I run a digital agency in a small Polish town. I build websites, do SEO, run campaigns. And I regularly look at online stores whose owners tell me: "it's not selling." I open the site. And I see why.
It's not about magic. It's not about some secret TikTok hack. It's about basics that 90% of stores get wrong.
Product pages that don't sell
I had a client. Garden furniture store. Nice products, fair prices. Conversion rate? 0.4%. I open a product page. One photo. Description copy-pasted from the wholesaler. Zero reviews. The "Add to cart" button buried somewhere at the bottom.
We did three things:
- Professional photos — 5-6 per product, close-ups, wide shots, lifestyle shots
- Copy written for the customer, not for Google. Benefits, not technical specs
- Reviews with real names and cities — people trust people, not star ratings
After one month, conversion jumped to 1.8%. On their revenue, that was an extra $3,000 per month. For a one-time optimization cost.
Checkout that scares people away
Another story. A natural cosmetics store. Instagram traffic — beautiful, targeted, expensive. People add to cart. And disappear.
I check the checkout. Five steps. Mandatory registration. A form that looks like a mortgage application. Practically asked for a DNA sample.
We cut it to two steps. Added guest checkout. Threw in express payment options. Result? Cart abandonment dropped from 78% to 51%. That's the difference between "not worth it" and "profitable."
The rule is simple: every extra step in checkout costs you roughly 8-12% of completions. Do the math on your own revenue.
Trust — the thing nobody talks about
You know what's funny? Stores spend thousands on ads and then don't have on their site:
- A phone number
- A physical address
- A return policy that's actually readable
- Shipping info visible WITHOUT searching for it
I ran a test with one client. We added a bar saying "Free shipping over $40 | 30-day returns | Over 2,000 happy customers" at the top of the page. Conversion went up by 0.3 percentage points. From nothing. From one line of text.
Because people don't trust stores that look like they were set up yesterday. And they're right not to.
Remarketing — the second chance
A stat I like to repeat: 97% of visitors to your store won't buy on the first visit. Ninety-seven percent.
So you either give up on that 97%, or you come back to them. Remarketing on Facebook and Google isn't optional — it's essential. But you do it smart. And to make it work well, you need a solid CRM that shows you who's coming back and when.
You don't show people the same banner 47 times. You show them the specific product they looked at. With a customer review. Or with a discount code that expires in 48 hours.
For one client, dynamic remarketing accounts for 23% of all sales. At a campaign cost of $300/month, it generates $4,500 in revenue. I don't think that needs commentary. But to get campaigns performing this well, you need to avoid the common Google Ads mistakes I see beginners make.
Data, not hunches
Last thing. Most important thing. Stop guessing.
Set up Analytics. Add Hotjar or Microsoft Clarity. Watch where people click, where they pause, where they leave. Run A/B tests on buttons, headlines, images. If you don't know where to start with analytics, I wrote a separate guide on web analytics for beginners.
I had a client convinced the problem was pricing. "We're too expensive." We fired up session recordings. People weren't even reaching the price. They got stuck on the homepage because the navigation was so convoluted they couldn't find the product.
We fixed the navigation. Sales went up 40%. Price didn't change.
Don't ask customers why they don't buy. Watch what they do on your site. The camera doesn't lie.
Online sales isn't a lottery. It's a system. Every element either helps or gets in the way. Your job is to find those obstacles and remove them. One by one. No revolution, no fireworks. Just solid work.