|---|---|---|
| Age | 25-54 | Highest booking rate zone |
| Gender | All genders | Age has more impact than gender on booking rates |
| Country/Region | AU priority → US | AU CPO is half that of US |
Recovering 1,000 Installs/month → +30-50 BK/month increase expected after ~2 weeks (based on correlation r=0.66-0.76).
| Placement | Normal Day CPC | Good Day CPC | Difference |
|---|---|---|---|
| Stories | ¥27-37 | ¥19-26 | -30% |
| Reels | ¥70-92 | ¥49-64 | -30% |
| Feed | ¥100-170 | ¥70-119 | -30% |
Evaluate by In-App Event/Install rate, not CPI. Since post-install in-app activity drives bookings, we prioritize Install "quality" over volume alone.
| Tier | Criteria | Action |
|---|---|---|
| S Tier | High In-App Rate + Low CPI | Scale budget |
| A Tier | Above-average In-App Rate | Continue delivery |
| B Tier | Standard | Room for improvement |
| C Tier | Low In-App Rate | Consider pausing |
| Placement | Recommended Share | Rationale |
|---|---|---|
| Reels | 40-50% | Video impact + reach |
| Feed | 30-40% | Information density + WEB traffic |
| Stories | 15-20% | Lowest CPC + frequency |
The 2/20 Create Account flow change significantly altered the correlation structure.
| Comparison | Pre (Before Change) | Post (After Change) |
|---|---|---|
| CA Required At | App browsing start | Just before booking |
| Install → BK Correlation | r=0.40-0.55 | r=0.66-0.76 |
| Weekly CA Count | 370 | 30 |
| Welcome Email Delivery | 370/week | 30/week (82% drop) |
The stronger Install → BK correlation in the Post period confirms that Install volume itself is critical. However, pushing CA later means we lose early User ID acquisition, shrinking the pipeline for cohort analysis, Push, and email.
Explore early CA without sacrificing browsability. For example: prompt a lightweight CA (email-only capture) after the first tour view. Once we have User IDs, we unlock cohort analysis, Push notifications, and email delivery. The coupon distribution pipeline also depends on CA.
| Segment | In-App Event/Install | Purchase | Daily Budget | Verdict |
|---|---|---|---|---|
| SegA | 2,967% | 4 | ¥20K | Continue |
| SegB | Low | 0 | - | Pause Justified |
| SegC | Low | 0 | - | Pause Justified |
SegA's In-App Event/Install rate of 2,967% is exceptional. Continue the ¥20K daily budget and use this segment's characteristics to inform targeting across other campaigns.
Product improvement opportunities identified from marketing data. Engineering team collaboration required.
| Issue | Data Evidence | Impact | Owner |
|---|---|---|---|
| APP Onboarding Improvement | 7x In-App rate gap: WEB vs Direct Install | High | Engineering + Design |
| TL Selection Loop Fix | CartAdd r=-0.82 | High | Product + Engineering |
| WEB Funnel Improvement | WEB 91% drop-off | Medium-High | Marketing + Engineering |
| Coupon UX Improvement | 75.7% unused | Medium | Engineering + Design |
Users loop on the Tour Leader selection screen, firing multiple CartAdd events. This is the root cause of the negative CartAdd/Purchase correlations. The UX flow itself needs to be redesigned.
WEB-sourced users are 7x more active in-app because they've already engaged with tour content on the web. If we provide Direct Install users with an equivalent initial experience, we can expect significant conversion rate improvement.
Expected Impact: +26-31 bookings/month
| Phase | Timing | Cumulative BK/Month | BK/Day |
|---|---|---|---|
| Current | Now | 42 | 1.8 |
| Phase 1 Complete | Week 2 | 68-73 | 2.3-2.4 |
| Phase 2 Takes Effect | Week 6 | 100-120 | 3.3-4.0 |
| Phase 3 Validation Done | Week 10 | 150-180 | 5.0-6.0 |
| Phase 4-5 | Month 3+ | 300 | 10.0 |
Each Phase requires collaboration across multiple teams. Please discuss role assignments and priorities within your teams, then assign owners and timelines for each Phase before proceeding.
| KPI | Current | Phase 1 Target | Phase 2 Target | Final Target | Data Source |
|---|---|---|---|---|---|
| Daily Bookings | 1.8 | 2.5 | 4.0 | 10 | Admin Raw |
| Weekly Installs | ~10 | Maintain | 250 | 500 | AppsFlyer |
| View Tour Detail Page Rate | - | 40%+ | 40%+ | 50%+ | AF/Amplitude |
| Coupon Usage Rate (First BK) | 24.3% | 35% | 40% | 50% | Admin Raw |
| Push Bookings | 5/month | 15/month | 20/month | 50/month | MoEngage/Amplitude |
| Email Delivery Volume | 30/week | 100/week | 200/week | 500/week | Newsletter |
KPIs should be reviewed weekly and shared across all teams. Track which Phase and initiative each metric change is attributable to, and reflect findings in the following week's actions.