Your Funnel Halves at Every Step or You've Got a Blocker

There's a rule I use all the time when I analyze an account. Simple, and it spots a problem in 30 seconds. Basically, at every step of your funnel, your rate divides by 2. You've got 10% add-to-cart, you drop to 5% reaching checkout, and you end at 2.5% conversion. It's not a law of physics, but it's what we see with most of the brands we work with.

The number itself, you don't care. What matters is the ratio. The moment a step breaks that ratio, you've got an abnormal drop-off somewhere. And that's where you need to dig.

The divide-by-2 rule, and why it works

A classic conversion funnel is three big measurable steps on a site. Add-to-cart, reaching the checkout page, then completed payment. Each time, you lose roughly half your people.

So if I take a site with 10% add-to-cart, I expect to see roughly 5% on the checkout page and 2.5% final conversion. When that's the case, all good, the journey is broadly healthy.

My method is simple. I pull the three rates, put them side by side, and check if the ratio holds. If you lose way more than half between two steps, there's a blocker. If you lose less, great, but it often hides something else, like botched tracking.

Don't look at the absolute value of each rate. Look at the slope between steps, not the raw value. A site that drops cleanly by half each time, even with average rates, is healthier than a site that collapses at one specific step.

A real case that landed on my desk

I had this with a multi-brand apparel retailer. I pull the numbers, and I see add-to-cart at 2.6%. Already, that's low. And a final conversion rate that was actually higher than my rule predicted.

If I apply my rule, 2.6 divided by 2 is 1.3%, then 0.65% final conversion. Except their conversion rate was above that expected 0.65%. So the problem was clear: between add-to-cart and checkout, then at the payment step itself, there was a huge and completely abnormal drop-off. The numbers didn't add up, and that's exactly what should set off your alarm.

Digging in, I find the culprit. To reach the checkout page, the site forced the customer to create an account with a password. You add to cart, you're ready to buy, and now you're asked to invent a password and create an account. The 2 out of 3 abandoned payments came largely from that.

This forced account creation, I kill it almost every single time I come across it. It's a massive conversion blocker, and 9 times out of 10 it serves absolutely no purpose.

The mandatory account vs. mandatory phone confusion

There's something brands often mix up, and I want to clear it up because it matters.

Making the phone number mandatory at the delivery step is justified. The carrier needs it to notify the customer, manage time slots, missed deliveries. There, I agree, you can keep it mandatory.

But forcing account creation with a password to pay is a dealbreaker. Nobody wants to sign up for your club just to pay. They want to buy and get out. And on Shopify, it's even more absurd because the platform already centralizes everything via email as the identifier. The customer gets their order confirmation by email, their shipping tracking by email, everything goes through there. They have zero need for an account to find their info.

On this one, we just disabled the mandatory account in the theme. The checkout went back to being smooth, single-page. The customer lands on one page, fills it out, pays. No more detour through signup.

This kind of friction is the same deal as tracking that distorts your data and makes you decide half-blind. A technical detail eating your sales every day without you ever noticing.

What rates to aim for, and why the source changes everything

Before you panic over your conversion rate, you need to know what you're comparing yourself to. And that all comes down to your traffic.

On organic or direct traffic, meaning people actively searching for you with buying intent, an eCommerce funnel conversion rate between 1 and 2% is normal. If you hit around 3%, you're already very high, that's rarely beaten. The only brands that go higher are high-recurrence ones. Think supplements for example, where people come back on their own to reorder their cure, so the rate goes through the roof.

Now, if you run a lot of Meta Ads, your conversion rate will mechanically drop. That's normal. Meta traffic is less intent-driven. You're tapping the shoulder of someone who was casually scrolling, they click out of curiosity, they're not necessarily there to buy right now. So you don't compare the conversion rate of paid traffic with that of traffic actively searching for you.

That's why looking at your overall conversion rate, all sources combined, doesn't mean much. You have to segment. A visitor coming from Google by typing your name and a visitor coming from a Meta ad are two different businesses with two different values. You can't throw them in the same bucket.

Where to start when your funnel is leaking

Once you've spotted the leaking step, you work on it. The levers are well known.

If the drop-off is at the top of the funnel, on add-to-cart, you look at your pages. The PLP, your collection pages, and especially the PDP, your product page. That's where the desire to buy plays out, above all the photos and the clarity of the info, and trust signals in the right spot. And you can run A/B tests to settle it instead of guessing.

If the drop-off is at the bottom, between cart and checkout, you look at the buying journey itself. That's exactly my case from earlier. A mandatory account, too many steps, useless fields, no trust signals at the right moment. Because here's something important, trust signals at the payment step aren't the same as at the add-to-cart step. At payment, the customer wants reassurance on security, delivery times, the return option.

And if you've got, like in my example, 2 out of 3 payments going up in smoke, alongside fixing the journey you absolutely need to work on your abandoned checkout flows. It's the same principle as your other CRM flows that can quietly break without you knowing why. You've got people who are red-hot, they half-entered their card, you follow up by email and recover a chunk of them.

Concretely, here's how I work on every audit. I pull the three funnel rates, add-to-cart, reaching checkout, conversion. I check the divide-by-2 ratio between each step. The moment a step breaks the rule, I know where the problem is and I dig there first. I segment by traffic source, because an overall conversion rate lies. And I always look at the real journey, not just the numbers, sometimes all it takes is disabling one field to unlock a big part of the revenue.

No need for a full rebuild in 80% of cases. Often, a few tweaks are enough. A mandatory account to kill, or a field to make optional. Sometimes just reactivating the abandoned checkout flow. Ten-minute tweaks that win back a share of the revenue you were letting slip away.