Before I recommend anything, I map the customer journey across five stages — and mark what's working, what's leaking, and what I can't see from outside. Here's what that looks like in practice.
Working
Leaking
Need to confirm
01Visitor
Concern-based catalog. Products organised by what customers are experiencing, not by SKU type — already rare for this category.
No guided product finder. A first-time visitor self-navigates a deep catalog. Most won't.
↓
02Subscriber
Popup captures species at signup. Zero-party data is being collected at the right moment.
Welcome flow stops at day one. Three emails arrive within an hour, then silence for the next 30 days.
Is the species data being used downstream, or only for the first email's imagery?
↓
03First-time buyer
Subscribe-and-save offered at checkout. Subscription option is clearly available.
Is there a browse-abandonment flow, or only cart-abandonment?
↓
04Repeat buyer
Cross-sell happens by chance. Customers who bought one product are statistically likely to need another — but nothing routes them there.
Is there a replenishment reminder timed to the supplement cycle?
↓
05Loyal advocate
Rewards program runs. A real persistent loyalty mechanic — more mature than most at this size.
No referral mechanic. The cheapest source of new customers — loyal ones — isn't being asked.
The wedge
Acquisition is working. The matching layer is missing.
The infrastructure is here — concern-based catalog, signup segmentation, subscription, rewards. What's missing is the routing between subscriber and first purchase: the welcome flow that uses the data the popup already captured. That's usually the highest-leverage fix at this stage, but the conversation tells us for sure.
If we spoke for 20 minutes
I'd want to understand the system before suggesting changes to it.
Most of what's wrong on this map is what I can't see from outside. A short call walks through the questions together — and whether or not a project comes from it, you'll have a structured view of where the leaks are.