+220% net subscriber growth and +29.9% customer lifetime value, while spending less on acquisition
The Challenge
When we first spoke with the Mersey Raw team, the frustration was palpable.
They'd built a product customers genuinely loved. Reviews were strong. Repeat buyers were coming back. The subscription model made sense for a consumable product. On paper, everything should have been working.
But the numbers told a different story.
Acquisition costs had jumped 30% in six months. The paid media that once delivered affordable customers was now squeezing margins on every first order. Site traffic was down. And despite being a "subscription-first" brand, subscriptions weren't growing fast enough to offset the pressure.
The team could feel the squeeze, but couldn't pinpoint exactly where the leaks were.
Here's what was actually happening under the hood:
Customers were signing up, but not converting to subscribers at the rate they should have been. Subscribers were churning early in the subscription lifecycle at alarming rates. The Klaviyo account was taking credit for subscription renewal revenue, making it impossible to know what was actually working. And the email deliverability score had slipped to 60, meaning a significant chunk of emails weren't even reaching inboxes.
Meanwhile, every customer - whether a first-time browser, a one-time buyer, or a loyal subscriber - was receiving essentially the same messaging.
The raw feeding category adds another layer of complexity. This isn't like buying socks. Switching a dog to raw food is a considered decision. Customers need education before they buy, reassurance after they buy, and ongoing support to stick with something unfamiliar. Without that, they drift away.
The real question wasn't "how do we get more customers?" It was "how do we make the customers we already have worth more?"
Everboost are the best of all the agencies we've used and runs a tight ship. I have thoroughly enjoyed working with the wider team.
The Turning Point
We started where we always start: understanding where customers were falling out of the journey and why.
The diagnosis revealed five priority areas, where fixing them would create the biggest impact fastest:
The sequencing mattered. Quick wins (offer testing, flow fixes) would show results within weeks and build momentum. Foundational work (quiz infrastructure, rewards program) would take longer but compound over time.
What We Did
The existing popup offered 15% off the first two orders (a decent offer), but it didn't collect any information about the customer or their dog. We tested alternatives: free treats, free shipping, different discount structures.
What we learned was customers wanted to try the product before committing to a subscription. A high-consideration category requires trust first. We landed on 15% off the first order, plus new fields collecting dog size and raw feeding experience. This data became the foundation for everything downstream.
For returning visitors who'd already seen the popup, we introduced a path to a nutrition quiz, giving them a reason to re-engage rather than bounce again.
The quiz asked about their dog: weight, allergies, feeding experience, goals. The output was a personalised feeding plan with clear product recommendations and a natural path to subscription.
But the quiz alone wasn't enough. We built flows for people who abandoned after completing the quiz (addressing their specific objections) and for those who finished but didn't convert (using their quiz answers to tailor the follow-up).
The quiz became our highest-converting mid-funnel asset. It transformed anonymous traffic into known prospects with stated needs.
The welcome flow now branches based on feeding experience and dog size - a first-time raw feeder sees completely different content than someone transitioning from another brand.
The abandoned cart flow stopped being purely transactional ("You left something behind!") and started educating. In this category, uncertainty kills conversions more than price. We addressed transition concerns, safety questions, and convenience objections.
The post-purchase flow was redesigned to eliminate buyer's remorse, set realistic expectations (yes, there might be a transition period), and collect feedback early. OTP customers got pushed toward subscription; subscribers got a dedicated onboarding experience recognising their commitment.
The replenishment flow became smarter - timing adjusted based on order size and dog size, because a Chihuahua owner and a Great Dane owner have very different reorder windows.
We stopped treating the list as one audience. Prospects, one-time buyers, and subscribers now receive fundamentally different campaigns with different offers:
Subscribers get Loop-triggered campaigns with subscriber-only perks - no discounting that undermines their commitment. OTP customers get subscription conversion incentives. Prospects get trial bundles and free shipping offers.
We also layered engagement windows to measure incremental lift and protect deliverability. The result: we increased send volume by 27% while improving engagement metrics across the board.
For BFCM, we tested direct mail targeting unsubscribed and unengaged profiles - people we couldn't reach via email or SMS. It converted at 22.6% with a 16.6x ROAS.
Month two was the danger zone. Subscribers were churning before the habit formed. We extended the discount through the second order, then stepped it down, giving customers two months at the better rate before the price adjusted.
We introduced a rewards program timed to common churn points, added a benefits page that appears when someone tries to cancel, and built in-portal upsells to increase order value.
Campaign cadence to subscribers was refined to stop over-emailing, which was causing unnecessary cancellations.
A Look Inside
From data-capturing popups to lifecycle-segmented email campaigns — every touchpoint was redesigned with intent.
The Results
Where they were: Rising acquisition costs, stagnant subscription growth, leaking customers at every stage, and no clear picture of what was working.
Where they are now:
Mersey Raw's subscription business is fundamentally stronger. The engine now compounds, with every new customer acquired being worth significantly more than before.
The pressure from rising acquisition costs hasn't disappeared, but it matters less now. When each customer generates meaningfully more revenue over their lifetime, the math on acquisition changes. Mersey Raw can now afford to acquire customers that would have been unprofitable before.
The subscription base is no longer a leaky bucket. With churn reduced and upsells working, recurring revenue is growing faster than new customer acquisition, exactly what a subscription business needs.
And perhaps most importantly, they now have visibility. Clean attribution means they can see what's actually driving revenue. Segmented reporting means they can spot problems before they become crises.
What This Means for Similar Brands
If you're running a £2M–£50M CPG brand with subscriptions or repeat purchases, here's what transfers:
If subscription renewals are inflating your campaign attribution, you're optimising blind. Clean it up before you change anything else.
Every zero-party data field should unlock either a personalisation opportunity or funnel insight downstream. If you're not going to use it, don't ask for it.
A prospect, a one-time buyer, and a subscriber have fundamentally different needs. Same message to all three is money left on the table.
If your product requires behaviour change, your abandoned cart flow should teach, not just offer 10% off.
Use cohort data to find when customers leave, then build something to catch them.
Moving from 60 to 82 means dramatically more emails reaching inboxes. That's pure incrementality.
The profiles you can't reach via email or SMS still have mailboxes. Test it during peak periods.
Ready to Talk?
If you're a £2M–£50M CPG brand and retention feels like it's underperforming, book a discovery call.
If it sounds like a fit, we'll offer a comprehensive audit where we walk through your current setup, identify the biggest leaks, and show you what a 6-month roadmap could look like.