Conversion rate optimization (CRO)
Conversion rate optimization is the practice of increasing the share of website visitors who take the action you want, usually by finding where people abandon and removing the reason. It is the discipline; [website conversion rate](/glossary/website-conversion-rate/) is the number it moves.
In practice
In B2B the action is rarely a purchase. It is an enquiry, a quote request, a specification download or a phone call, and the volumes are small enough that a single extra enquiry a week is a meaningful result. That changes the method considerably from the ecommerce version of this discipline that most published advice describes.
The first thing worth understanding is that at typical B2B volumes you cannot test your way to an answer. An A/B test needs enough conversions in each variation to separate a real difference from chance, and a site producing twenty or thirty enquiries a month will take many months to get there, by which point the season and the market have both moved. Below that volume, watching what people actually do is faster and more reliable than experimenting on them.
That makes the qualitative tools the primary ones here. Session replay shows you where someone stopped. A heatmap shows what they never scrolled to. Form analytics names the field they abandoned on, which matters because the average first-contact form on the B2B sites in our audit asks for 12 fields.
The second thing is that most B2B conversion problems are defects rather than hypotheses. In our audit of more than 55,000 US B2B homepages, 57% gave a visitor no clear next step above the fold and 71% showed no proof a first-time buyer could verify. Those do not need testing, they need fixing, and pointing an experimentation platform at a page with no clear call to action will spend three months proving neither variation works.
A test you can run this week without any tooling: open your own site on a phone, arriving from a search result rather than a bookmark, and try to make contact the way someone who has never dealt with you would. Count the taps. If it takes more than three and you have to scroll to find a phone number, you have found more than most testing programmes turn up in a quarter.
Common questions
What is a good conversion rate for a B2B website?
Benchmarks in this category are close to useless because the definition of a conversion varies so much between companies. A more useful comparison is against your own site over time, and against the specific step where people drop out. If someone tells you a single industry benchmark figure, ask them what they counted as a conversion.
Is CRO the same as A/B testing?
No. A/B testing is one method within conversion rate optimization, and at B2B traffic volumes it is often the wrong one because tests take too long to reach significance. Session replay, heatmaps, form analytics and simply fixing obvious defects are all conversion rate optimization and none of them require statistical power.
Where should a B2B company start with CRO?
With the enquiry path, because that is where intent is highest and losses are most expensive. Check that there is one obvious next step above the fold, that the phone number does not require scrolling, and that the form asks for the minimum needed to reply. Those three account for most recoverable loss on a typical B2B site.