Case study · Health & Wellness
NutriBlend: fixing the churn window instead of buying more subscribers
A UK supplements brand watching acquisition cost climb past lifetime value. The churn was not random — it clustered in months two and three, for a reason the data made obvious once anyone looked.
PLATFORM
INDUSTRY
MARKET
NutriBlend’s core subscription range — where the replenishment interval and the actual consumption rate had drifted apart.
Results snapshot
What changed in two quarters
Longer subscriber life meant acquisition budget could be raised rather than cut.
+29%
Subscriber lifetime value
Measured on 12-month cohorts
-34%
Month-three churn
The window where cancellations clustered
+18%
Revenue per email
Across the lifecycle programme
11 wks
To payback
On the engagement fee
Lifetime value here is net of refunds and payment failures, not gross subscription value at signup.
About the client
A subscription-led supplements brand
NutriBlend sells powders, capsules and gummies on a subscription model, direct to consumers in the United Kingdom.
NutriBlend built its business on subscription from the start. The acquisition side worked well: the creative was strong, the offer converted, and new subscribers came in at a cost the founders were happy with in year one.
By the time we were introduced, that had stopped being true. Media costs had risen, the offer had been discounted to compensate, and the lifetime value assumption underpinning the whole model had never been re-tested against what subscribers actually did.
- WooCommerce storefront with a subscriptions extension
- Around forty SKUs across powders, capsules and gummies
- Roughly three quarters of revenue on subscription
- Lifecycle email running on a template set never revisited since launch
The challenge
Acquisition cost had overtaken lifetime value
The brief was to lower cost per acquisition. The actual problem was at the other end of the funnel.
Cancellations were not spread evenly across the subscriber base. They clustered sharply in months two and three — and the shape of that curve pointed at the delivery schedule rather than at the product.
Every subscription shipped on a fixed 30-day cycle regardless of what was in it. For the powders, a tub lasted closer to 40 days. Subscribers were accumulating unopened product, noticing around the third delivery, and cancelling.
- One fixed 30-day cycle applied to every product regardless of size
- Churn concentrated in months two and three, not spread evenly
- Onboarding email sequence ending after day seven
- Win-back campaigns firing 90 days after the decision was already made
Our analysis
The interval, not the product
Consumption data existed. Nobody had compared it to the shipping schedule.
We rebuilt the cohort view by product and by month, then laid the cancellation curve over the delivery calendar. The overlap was almost exact: for every SKU where the pack size outlasted the cycle, churn spiked one delivery after the surplus became visible in a customer’s cupboard.
The second finding was cheaper still to fix. The win-back programme fired at 90 days, long after the customer had mentally moved on. The real decision window was the fortnight before the third delivery, and nothing was reaching them in it.
- Cancellation curve mapped against the delivery calendar per SKU
- Pack size outlasting the cycle on roughly half the range
- The real decision window found two weeks before delivery three
- Payment failures quietly counted as churn, inflating the problem
Lowering acquisition cost would have bought more subscribers into the same leaking bucket.
Strategy
Make the schedule match the customer
Three commitments, in the order they had to happen.
Ship on real consumption intervals
- Cycle length set per SKU, not once for the catalogue
- Customers given easy control over timing before they cancel
- Surplus treated as a signal to skip, not a reason to churn
Reach people inside the decision window
- Onboarding extended across the first three deliveries
- Pre-delivery prompts before the fortnight that matters
- Win-back rebuilt around the real churn window
Report on value that survives
- Lifetime value measured net of refunds and failed payments
- Payment failures separated from genuine cancellations
- Acquisition budget set against cohort value, not signup count
None of this required a new subscriptions platform. It required the schedule to reflect how the products are actually used.
What we did
The work, service by service
Four service lines, one team, one shared plan.
RETENTION
Onboarding stretched from seven days to cover the first three deliveries, with pre-delivery prompts landing inside the window where subscribers actually decide.
- Onboarding extended across deliveries one to three
- Pre-delivery skip and adjust prompts added
- Win-back moved from day 90 to the real decision window
CONVERSION
The account area offered one obvious action: cancel. Adding skip, delay and swap as equally visible options converted a large share of would-be cancellations into pauses.
- Skip, delay and swap surfaced alongside cancel
- Cancellation flow rebuilt to offer the right alternative first
- Failed-payment recovery flow added
CREATIVE
Usage guidance, dosing explanations and routine content replaced discount-led emails, which had been training subscribers to wait for offers.
- Usage and routine content replacing discount emails
- Product education mapped to each delivery
- Creative refreshed on a monthly cadence
MARKETING
Once subscriber life extended, the affordable acquisition cost rose. Budget went up rather than down — on the segments whose cohorts actually justified it.
- Target acquisition cost rebuilt from cohort value
- Spend concentrated on the longest-lived segments
- Discount depth at signup reduced
Implementation
How it was sequenced
Fourteen weeks, ordered so the retention fixes landed before any acquisition spend increased.
WEEKS 1-2
Separate the signals
Failed payments split out from genuine cancellations, and lifetime value rebuilt net of refunds. The churn number turned out to be smaller and more fixable than reported.
WEEKS 3-6
Fix the schedule
Cycle lengths set per SKU and existing subscribers migrated with a clear explanation rather than a silent change.
WEEKS 7-10
Rebuild the lifecycle
Onboarding, pre-delivery prompts and the new cancellation flow launched together, since each depends on the others to work.
WEEKS 11-14
Re-open acquisition
With cohorts holding, target acquisition cost was raised and budget increased on the segments the data supported.
Migrating existing subscribers to a new cycle was the risky step. Explaining it plainly, rather than changing it quietly, is why it did not trigger cancellations.
Results
The KPI breakdown
Twelve-month cohorts, compared against the equivalent cohorts a year earlier.
Metric
Before
After
Change
Subscriber lifetime value
£118
£152
+29%
Month-three churn
19.2%
12.7%
-34%
Revenue per email sent
£0.44
£0.52
+18%
Average subscription length
4.1 mo
5.6 mo
+37%
Skip rate (in place of cancel)
2%
14%
+12pt
Failed-payment recovery
31%
68%
+37pt
Signup discount depth
25%
15%
-10pt
Skip rate rising is a good outcome here: a paused subscriber is worth far more than a cancelled one.
Supporting evidence
Before and after, side by side
The operating changes behind the numbers above.
BEFORE
• One 30-day cycle applied to every product regardless of pack size
• Onboarding stopping at day seven
• Cancel as the only visible option in the account area
• Win-back firing at day 90, long after the decision
• Failed payments counted as churn, inflating the problem
AFTER
• Cycle length set per SKU against real consumption rates
• Onboarding covering the first three deliveries
• Skip, delay and swap offered alongside cancel
• Prompts landing in the fortnight before delivery three
• Payment failures separated out and actively recovered
The cheapest single change was separating failed payments from cancellations. It cost nothing and immediately made the rest of the numbers honest.
We were about to cut acquisition spend. It turned out we could afford to raise it — we just had to stop shipping people a tub they had not finished.
Founder
NutriBlend
Services used
The service lines behind this engagement
Four of our nine service lines worked on this account, under a single plan and a single owner.
Lifecycle, onboarding, pre-delivery prompts and win-back — the service that owned the change that mattered most here.
Account area and cancellation flow, where a pause became easier to choose than a cancel.
Usage and routine content that gave the lifecycle programme a reason to be opened.
Acquisition budget reset against cohort value once subscriber life extended.
Classification
Platform, industry and market
Every case study on this site is classified on four axes so you can find the one that matches your situation.
This engagement is filed as Email & Retention + WooCommerce + Health & Wellness + UK. It appears on each of those four pages.
Related work
Other case studies worth reading
Matched on shared services, platform or market.
Is your churn a product problem or a schedule problem?
For most subscription brands it is the second, and it is far cheaper to fix. Tell us what you sell and how often you ship it, and we will show you where your cancellations actually cluster.