Conversion rate optimization (CRO) is the practice of increasing the share of website visitors who complete a defined action, such as requesting a demo, starting a signup or submitting a form, by finding what stops them and testing changes that remove it. It is a research and experimentation discipline, not a list of design tweaks: the output is evidence about what your buyers respond to.
The metric at the center is the conversion rate. Nielsen Norman Group defines it simply as "the percentage of users who take a desired action" (Nielsen, 2013).
The conversion rate formula
Conversion rate = conversions / visitors x 100
Two choices decide whether the number means anything. First, the denominator: sessions, users or identified accounts give different rates, so pick one and keep it. Second, the conversion: a newsletter signup and a sales-qualified demo request are not the same event. In B2B, track the primary conversion (the one that creates pipeline) separately from micro-conversions like content downloads. Our conversion rate calculator does the arithmetic and shows the uplift between two periods.
How conversion rate optimization works
- Measure. Confirm the conversion is tracked correctly and segment the rate by page, channel, device and audience. Broken tracking is the most common "conversion problem".
- Research. Find where and why people drop: funnel reports, form analytics, session recordings, on-site polls and sales call notes. Quantitative data shows where; qualitative data shows why.
- Hypothesize. Write each idea as a sentence that names the evidence you saw, the change you will make, the outcome you expect and the audience it applies to. For example: "Because we saw mobile visitors miss the SSO row, we believe a plan comparison toggle will cause more plan selections for mobile pricing-page visitors." No evidence, no test.
- Prioritize. Score ideas on expected impact, confidence and effort, and against how much traffic the page actually gets.
- Test. Run an A/B test with a sample size fixed in advance. Pages with too little traffic get changes shipped and monitored instead.
- Learn. Record the result, including losers, so the next hypothesis starts from evidence.
Example
Example (illustrative numbers): a B2B software site gets 8,000 visitors a month to its demo page and 160 demo requests, a 2.0% conversion rate. Form analytics show most abandonment happens at a "budget" field. The team tests a version without it.
- Control: 2.0% (80 requests from 4,000 visitors)
- Variant: 2.5% (100 requests from 4,000 visitors), a 25% relative lift
Before calling it, they check two things: whether the sample was large enough to detect a lift that size (here it is not; one month of traffic is under-powered, so the test keeps running), and whether the extra requests convert to qualified meetings at the same rate. A lift that floods sales with unqualified meetings is not a win.
How B2B teams use CRO
- Demo and contact forms: removing fields, clarifying what happens after submit, adding a calendar step. See our lead capture form breakdown.
- Pricing pages: clarifying which plan fits which buyer and what the sales process looks like.
- Landing pages for paid traffic: matching the headline to the ad and keyword that brought the visitor.
- Segment-specific experiences: showing industry or company-size relevant proof to identified accounts, then testing it against the generic page.
- Audits: a structured pass over the funnel, using the CRO audit checklist, before choosing what to test.
What most teams get wrong
Our view: the usual failure is not bad ideas, it is optimizing a number nobody trusts. Teams test button colors on pages with 300 visits a month, stop tests the day a variant looks ahead, and report form fills without checking pipeline. Fix measurement, test only where traffic supports it, and judge winners on qualified pipeline.
| CRO | A/B testing | Personalization | |
|---|---|---|---|
| What it is | The overall discipline | One method for validating changes | One type of change: different content per segment |
| Output | Higher conversion rate, documented learnings | A decision on one hypothesis | A tailored experience |
CRO is also a career path in its own right; our guide to conversion rate optimisation jobs covers the roles, skills and published salary data.
Frequently Asked Questions
What is a good conversion rate?
There is no universal benchmark worth chasing. Rates depend on the conversion you count, the traffic source and the price of what you sell; a demo request for a six-figure product and a free signup are not comparable. The useful comparison is your own page against itself over time, split by channel and segment, and the conversion rate of each step into pipeline.
How long does a CRO test take?
As long as it takes to reach the sample size you calculated before launch, and at least one or two full weekly cycles so weekday and weekend behavior are both included. Low-traffic B2B pages often need several weeks per test. If the required sample would take more than about two months, test a bigger change or a higher-traffic page instead.
Is CRO the same as A/B testing?
No. A/B testing is one tool inside CRO. Conversion rate optimization also includes measurement, user research, prioritization and deciding when not to test at all. Many valuable CRO changes, like fixing broken tracking or a confusing form error, are shipped directly because there is nothing to debate.
Related
For the full process, read the conversion rate optimisation guide. To plan tests, use the A/B test plan template and check required traffic with the A/B test calculator. For tooling, see our comparison of conversion rate optimization tools.