An online store owner in Bandar Lampung opens her ad dashboard on Monday morning. The click numbers look great: 2,400 people visited the website in the past week. Then her eyes drop to the number below and stop. Of those 2,400 visitors, only 9 actually bought something. She paid for 2,400 opportunities, and 99.6 percent of them walked away without leaving anything behind.
Not because the product is bad. Not because the ads targeted the wrong people. The problem sits between the click and the purchase: a page that does not convince, a process that is too long, or an offer that is unclear. All of that can be fixed, and this discipline is called conversion rate optimization, or CRO.
CRO is not about adding visitors. It is about making the visitors you already pay for actually take action. This article is a step-by-step guide to doing exactly that, with honest numbers and a way of measuring results that does not fool yourself.
What Conversion Rate Actually Means
Conversion rate is the percentage of visitors who perform the action you want. That action might be a purchase, filling out a form, signing up for a trial, or making a phone call. The formula is simple:
Conversion rate = (number of conversions ÷ number of visitors) × 100%
If 100 people visit a page and 3 buy, your conversion rate is 3 percent.
This number matters, but do not obsess over ideal figures from the internet. Average e-commerce conversion rates in Indonesia vary between 1 and 3 percent depending on product category, traffic source, and season. More useful than comparing with global standards is comparing with yourself: is this month better than last month?
There is a common misconception that a low conversion rate means the product does not sell. In reality, the product can be in high demand while a poor page sabotages the sale. That is where CRO works. It does not change your product; it changes how the product is presented.
One more note: conversion is not a single thing. There are micro-conversions (signing up for a newsletter, adding to cart, viewing the pricing page) and macro-conversions (purchasing, requesting a quote). Healthy businesses optimize both, because micro-conversions are often the bridge to macro-conversions.
Why So Many Visitors Do Not Buy
Before changing anything, it helps to understand why visitors leave. These patterns are nearly universal, and recognizing them tells you where to start.
1. Visitors Do Not Know What to Do
A cluttered page leaves visitors confused. Which button should they click? What is the next step? If the answer is not clear in three seconds, they leave. User behavior studies show visitors decide within seconds whether a page is worth trusting, and that decision is heavily influenced by the clarity of the first message.
2. The Offer Is Unclear
Many websites present features instead of benefits. "We provide integrated digital solutions" does not tell visitors what they get. Compare that with "Get automated sales reports every Monday morning without retyping anything." The second speaks about outcomes, not features.
3. Trust Has Not Been Built
New visitors who have never heard of you hesitate before handing over data or money. They look for signals: reviews, a reachable phone number, a clear address, guarantees, testimonials. If those signals are missing, doubt wins.
4. The Process Is Too Long
Every extra step between "interested" and "done" is a place where visitors drop off. A form with twenty fields drives away people who only wanted a quote. A checkout that forces account creation drives away buyers who just want to pay. The less friction, the higher the chance of conversion.
5. The Page Is Too Slow
This is the most underestimated factor. Indonesian visitors, most of whom access the internet on phones with varying connection quality, have little patience for slow pages. Google's research found that the likelihood of visitors abandoning a page rises sharply with load time, and every second of delay cuts conversion measurably. If your website takes more than three seconds to appear on a phone, you are losing sales before visitors see anything.
The good news: all of these problems can be measured and fixed. CRO is the discipline of finding them systematically, not guessing.
Start with Data, Not Feelings
The biggest mistake in CRO is changing something because "I think so." Feelings are subjective and often wrong. The replacement is data, and you probably already have plenty of it without realizing.
Google Analytics: Where to Begin
If your website does not have Google Analytics installed, install it first. The tool is free and gives you the basics: how many visitors, where they come from, which pages get the most visits, and which pages they leave from.
Start with simple questions:
- Which pages do visitors leave from most often?
- Which paths lead visitors to a purchase?
- Which pages get lots of visits but few conversions?
Pages with high traffic but low conversion are the clearest CRO targets. You are already paying for that traffic; now make it work.
Heatmaps and Session Recordings
Google Analytics tells you "how many" but not "why." For that, use heatmaps that show where visitors click, scroll, and pause. Tools like Hotjar or Microsoft Clarity (free) display heat maps and visitor session recordings.
The most common things found in session recordings: visitors scrolling past the most important section, clicking elements that are not clickable, or stopping confused in the middle of a form. You see with your own eyes, instead of guessing, where visitors get lost.
Exit Surveys
Sometimes the easiest way to learn why people do not buy is to ask. A short survey that appears just before a visitor leaves the page can capture one honest sentence: "too expensive", "still comparing options", "the form is too long". These answers are often the best CRO hypotheses you can get.
Building a Proper Hypothesis
CRO is not changing button colors and hoping. Every change should come from a hypothesis that can be tested. A good hypothesis format:
Because [reason based on data], if [change], then [expected result].
Real examples:
- "Because session recordings show visitors stopping at a 12-field form, if I reduce it to 5 fields, then the form completion rate will rise."
- "Because analytics show 70 percent of visitors come from mobile, if I fix the mobile version that is currently hard to scroll, then mobile conversion will rise."
- "Because exit surveys show visitors worry about shipping, if I display the shipping policy on the product page, then add-to-cart will increase."
A good hypothesis has two traits: it is based on evidence rather than feelings, and it can be measured. If you cannot measure the outcome, it is not an experiment, just an unfocused change.
Areas That Most Often Deliver Results
Certain parts of a website almost always hide big conversion opportunities. Prioritize from the most likely to make an impact.
1. The Landing Page
The landing page is the most-viewed front door, and often the messiest. Five things to check:
- A clear headline. In one sentence, visitors must know what you offer and who it is for.
- A clarifying subheadline. Reinforce the headline with one sentence of concrete benefit.
- A prominent call to action. One primary button, with specific action words, not a vague "submit".
- Social proof. Reviews, customer counts, or logos that build trust.
- No distractions. Elements that do not support conversion should be removed, not added.
Bad landing pages often try to say too many things at once. A page that converts focuses on one offer and one action.
2. Product and Pricing Pages
In e-commerce, the product page is where decisions are made. Check: photos that are large and clear, descriptions that explain benefits (not just specifications), a visible price, transparent stock and shipping costs, and a buy button that does not require scrolling to find.
Pricing pages have their own psychology. Showing three options with one more expensive makes the middle option feel reasonable. This is not manipulation; it is presenting information so visitors can compare easily.
3. Forms
Forms are where conversion most often dies. Every extra field lowers the chance of completion. The principle: only ask what you truly need for the next step. Name and email may be enough for the first stage; details can be collected later.
Clear labels, helpful placeholders, and error messages that explain how to fix problems also reduce frustration. Long forms can be split into several short steps, which feels lighter even when the total number of fields is the same.
4. The Mobile Version
This is the most neglected area and the one with the biggest opportunity. The majority of visitors in Indonesia come from phones, yet many websites are treated like shrunken desktop versions. Hard-to-tap buttons, tiny text, forms that force zooming — all of these drive buyers away.
If you must choose one CRO investment, improve your mobile experience. The impact is almost always the largest.
A/B Testing: Testing Without Guessing
Once you have a hypothesis, it is time to test. A/B testing (or split testing) compares two versions of a page: the old version (control) and the new version (variation). Visitors are split randomly, and results are measured to see which converts better.
There are tools that make this easy, from free to paid: Google Optimize (free), VWO, Optimizely, or simply making a change and carefully comparing against the previous period. What matters is not the tool but the discipline of testing properly.
A/B Testing Rules Not to Break
One change per test. If you change three things at once and conversion rises, you do not know which one caused it. Test one variable at a time, even if it feels slow. The real speed is learning speed, not change speed.
Run long enough. Testing for one day with 200 visitors gives you results you cannot trust. A general guideline: run until you have enough data to make a statistical decision, usually several thousand visitors or two to four weeks, depending on traffic. Results that come too early are often misleading.
Measure the right things. Conversion is not one metric. Also watch secondary metrics: time on page, bounce rate, clicks on the button. Sometimes a change improves one metric but hurts another; decisions must be based on the full picture.
Do not stop at small changes. Changing a button color rarely changes conversion dramatically. Meaningful changes — a new headline, a clearer offer, a shorter process — deliver real impact.
How to Read Results
A/B test results come as a percentage improvement and a confidence level. Do not celebrate a "20 percent lift" without checking whether the result is statistically significant. With small samples, an improvement can happen by chance.
A result is called significant when the difference is too large to be coincidence. Most tools display a confidence level; above 95 percent is generally considered safe enough to make a decision.
But do not be overly rigid. There is a difference between an experiment that gives a definitive answer and a reasonable business decision. If a change is low-risk, cheap, and logically sound, you can apply it even when results are not perfect. CRO is an art of balancing rigor with momentum.
Metrics You Should Track
Healthy CRO does not stop at one number. Several metrics complement each other:
| Metric | What it tells you | How to read it |
|---|---|---|
| Conversion rate | Percentage of visitors who act | Rising means the page is more effective |
| Bounce rate | Visitors who leave without interacting | High could mean the page is irrelevant or slow |
| Time on page | How deeply visitors engage | Rising is usually good, but not alone |
| Call to action clicks | Whether the action prompt gets attention | Shows effectiveness of message and design |
| Average order value | How big transactions are | Rising means visitors buy more |
| Cost per acquisition | How much each customer costs | Good CRO lowers it without cutting traffic |
The key relationship to understand: CRO lowers your cost per acquisition. If you pay for ads to bring visitors, and every 100 visitors now produce 4 purchases instead of 2, your cost per purchase drops by half. You get more customers with the same budget. This is why CRO is often more profitable than increasing your ad budget.
How Much Should You Invest?
A fair question: what does CRO cost? The answer depends on the approach.
- Starting from zero rupiah. Many basic tools are free: Google Analytics, Microsoft Clarity for heatmaps and recordings, Google Optimize for A/B tests. If you or your team can spare the time, CRO lessons can start with no software cost.
- Paid tools. Hotjar, VWO, or Optimizely range from hundreds of thousands to several million rupiah per month, depending on traffic and features.
- Working with professionals. CRO consultants or web development services focused on conversion typically range from Rp 5-50 million per project, depending on scope. That is cheap compared with the money lost every month to a page that does not convert. For an overview of general website costs, read our website cost guide.
A wise approach: start with free tools and fix the most obvious problems in your data. Once you find areas that deliver impact, consider a bigger investment to accelerate. The Kartech team in Bandar Lampung can help you analyze your pages, build proper hypotheses, and rebuild the parts of your website that do not convert. We start from your data, not from a package, and can be reached through the contact page.
Strategies You Should Not Forget
CRO is not only about the website itself. Several supporting strategies often deliver big results.
Page Speed
Mentioned earlier, but worth repeating: speed is conversion. Indonesian visitors accessing from phones on unstable 3G or 4G connections have no patience for slow pages. Compress images, cut unnecessary scripts, and use responsive hosting. We discuss website security and performance in more depth in our business website security guide.
Copy That Sells
Words decide outcomes. Headlines, product descriptions, and calls to action written well turn visitors into buyers. Focus on benefits, not features; speak in the customer's language, not internal jargon; and give a clear reason to act now.
Trust Signals
Trust is the currency of conversion. Customer reviews, certifications, an easy-to-reach phone number, a clear address, transparent return policies. All of these reduce the doubt that blocks purchases. Hesitant visitors will not buy; they will keep searching until they are convinced, or they go to a competitor.
CRO Mistakes That Happen All the Time
Changing Too Much at Once
The temptation to redesign the entire website because one number is unsatisfying is strong. Resist it. Big overhauls are hard to measure: if conversion changes, you do not know which part caused it. Measurable, incremental improvements are always better than a confusing leap.
Stopping at a Failed A/B Test
A failed experiment is not waste; it teaches something. "The new headline did not change conversion" is valuable information: it means the problem is not the headline. Every result, up or down, narrows the next direction. CRO is an iteration process, not a search for one magic solution.
Copying Competitors Without Data
Seeing a competitor change their design and copying them is a natural reaction, but risky. What works for them may not suit your visitors. Always test based on your own data.
Ignoring External Factors
Rises and drops in conversion can come from outside your website: seasons, holidays, price changes, or competitor campaigns. When reading experiment results, pay attention to context. Do not blame your page for a decline actually caused by a long holiday break.
A Short Walkthrough Without Overclaiming
CRO becomes clearest through real examples, though every business is different. Imagine an online store selling household goods. Analytics shows the product page gets many visits but few add-to-carts. Session recordings reveal visitors stop, confused, at a shipping section that only appears at the checkout step.
Hypothesis: if shipping costs are shown earlier, doubt drops and add-to-cart rises. This small change gets tested, and the result is measured. Maybe it rises, maybe not; either way, the business now has data to make a better next decision.
That is the essence of CRO: not finding a single solution, but building a habit of asking, testing, and learning from every decision.
Your First Step This Week
CRO feels confusing when viewed as a big project. In reality it can start with three simple steps:
- Install measurement tools. Google Analytics and Microsoft Clarity are free, and they give you data that has been missing all along.
- Find one leak. Which page gets the most visits but few conversions? Start there.
- Make one hypothesis. Based on what you saw in the data, change one thing and measure the result.
One experiment per week means 52 lessons in a year. Compared with "redesigning the whole website once a year without knowing what worked", this approach is far faster and cheaper.
If you want to move faster in the right direction, the Kartech team in Bandar Lampung can help you analyze your pages, build the right hypotheses, and create a website that truly converts. We build websites and business applications that start from your conversion problem, not from a template. Reach us through the contact page to begin with your data.
If you do not have a website yet or it needs rebuilding, our guide to choosing a web development service is a good starting point before you begin CRO experiments.
