Increasing website sales starts with one question: why don't the people who already reached the site get in touch or buy? Most businesses pay for more traffic, when the very same visitors could bring twice as many leads. I find the reason in the data, fix it in code, and measure that sales really went up.
What conversion rate optimization is, and what it isn't
Revenue from a website is the product of three numbers: how many visitors arrive, what share of them get in touch or buy (the conversion rate), and what each lead is worth. Advertising and SEO work on the first number. Conversion rate optimization (CRO) works on the second.
A simple example: a site with 2,000 visitors a month and a 1% conversion rate brings 20 leads. The same 2,000 visitors at a 2% conversion rate bring 40 leads, on exactly the same ad budget. To get to 40 leads through advertising alone, you'd have to double the budget, every month, forever.
What it isn't: it isn't "a prettier design" and it isn't red-button tricks. It's systematic work: measure where visitors stop, form a hypothesis about why, change one thing, and check whether the number moved. What isn't measured doesn't count.
More traffic vs more conversions
| Compared | More traffic (advertising) | More conversions (CRO) |
|---|---|---|
| Cost | Every extra lead costs more clicks, every month | One-off work, the result stays |
| What happens to the ad budget | Same conversion rate, same cost per lead | Every ad dollar brings more leads |
| Time to results | Immediate, as long as you pay | Quick fixes within weeks, tests within a month or two |
| Risk | Budget wasted on a site that does not convert | Every change is measured; what did not work is rolled back |
| When it is the right answer | A site that converts well and wants to grow | Existing traffic, few leads |
The two work together: a site that converts well makes every campaign more profitable, which is why the site is usually fixed before the budget grows. If there's no traffic at all yet, start with organic SEO or a landing page for a campaign, and optimize after that.
Why visitors don't buy: the 7 leak points I check
Almost every site loses customers in the same places. These are the points I check first, in the order of impact I see in practice:
- Speed on mobile. A 0.1-second faster mobile load lifted retail conversions by 8.4% (Google/Deloitte, 37 brands). A site that takes 6 seconds to load loses some visitors before they've seen anything.
- The first screen. Within 5 seconds the visitor needs to understand what you offer, to whom, and what to do now. A slider, a generic slogan and three different buttons are the surest way to lose them.
- The form. Every unneeded field costs leads. A name and a phone number are enough to start a conversation; the rest can be asked on the phone. And a form that doesn't work on a certain phone is a leak nobody sees without measurement.
- Trust. Real reviews, prices, a photo of the person who answers, a clear cancellation policy. According to Baymard, 19% of those who abandon don't trust the site with their card details.
- Checkout. Costs that pop up at the end (40% of abandonments), a required account (18%), a long process (17%).
- Mobile. Most traffic comes from phones, and that's where the site is tested least: small buttons, a form hidden behind the keyboard, a pop-up that can't be closed.
- Measurement. If phone calls, chat messages and forms aren't counted, there's no way to know what works. Half the sites I check measure only part of their leads, and make decisions on half a picture.
Want to know where your site loses customers? Send me your site address and the action you want visitors to take.
A free 30-minute call, no obligation · reply within 24 hours
Book a free callHow to increase website sales: what I actually do
These are the actions themselves. Each one is documented, with the number before and the number after.
- Full measurement. Google Analytics 4 with goals for every kind of lead: form, phone, chat, order. Without it there's no baseline to compare against, so it's always the first step.
- Data analysis. Where visitors come from, on which pages they leave, where in checkout they stop, and how phone differs from desktop. Recordings and heatmaps show what the numbers don't tell.
- Hypotheses by impact. A list of what probably blocks leads, ranked by how many visitors it affects and how easy it is to fix. Start with what affects everyone: speed, the first screen, the form.
- Fixes in code. Speed, forms, page structure, call-to-action buttons, checkout. I fix them myself, the same week, without handing "recommendations" to another developer.
- A/B tests. When there's enough traffic, two versions of a page or a form run side by side, and the numbers decide. Without enough traffic, compare before and after on the same channel.
- A one-page monthly report. The conversion rate, the leads, what changed, what worked and what didn't, and the plan for next month.
How it works: 6 steps from the first call to the report
Send the site address and the action you want. I check speed, mobile, the first screen and the form, and tell you on the call whether there's something to improve, and roughly how much.
Connecting or fixing the measurement, then two weeks of data and recordings. At the end: a one- or two-page report with the leaks by impact, and what each fix is expected to bring.
What affects every visitor: speed, the first screen, the form, trust. In code, within a week or two, measured before and after.
Version against version on the key pages: the home page, the service or product page, checkout. Every test with a written hypothesis and a target number.
What won goes live for everyone; what didn't is dropped. No ego and no "but it looks nicer".
A monthly report with the numbers, and the list of the next tests. The work continues as long as every month brings in more than it costs. You can stop in any month.
Why a developer, not a consultant with a slide deck
Most CRO services end with a report: thirty recommendations, and a client's developer who has to implement them sometime. The best recommendations, never implemented, are worth zero. The person who diagnoses the problem is the one who fixes it: the form loses fields, the page loads faster and checkout drops steps the same week, in code, and it's all measured.
Two examples already published on this site: a landing page that replaced Elementor with clean code went from 6.2 seconds of loading time to 1.4, and its conversion rate rose 38% within a month without changing a word of the content. The Hebrew home page of this site went from 64 to 100 in mobile PageSpeed, with a two-field form on the first screen. That isn't magic, it's code.
Three things aren't up for discussion: measurement comes before any change, because without a number before there's nothing to compare after. Change one thing at a time, because changing ten things teaches you nothing. And the data, the accounts and the code are yours.
My experience
I'm Vadim Leviev, a developer with 25+ years of experience. I was Tech Lead at William Hill Online, on systems serving millions of users, where every change to a page was measured in money the same day, and later a developer at Mars Media Group and BuyMe. That's where the habit comes from: checking every change in numbers, not by the feeling of whoever designed it.
I've worked independently since 2006, full time since 2019, directly with businesses: building and improving WordPress sites and WooCommerce stores, PHP and Laravel systems and AI systems, and developing verifystring.com. My full background is on LinkedIn, and there's more about me here.
Common CRO mistakes, and how I avoid them
- Raising the ad budget for a site that doesn't convert
It doubles the spend, not the leads. Fix the site first, then grow.
- Changing ten things at once
If leads went up, you don't know why; if they went down, you don't know what caused it. One thing at a time.
- Drawing conclusions after two days
Twenty visitors tell you nothing. A test runs until there's enough data, not until the result looks good.
- Testing a button color while the form is broken on mobile
The big leaks first: speed, mobile, the form, checkout. Fine-tuning tests later.
- Measuring only forms
Phone calls and chat messages are sometimes most of the leads. Without counting them, decisions are made on half a picture.
- Copying what works for Amazon
A site with millions of visitors has different problems. Your site needs fixes based on your own data.
When not to start: an honest comparison
Not every site needs this now. Four situations where I say start somewhere else, and one where CRO pays off most:
| Situation | Conversion optimization now | Better first |
|---|---|---|
| Fewer than a few hundred visitors a month | Not enough data for tests | Basic fixes based on an audit, then traffic: SEO or advertising |
| The site is old, slow or broken | Optimizing on a shaky foundation | First upgrade the site, then test |
| The traffic is the wrong traffic | No site sells to the wrong audience | Fix the campaigns and the keywords, then the site |
| The product or the price is not competitive | The site will not fix the offer | Check the offer against competitors before anything else |
| Existing traffic, few leads | Exactly where it pays off most | Keep paying for more clicks |
CRO cost: price ranges for 2026
The numbers are estimates; the exact price is set after the free call, in a fixed quote that doesn't move halfway through. For comparison: doubling a $1,500 monthly ad budget costs $18,000 more a year, and doesn't change the conversion rate at all.
| What | What is included | Estimated range |
|---|---|---|
| Measurement setup | GA4, goals for form, phone and chat, connection to Google and Meta, recordings and heatmaps | $220-420, one-off |
| Full conversion audit | Two weeks of data, speed, mobile, forms, checkout, a written report with priorities | $420-970, one-off |
| Quick fixes | Speed, the first screen, form, trust, checkout, in code, measured before and after | $690-2,200 |
| Sales page or landing page rebuilt | Structure, copy, design and clean code, connected to measurement | $830-2,200 |
| Monthly CRO | 2-3 A/B tests or changes a month, monitoring, a monthly report | $690-1,700/month |
No annual commitment: the retainer is monthly and you can stop in any month. The measurement, the data and the code stay with you.
International-level service and knowledge. Vadim built my online store step by step, met all my needs, and whenever something new came up, thought it through and found a solution. All along the way I felt safe and in good hands - even under pressure there was always a calm answer. On top of that, the prices are fair, and the service and support are among the best I have come across in Israel.
Conversion rate optimization: frequently asked questions
What is a conversion rate, and what counts as a good one?
The conversion rate is the share of visitors who do the action you wanted: a lead, an order or a sign-up. In e-commerce the global average is 2-3%, and on a services site a form that converts 3-5% of visitors is a good result. The number that really matters isn't the average but yours, before and after, on the same channel and the same budget.
How long until I see results?
Quick fixes, such as speed, a shorter form and a clear message at the top of the page, have an effect from the week they go live. An A/B test usually needs 2-4 weeks and a few hundred conversions per version for the result to be reliable. A site with little traffic is fixed according to the data and the recordings, without A/B tests.
I have little traffic. Is CRO relevant for me at all?
Yes, even more so: when every visitor costs money, every point of conversion is worth more. You just don't run A/B tests, which need a lot of traffic. Instead you fix what the data, the recordings and user tests show, and measure before and after.
How is this different from redesigning the website?
A redesign replaces everything at once and hopes the new version works better. Conversion optimization changes one thing at a time, measures, and keeps only what raised the number of leads. For most sites this is cheaper, faster and safer, because you don't lose what already works. A truly old or broken site gets an upgrade first, and then optimization.
Does it include ad management?
No. Managing the campaigns themselves is a separate service, and when needed I work together with the person who runs them. What is included is the measurement: every lead, from the form, the phone or chat, is counted in Google and Meta, so the campaigns learn from what works and not from what looks good.
How much does conversion rate optimization cost?
A full conversion audit costs $420-970, quick fixes $690-2,200, and monthly CRO with A/B tests $690-1,700/month. The first call costs nothing, and the exact price is fixed in a written quote before the work starts.
Same traffic, more customers. Send your site address and I'll tell you where it loses them.
A free 30-minute call, no obligation · reply within 24 hours
Book a free call