You're Getting Traffic but No Sales: How to Find Where Visitors Drop Off
Traffic looks healthy but nobody converts. The problem usually isn't your traffic — it's a specific step where visitors quit. Here's how to find that step and fix it.
You checked analytics this morning. Traffic's up. Sessions look great. You should feel good.
Then you check sales, and it's the same flat line it's been for weeks.
This is one of the most frustrating gaps in running a website, and also one of the most misunderstood. When traffic is fine but conversions aren't, the instinct is to go get more traffic. That's usually the wrong move. You don't have a traffic problem. You have a leak — a specific step where people show up, look around, and quit.
Find the leak and you fix the whole thing. Here's how.
Stop looking at the total. Look at the steps.
"Conversion rate" is a single number, and single numbers hide the truth. A 1.5% conversion rate could mean everyone's mildly unconvinced, or it could mean 98% of people bail at one exact moment. Those are completely different problems with completely different fixes.
So instead of one number, draw the path. For most sites it looks something like this:
- Landed on the site
- Looked at a product, service, or key page
- Started the action — began a signup, added to cart, opened the form
- Finished — bought, booked, or submitted
At every arrow, some people leave. That's normal. Your job is to find the arrow where too many leave.
Find the step with the biggest cliff
Go through your funnel step by step and write down how many people reach each one. You're looking for the single biggest percentage drop between two steps.
Say 10,000 people land, 4,000 look at a product, 800 start checkout, and 120 finish. The drops are:
- Landed → Looked: 60% leave (10,000 → 4,000)
- Looked → Started: 80% leave (4,000 → 800)
- Started → Finished: 85% leave (800 → 120)
That last step — people who started checkout and didn't finish — is your cliff. These are your warmest visitors. They wanted it enough to begin. Something in those final steps stopped them. That's where a fix pays off the most, because you're not convincing strangers, you're rescuing people who already said yes.
The usual suspects at each drop
Once you know where, the why is usually one of a short list.
People land and immediately leave:
- The page doesn't match what they searched for
- It loads slowly on a phone
- The headline doesn't make it obvious what you do
People look but never start:
- No clear next step — the button is buried or vague
- Price surprise, or price hidden until too late
- Not enough trust — no reviews, no proof, no faces
People start but don't finish:
- The form asks for too much, too soon
- Forced account creation
- Unexpected shipping cost or a confusing final screen
- It breaks on mobile
Notice how few of these are "SEO" problems. They're clarity and friction problems. The good news: they're almost all fixable in an afternoon once you know which one you have.
Watch a few real sessions
Numbers tell you where. Watching tells you why. Most analytics setups won't record screens, but you don't need fancy tooling to start — sit down on your own phone, come to your site the way a customer would, and try to buy or sign up. Do it slowly. Notice every moment you hesitate, squint, or think "wait, where do I…". Those hesitations are your drop-offs, live.
Better yet, hand your phone to someone who's never seen the site and say "buy this, think out loud." You'll learn more in five minutes than in a week of dashboards.
Fix one step, then measure again
Resist the urge to redesign everything. Pick the single biggest cliff, form one honest guess about why, and change one thing:
- Cut a form from seven fields to three
- Put the price on the page instead of behind a click
- Add three real reviews near the button
- Make the call-to-action say what happens next ("Start free scan") instead of "Submit"
Then leave it alone for a week or two and check the same funnel again. Did that step's drop shrink? Keep it. Did nothing change? Revert and try the next guess. One change at a time is slower but it's the only way you'll actually know what worked.
The honest part
Not every drop is fixable, and not every visitor was ever going to buy. Some people are just browsing, comparing, or clicked by accident. Chasing a 0% drop-off is a fool's errand. The goal isn't perfection — it's finding the one step that's leaking far more than it should, and plugging it.
If you'd rather not build the funnel by hand, this is exactly what our [conversion journeys](/features/conversion-journeys) view does: it reads your real GA4 data, shows you where the biggest drop-offs are, and hands you specific things to fix — no spreadsheets, no guessing. You can [run a free scan](/scan) and see your funnel in a couple of minutes.
But even if you never touch a tool: draw the steps, find the cliff, fix one thing, measure. That loop, repeated a few times, is how "traffic but no sales" quietly turns into sales.