In This Article
When an online store isn't making enough money, the reflex is to go get more traffic. More ads, more posts, more email. Sometimes that's right. More often the store is already getting enough people and losing them somewhere specific, and adding traffic just pushes more of them through the same leak.
Three leaks account for most of it in the small stores I've worked on. None of them show up as a problem on a dashboard. That's exactly why they persist.
Leak one: your checkout works fine on your laptop
Most people browsing your store are on a phone. Most store owners test their checkout on a desktop, because that's where they built it and that's where they work.
So the failures are invisible to the person who could fix them. A form that needs a zoom to read. A card field that pushes the "place order" button below the fold. A discount box that opens the keyboard and hides the total. A shipping selector that resets when someone rotates the phone. None of these throw errors. They just make people give up, and giving up is silent.
Cart abandonment gets discussed as though it's a psychological problem to be solved with urgency timers and exit popups. Often it's a much dumber problem: the checkout is annoying on the device most people are using.
The fix is not clever. Buy something from your own store on your own phone, on mobile data rather than wifi, and pay real attention to where you get irritated. Then do it again on a cheap Android handset, because iPhones hide a lot of sins. Most owners find two or three fixable things in fifteen minutes.
The things worth checking first
- Guest checkout exists. Forcing account creation before purchase is the single most reliable way to lose a first-time buyer.
- The total is honest early. Shipping that appears at the last step is the most common reason people walk, and it feels like a trick even when it isn't.
- Wallets are enabled. Apple Pay and Google Pay remove the worst part of mobile checkout, which is typing a card number one-handed.
- The form doesn't fight the phone. Correct keyboards for each field, autofill that works, no zoom required.
Leak two: you're renting your customers
If most of your sales come through Amazon, Etsy, eBay or a marketplace app, you're not really building a customer base. You're renting access to somebody else's, on terms you don't set.
That trade can be entirely worth making. Marketplaces bring volume you would struggle to reach alone, and plenty of good businesses run mostly on them by choice. The problem is when it's the whole business by accident, and nobody has done the arithmetic on what it costs.
What you're paying is not just the commission. It's that you usually don't own the customer relationship, the marketplace can change the rules or the fee structure whenever it likes, your listing sits beside competitors including sometimes the marketplace's own version of your product, and if the account is ever suspended the revenue stops the same day.
A business where one platform can end your revenue with an automated email is not a business you fully own.
The fix is to treat marketplaces as acquisition and own what happens next. Put something in the box that gives a reason to buy direct next time. Collect the email where the rules allow it. Make sure your own store is good enough that a returning customer prefers it. The goal is a business where the marketplace is one channel rather than the only one, and that's a store architecture question more than a marketing one.
Leak three: nobody chases the second purchase
Almost every small store I've looked at spends nearly all its money and attention getting a stranger to buy once, and almost nothing on getting that same person to buy again.
That's backwards for most product businesses. The person who already bought from you has done the hard part. They know the product arrives, they know the quality, and they don't need convincing that you're real. Reaching them costs a fraction of what it cost to find them.
A few things that consistently move this, none of which are complicated:
- A good post-purchase sequence. Not just tracking numbers. What to expect, how to use it, what people usually buy next.
- Reordering that takes one tap. If you sell anything consumable, the reorder should be trivially easy and timed to roughly when they'd run out.
- Knowing who bought what. Sounds obvious. Most small stores can't segment their list beyond "everyone", so every email is written for a stranger.
- Asking for the review at the right moment. After they've had it long enough to like it, not the hour it shipped.
How to tell which leak is yours
You don't need a full audit to find out. Three numbers will point you at the right one.
Look at what share of your traffic is mobile against what share of your revenue is mobile. If the gap is wide, and it usually is, leak one is costing you the most and it's also the fastest to fix.
Look at where your orders come from. If one marketplace is most of your revenue, leak two is your biggest risk even if it isn't your biggest cost this month.
Look at how many customers have ordered more than once. If it's a small share and your product is something people could reasonably buy again, leak three is the cheapest growth available to you and you're leaving it on the table.
Why this beats buying more traffic
More traffic is the most expensive way to grow, and it's the only one most stores try. Every one of these three fixes makes the traffic you already have worth more, which means every ad you run afterwards performs better too.
It's the same argument as site speed. Fixing conversion improves everything upstream of it at once, and it compounds, whereas traffic stops the moment you stop paying for it.
If you want a second pair of eyes on which leak is yours, that's most of what conversion work is: finding the specific place the money is going, rather than redesigning things that were already fine.
