8 E-commerce Numbers That Tell You Less Than You Think
The metric rises. You celebrate. But something's still broken.
Here's the thing about data: the numbers everyone tracks aren't always the numbers that matter.
Some metrics feel safe. They show up in every dashboard. Everyone nods when you mention them. But dig one layer deeper and they fall apart.
This post covers 8 metrics that deceive more often than they inform — and what to check instead.
These aren't useless numbers. But they're dangerous precisely because they look so reliable.
#1 Session Duration
What it measures: How long people stay on your site during a single visit.
What you think it means: People are interested. They're reading. They're engaged.
What it might actually mean: They can't find what they need. Your navigation is confusing. They're clicking around aimlessly.
What to do instead: Look at scroll behavior. Check which pages they visit. See where they leave. Long sessions with no clicks? That's frustration, not interest.
#2 Bounce Rate
What it measures: Percentage of visits where someone views one page and leaves.
What you think it means: Your site isn't interesting enough.
What it might actually mean: On a landing page or blog post, one-page visits are completely normal. They found what they needed and left. That's not failure.
What to do instead: Check if they scrolled. Did they click your CTA? Did they spend time reading? Context matters more than the number.
#3 Churn
What it measures: How many customers you lose in a period.
What you think it means: Fewer people leaving = things are improving.
What it might actually mean: You lost a bunch of low-spenders. Your best customers are still leaving at the same rate. The average just looks better.
What to do instead: Split churn by customer value. Check which cohorts are leaving. Track retention by segment, not just overall.
#4 Average Order Value
What it measures: How much people spend per transaction.
What you think it means: Customers are buying more expensive items.
What it might actually mean: Your cheap products stopped selling. Or you're simply getting fewer orders. The average goes up, but revenue might be flat—or down.
What to do instead: Look at units per order. Check product mix. Compare total revenue, not just averages.
#5 Conversion Rate
What it measures: Percentage of visitors who complete a purchase.
What you think it means: Your site is converting better.
What it might actually mean: Your traffic changed. Bad visitors stopped coming. The percentage improved, but you might be selling less overall.
What to do instead: Compare conversion across traffic sources. Check total conversions, not just the rate. A 10% conversion on 100 visitors is worse than 2% on 10,000.
#6 Cart Abandonment
What it measures: How often people add items to cart but don't buy.
What you think it means: Fewer people are bailing at checkout. Progress.
What it might actually mean: Fewer people are reaching the cart in the first place. Maybe your product pages are broken. Maybe traffic dropped. The rate improved because the funnel narrowed.
What to do instead: Track how many people reach the cart. Compare that to checkout starts. Watch absolute numbers, not just percentages.
#7 Site Traffic
What it measures: Number of visits to your site.
What you think it means: More visits = more opportunities.
What it might actually mean: Most of that traffic is junk. Bots. Wrong audience. Accidental clicks from irrelevant ads.
What to do instead: Split traffic by source and device. Check engagement per segment. A thousand useless visits are worse than a hundred good ones.
#8 Cost Per Click
What it measures: What you pay for each ad click.
What you think it means: Cheaper clicks = better performance.
What it might actually mean: You're buying low-quality traffic. People click, then immediately leave. You're paying less but getting nothing.
What to do instead: Check what happens after the click. Bounce rate. Time on site. Purchases. Cheap clicks that don't convert are expensive.
How to Actually Read Metrics
A number by itself means nothing.
It's not about whether a metric goes up or down. It's about why—and what else changed around it.
Here's how to stop getting fooled:
See the whole picture Conversion up? Great. But why? More traffic? Better product pages? Different audience? One metric never tells the full story.
Break everything down New vs returning. Mobile vs desktop. Paid vs organic. The overall number hides what's really happening.
Watch volume, not just rates A 50% conversion rate sounds amazing. Until you realize it's based on 10 visitors. Meanwhile, your main traffic source converts at 2%—and that's where all your revenue comes from.
Add behavior to the numbers Did they scroll? Click? Read? Spend time? Numbers show outcomes. Behavior shows intent.
Final Thought
Metrics don't lie. But they don't explain themselves either.
A number can look good and still mean nothing. Or look bad and hide something important.
Next time a metric moves, don't just react.
Ask: what changed? Who's behind this number? What am I not seeing?
That's where the real answer is.