I wrote "10 Things You Need to Run a Successful eCom Growth & Experimentation Program" a while back, diving deep into what makes experimentation programs work.
But after implementing these programs across multiple Subscription eCommerce brands, I've identified five must-have elements that separate winners from the rest.
A North Star Metric keeps your entire company aligned on what provides value to customers while driving business growth. For traditional eCommerce, it might be as simple as customer count or revenue.
But subscription eCommerce? That's a different beast entirely.
The number of subscribers might seem obvious, but it completely neglects non-subscribers – who can represent a massive percentage of revenue.
Your ideal North Star needs to account for both new and returning customers, subscribers and one-time buyers, AOV, LTV, and (ideally) profit margins. If that sounds complex, it's because it is.
But anything less risks neglecting a critical component of your business model and steering you in the wrong direction.
In standard eCommerce, your A/B testing tool usually handles the analytics side. You run an experiment aimed at conversion or AOV, and the tool tells you if you won or lost.
Subscription businesses don't have that luxury.
Here's why: If your experiment successfully increases subscription signups (which typically come with a discount), your AOV will likely decrease. The A/B testing tool will declare your variant a loser, showing decreased revenue per visitor.
But that's not the truth.
To determine the real winner, you need sophisticated LTV analysis – something most testing tools don't provide out of the box. You'll need to develop custom solutions using Google Sheets, Looker Studio, or similar tools that incorporate longer-term value into your decision-making process.
Running effective experiments in subscription eCommerce requires intimate knowledge of your business metrics, including:
These barely scratch the surface. There are dozens more questions you should be able to answer instantly.
This clarity enables you to craft strategies that genuinely unlock revenue and margins rather than superficial metrics that look good in reports but don't build a sustainable business.
Don't take this pillar for granted. Statistics isn't just complex – it's deceptively complex.
Someone on your subscription team needs deep statistical knowledge. Without it, you risk making million-dollar decisions based on faulty interpretations, potentially killing your growth flywheel before it gains momentum.
Unlike one-time purchase businesses, subscription testing operates on an entirely different timeline. What looks like a failed test today could be your biggest win in 60 days when subscription renewals start rolling in.
If you can't properly design cohort analyses and interpret non-binomial metrics across different time horizons, you're essentially gambling with your business's future.
I believe the term "CRO" (Conversion Rate Optimization) does a disservice to what growth teams actually do – and in subscription eCommerce, this narrow focus is even more damaging.
If you concentrate solely on acquisition metrics, you're leaving money on the table.
Start thinking about your business more like a SaaS product than traditional eCommerce, and multiple growth levers emerge:
Each of these represents a massive opportunity that most brands leave unexplored.
Intuition can be a powerful tool when you're just starting out, but it will only take you so far.
Once your subscription business matures beyond $5M in revenue, you need to shift your approach and embrace the scientific method. Experimenting your way forward will enable you to:
The alternative? Working tirelessly on initiatives that don't move the needle while watching hundreds of thousands in potential revenue evaporate.
Want to dive deeper into any of these principles? Let me know which one resonates most with your business challenges, and I'll explore it further in my next article.
