Chats / Q2 churn drivers by subscription plan
Churn Predictor

I want to understand Q2 subscription churn for Verdant. Pull from Shopify, cross-reference the Gorgias cancellation reasons, and isolate by plan tier. Focus on mid-tier — the 30-day subscribers spending $50–$120/mo.

JT
09:22
fct_subscription_events attached by auto — 842k rows from Shopify · sampled from the last 18 months of subscription activity. 09:22
Churn Predictor · 09:23

On it. I'll pull Q2 cancellations from subscription_contracts, cluster the Gorgias cancellation reasons against the same customer set, and check whether the Magnesium 30-day default is doing what I think it's doing. Starting the queries now.

Churn Predictor · 09:26

Here's what the data says for Q2, mid-tier subscribers ($50–$120/mo). 2,418 cancellations against a mid-tier base of 10,380 — a 23.3% quarterly churn rate, up from 19.1% in Q1 . The picture is less "customers don't love the product" and more "we're shipping product faster than they can use it."

Three drivers explain most of the lift:

  1. Cadence mismatch on 180-count SKUs. Subscribers on the 30-day cadence for Magnesium Glycinate 180ct, Marine Collagen 30-day jar, and Omega-3 Triple Strength are piling up bottles. 41% of cancellation tickets mention "too much product" or "still have last bottle" — it's the single biggest driver and it is almost entirely self-inflicted by the default cadence we set at checkout .
  2. Winback flow fatigue. The winback-30d-skippers flow is opening at 24% but only converting 0.8% back to an active subscription. It reads like a discount prompt, not a cadence conversation — subscribers who skipped twice don't need a coupon, they need a longer interval .
  3. Paid-social-acquired cohorts churn 1.6× faster. Mid-tier Magnesium subscribers acquired through Meta in Q1 are churning at 27.9% in Q2. The Meta hero creative has been running for 14 weeks and Triple Whale flags creative fatigue, so we're acquiring the wrong shape of customer — lower intent, higher unsubscribe rate .
"I keep getting the Magnesium shipped before I run out — shipping cadence is too aggressive for a 180-count bottle." — Gorgias ticket #VRD-18204, and 863 others almost like it .

Here's the segment view. Churn rate is Q2 cancellations over the mid-tier base of 10,380:

SEGMENTCHURN RATETOP DRIVER
Magnesium Glycinate · 180ct31.4%cadence-mismatch (58%)
Marine Collagen · 30-day jar24.8%cadence-mismatch (47%)
Night Restore18.2%price-sensitivity (29%)

One note on methodology: the cancel_reason_code field is free-text in ~31% of the mid-tier cohort, so I clustered the untagged notes against the tagged ones. The cadence-mismatch cluster absorbed most of the unlabeled rows, so if anything the 58% figure is understated.

Draft a save-play for the Magnesium subscribers who skipped their last two cycles. They're the biggest lost-revenue group — keep it friendly, not desperate.

JT
09:39
Churn Predictor · 09:40

Drafting now. Based on what we just saw, this isn't a discount problem — the winback-30d-skippers flow is already sending 15% off and converting at 0.8%. The lever is cadence, not price. Here's the v1:

Save-play · Magnesium 180ct · 2-skip cadence reset

Trigger: any mid-tier Magnesium Glycinate 180ct subscriber on the 30-day cadence who skipped their last two scheduled shipments and has

PLAN IN PROGRESS · 3 OF 6 COMPLETE
SOURCES
CASCADE can make mistakes. Always verify tool outputs. ⏎ TO SEND · SHIFT⏎ FOR NEWLINE
Maxed Out