Cobalt Robotics: Newer Customers Are Leaving Faster

Marketing
medium40 min0 submissions
Razorpay
Scenario

Cobalt Robotics is a industrial robotics business in Japan with about 60k active customers paying ¥750 per month on average.

Monthly churn has risen from 3.1% to 5% over four quarters. Acquisition has not slowed — the base is still growing — but the finance team has noticed revenue growth decelerating faster than the customer count would suggest.

What the analytics team has produced:

  • Cohorts from 12+ months ago still churn at about 3.1% a month
  • Cohorts acquired in the last two quarters churn at roughly 5%
  • Retention by tenure: month 1 72%, month 3 45%, month 12 29%
  • 32% of customers acquired in the last two quarters came in on a promotional discount, up from a much smaller share previously

The CMO believes the product has got worse. The Head of Product disagrees and points at marketing.

Supporting data

base

arpu monthly
750
active customers k
60

churn

older cohorts pct
3.1
recent cohorts pct
5

acquisition

discount acquired share pct
32

derived hints

monthly revenue at risk
0

retention by tenure pct

month 1
72
month 3
45
month 12
29
Your task

Diagnose the churn increase and recommend a response. Your answer should provide:

  1. Analysis — structure the churn problem and isolate where it actually sits.
  2. Risks — what your diagnosis depends on, and what would disprove it.
  3. Recommendation — specific actions, and what you would measure to know they worked.

State any assumptions you make.

Ready to move forward? Up next: Basil & Co: Build the Line or Keep Buying?Next question
How you'll be graded

100 points, 60% to pass.

  • root cause25
  • recommendation20
  • cohort analysis30
  • problem structuring25
Hint
Reveal suggested structure

Churn by cohort × tenure × acquisition source; early-life vs late-life; mix shift vs product change