Executive Summary
Apr 14 Bookings
0
-3 vs previous day
Apr 14 CA (APP)
2
Weekly avg 7.0 — down 75%
Apr CA-to-Booking Rate
17.6%
Mar 39.2% — down 55%
Apr Traffic-to-Booking Rate
0.4%
Mar 0.8% — halved
Conclusion: The zero bookings on 4/14 were directly caused by a sharp drop in CA (Booking taps) — CA(APP) fell from a weekly average of 7 to just 2 (-75%). Sessions were 142 (normal), but the Booking tap rate collapsed to 1.4% (typically 4-5%). Additionally, April's overall CA-to-Booking conversion rate deteriorated from 39.2% to 17.6%, indicating that account registration is the structural bottleneck.
April 14 — Direct Cause Analysis
Daily Trend (4/8-4/14) — Bookings
Daily Trend — CA (APP)
Detailed Daily Data
| Date | Bookings | Sessions | View Home | CA(APP) | CA(WEB) | Cost | AF Install | Ad CV |
|---|
| 4/8 | 1 | 109 | 164 | 8 | 0 | ¥94,954 | 47 | 0 |
| 4/9 | 3 | 116 | 162 | 7 | 0 | ¥66,795 | 54 | 0 |
| 4/10 | 3 | 137 | 173 | 9 | 0 | ¥88,159 | 57 | 0 |
| 4/11 | 2 | 130 | 134 | 6 | 0 | ¥86,446 | 36 | 0 |
| 4/12 | 0 | 113 | 157 | 9 | 0 | ¥123,822 | 60 | 0 |
| 4/13 | 3 | 140 | 167 | 8 | 1 | ¥98,298 | 30 | 2 |
| 4/14 | 0 | 142 | 147 | 2 | 0 | ¥89,687 | 42 | 0 |
Fact: On 4/14, sessions were 142 (2nd highest of the week) — traffic volume was adequate. However, CA(APP) plummeted from 8 to 2 (-75%). The Booking tap rate was just 1.4% against 147 View Home events (normally 4-5%). Zero bookings from both new and returning users indicates a universal drop in booking intent.
Interpretation: Traffic "volume" was not the issue — traffic "quality" was. While ad-driven installs were 42 (reasonable), the proportion of users with booking intent was unusually low. Note that 4/12 also had zero bookings — two zero-booking days in one week suggests structural vulnerability, not random variance.
April Structural Issue — Funnel Breakdown
Monthly Funnel Comparison (March vs April 14-day)
March CA-to-Booking Rate
39.2%
CA 265 → Bookings 104
April CA-to-Booking Rate
17.6%
CA 102 → Bookings 18
Funnel Flow — CA-to-Booking Collapse
4,447Traffic (Sessions + View Home)
▼
Traffic → CA: 2.3% (March: 2.1%, on par)
102CA (Create Account = Booking Tap)
▼
CA → Booking: 17.6% (March: 39.2% — down 55%)
18Booking Complete
Fact: Traffic-to-CA conversion is at 2.3%, in line with March (2.1%). The critical issue is CA-to-Booking conversion collapsing from 39.2% to 17.6%. Over 82% of users who tapped the Booking button and began account creation failed to complete a booking.
APP / WEB Funnel Breakdown
| Month | WEB Sessions | WEB CA | WEB Bookings | WEB CVR | APP View Home | APP CA | APP Bookings | APP CVR |
|---|
| March | 5,399 | 16 | 26 | 0.5% | 7,434 | 249 | 78 | 1.0% |
| April | 2,004 | 3 | 2 | 0.1% | 2,443 | 99 | 16 | 0.7% |
Interpretation: WEB CVR dropped from 0.5% to 0.1% (-80%); APP CVR declined from 1.0% to 0.7%. WEB performance is particularly concerning — only 2 bookings from 2,004 sessions. WEB CA is just 3, indicating a severe issue with reaching the Booking button on the web experience.
CA (Account Creation) Mechanism & Drop-off
APP Booking Flow
View Tour Detail Page
▼
Tap "Booking" Button
▼
Not Logged In → CA (Account Creation)
▼
Select Timeline → Confirmation Screen
▼
Booking Complete
Key Context: CA = "An unregistered user who tapped the Booking button with intent to book." These users have demonstrated booking intent. Yet 82.4% fail to complete — strongly suggesting that the account registration process itself is the primary drop-off cause.
Issue Separation
| Issue Layer | Metric | March | April | Change | Severity |
|---|
| Traffic Quality | Traffic → CA Rate | 2.1% | 2.3% | +0.2pt | Low |
| CA → Booking Conversion | CA → Booking Rate | 39.2% | 17.6% | -21.6pt | Critical |
| Overall CVR | Traffic → Booking Rate | 0.8% | 0.4% | -0.4pt | High |
SNS Performance — Active Despite Dashboard Gap
| Date | Account | Content | Views | Reach | Saves | Shares |
|---|
| 4/1 | tomogo.official | Asakusa Reel | 7,146 | 5,823 | 133 | 40 |
| 4/3 | tomogo.official | Japan Travel Tips Reel | 13,843 | 9,907 | 661 | 232 |
| 4/3 | tomogo.official | 50K Followers Milestone | 6,209 | 3,407 | 6 | 0 |
| 4/6 | tomogo.official | Toden Arakawa Line | 5,857 | 2,904 | 95 | 22 |
| 4/8 | tomogo.official | Bar Hopping Spots Reel | 13,470 | 10,602 | 493 | 208 |
| 4/10 | tomogo.official | Toden Arakawa Line Reel | 4,627 | 3,776 | 109 | 21 |
| 4/11 | ebifried_ | Bucket List (Collab) | 15,594 | — | 0 | 61 |
| 4/13 | tomogo.official | JR Pass Explainer | 3,671 | 1,758 | 20 | 8 |
| Total | 70,417 | 38,177+ | 1,517 | 592 |
Fact: SNS has been actively managed — 70K+ total views and 1,500+ saves in 14 days. The 4/3 "Japan Travel Tips" reel (V=13,843 / Sv=661) and 4/8 "Bar Hopping" reel (V=13,470 / Sv=493) showed strong engagement. The issue was only that data was not reflected in the dashboard — SNS itself is performing well.
Ad Efficiency Review
Daily Cost vs Bookings / Installs
| Date | Cost | AF Install | Bookings | CPI | Ad CV |
|---|
| 4/8 | ¥94,954 | 47 | 1 | ¥2,020 | 0 |
| 4/9 | ¥66,795 | 54 | 3 | ¥1,237 | 0 |
| 4/10 | ¥88,159 | 57 | 3 | ¥1,547 | 0 |
| 4/11 | ¥86,446 | 36 | 2 | ¥2,401 | 0 |
| 4/12 | ¥123,822 | 60 | 0 | ¥2,064 | 0 |
| 4/13 | ¥98,298 | 30 | 3 | ¥3,277 | 2 |
| 4/14 | ¥89,687 | 42 | 0 | ¥2,135 | 0 |
| Total | ¥648,161 | 326 | 12 | ¥1,988 | 2 |
Interpretation: ¥648,161 spent over 7 days generated 326 installs (CPI ¥1,988) but only 12 bookings — CPA of ¥54,013. Ad CV was zero on 5 of 7 days. Ads are effective at driving installs but not at driving bookings. The disconnect between install volume and booking volume points to a traffic quality issue.
Root Cause Summary
| Factor | Data Evidence | Impact |
|---|
| CA-to-Booking rate collapse |
39.2% → 17.6% (-55%) |
█████ Critical |
| Account registration bottleneck |
84 of 102 CA users dropped off post-registration |
█████ Critical |
| Ad traffic quality mismatch |
326 installs vs 12 bookings (CVR 3.7%) |
████☐ High |
| 4/14 Booking tap rate crash |
Normal 4-5% → 1.4% |
███☐☐ Medium-High |
| WEB CVR collapse |
0.5% → 0.1% (-80%) |
███☐☐ Medium-High |
Recommended Actions (Priority Order)
-
Implement Guest Checkout or Social Login Dev Proposal
Evidence: CA-to-Booking at 17.6% — over 82% drop off at registration. Removing the registration barrier is the highest-leverage fix.
Expected impact: Recover CA-to-Booking rate to 30%+ (+70% improvement)
-
Surface Tour Detail (TL Detail) Earlier in Flow UX Improvement
Evidence: Improve View Home → Booking tap conversion. Aligns with Hypothesis A. Present detailed tour info earlier to solidify booking intent.
Expected impact: Increase Booking tap rate from 4-5% to 6%+
-
Refine Ad Creative for Booking Intent Filtering Advertising
Evidence: 326 installs vs 2 Ad CV. Ads drive installs but not bookings. Shift creative messaging to emphasize "Book an Experience" to attract higher-intent users.
Expected impact: Improve Install-to-Booking rate from 3.7% to 8%+
-
Identify Exact Drop-off Step Post-CA in Amplitude Analysis
Evidence: Unknown where exactly users abandon between CA and booking. Is it email verification? Password setup? TL selection? Confirmation? Pinpointing the step is a prerequisite for targeted fixes.
Timeline: 1-2 days to complete analysis
-
Update Dashboard with SNS Raw Data Operations
Evidence: SNS is active but not reflected in the dashboard, creating a monitoring blind spot that affects decision-making.
Timeline: Same-day fix via GAS update
Measurement Gaps (Data Needed)
| Missing Data | How to Obtain |
|---|
| Step-by-step drop-off rate between CA and booking | Amplitude Funnel: CA → TL Select → Confirm → Booking Complete |
| Logged-in user booking behavior | Amplitude segment: returning user × booking event |
| Booking rate by ad channel | Appsflyer × Amplitude integration (deep link event tracking) |
| SNS-to-booking conversion path | UTM parameter implementation + Amplitude traffic source analysis |