.avif)
Primal Queen scaled subscription revenue from $2M to $100M+
Primal Queen scaled subscription revenue from $2M to $100M+
.avif)
.png)
KittySpout sells cat water fountains and the filters that keep them running. As subscriptions became the core of the business, the team wanted deeper visibility and more active control over churn and payment recovery.
After moving to Loop, they rebuilt two things alongside their Loop CSM: a personalized, continuously A/B-tested cancellation flow, and a dunning strategy they optimized over several iterations.
KittySpout measured churn fall from 13.83% (March 2026) to 11.7% (June 2026) and first-cycle payment recovery rise from 48.8% to 57% (about a 17% increase).
Results reflect KittySpout's measured experience. Outcomes vary by brand, vertical, product, and implementation.
About the brand
KittySpout makes wireless, self-filtering water fountains for cats plus the replacement filters that make it a subscription. The fountain is the front door; the filters are why customers come back every month.
As KittySpout grew, subscriptions became the core of the business - the fountains bring customers in, the filters keep them coming back. The team wanted a platform that could do more with that base.
They were looking for three things.
One thing was non-negotiable: the switch couldn't disrupt the experience their customers already trusted - no new logins to figure out, no confusion, just a smoother portal on the other side.
KittySpout rebuilt cancellation from a single "cancel" button into a multi-stage flow - one personalized to be relevant to each subscriber and the reason they're leaving, and built to keep improving through testing at every stage. The goal: lift the save rate.
Stage 1 - A save page they A/B test, down to the button.
Before anyone can cancel, the flow opens with a short video from a veterinary specialist and a reason to stay. It runs as a live experiment: a calm "Your price is locked" control against an urgent "WAIT - this is important" variant. They even A/B test the CTA copy - "KEEP PROTECTING MY CAT" vs "I want to protect my cat." The winning variant is saving 12.08% of attempts vs 8.43% on the control variant.
Stage 2 - A different page for every reason.
Pick a cancellation reason and the flow adapts with tailored copy and alternatives for each. "I have enough filters" gets "You don't have too many filters. You have too many deliveries," plus a nudge to change frequency or skip - not a generic "are you sure?" Each reason has its own page and its own measured save rate with the strongest saving ~8–10% of attempts.
Stage 3 - A reason-matched offer, also tested.
Still leaving? The final stage serves an offer matched to the reason - a delay + discount for the "too many filters" crowd, a straight discount elsewhere - with accept rates tracked and the offer pages themselves A/B tested for lift.
Stage 4 - A winback offer if they cancel anyway.
KittySpout was early to Loop's new winback stage: a one-time offer (30% off the next three orders) surfaced the moment a subscription cancels with a banner that runs only until it expires. One clean, time-boxed reason to come back.
In short: every stage is tailored to the subscriber and tested against a version of itself so the flow saves more subscribers over time.
The payment-recovery strategy: a dunning system built through iteration
Failed payments were the quieter leak and the one KittySpout and its Loop CSM optimized in steps, not one setting.
They started at 8 retries. Then moved to a 10-retry schedule - silent retries 1–5, notified retries 6–10 with gradual spacing so early attempts run quietly and later ones reach the customer at a sensible cadence.
Then they went further, to a 14-retry schedule: silent retries 1–4, notified retries 5–14 with gradual spacing, staggered across the billing cycle to catch the moment funds are available.
Alongside the retries, they turned on three more ways to recover:
KittySpout came to Loop with an ask: integrate AfterSell to power their post-purchase upsells and lift AOV from newly acquired customers. Loop's team handled the full integration and setup in roughly a week to ten days timeframe.
Post-purchase upsells appear right after checkout, on the order-confirmation step, so they don't touch conversion on the main purchase - the sale is already made. Payment is already on file, so accepting is one tap. And the offer lands at the moment a new customer's intent and trust are highest.
For KittySpout, that means putting the natural next thing in front of a new fountain buyer - extra filters, or the new supplement line - growing order value from day one with no extra ad spend. A quick turnaround on exactly what the merchant asked for.
None of this was a one-time setup. The retry schedule went from 8 → 10 → 14. The benefits page, CTAs, and offers are still being tested. That's the point - Loop's CSM works the configuration continuously, treating the merchant's ROI as the job. KittySpout's numbers moved because someone kept tuning the system alongside them.
After rebuilding both flowS, KittySpout measured:
The recovery line climbs as the dunning strategy is tuned - each iteration (more retries, backup payment, one-click card updates) compounds into more revenue recovered.
Results reflect KittySpout's measured experience. Outcomes vary by brand, vertical, product, and implementation.
.png)
"Loop has been amazing in helping us reduce our churn and scale our program. They act as a partner vs just a tool you can use which has made all the difference in working closely with the team to improve our retention and scale our business because of it."
Retention isn't one save offer and recovery isn't one retry. Both are systems you keep tuning - a personalized cancel flow you test at every stage and a dunning strategy that expects cards to fail and engineers around it. KittySpout treated them as products to iterate on, with Loop team that kept optimizing alongside them. That's what moved churn down and recovery up and keeps moving them.
.png)
.png)
.avif)
