What is conversion optimisation?

What does conversion optimisation mean?

A conversion is a predefined action. Conversion rate optimisation, or CRO, investigates why people complete that action or leave along the way. It informs changes to content, usability and decision journeys. The goal is a clear, working route from interest to a relevant action.

CRO is not a collection of universal button colours or layout rules. A change needs to fit the offer, information needs and actual use. The same change may have different effects on two websites.

Which conversion matters?

For a B2B website, a confirmed enquiry may be an initial goal. Sales often determine suitability later. A shop considers confirmed orders and their commercial value. A contact click or checkout start indicates interest, but does not establish a business outcome.

Define the primary goal, intermediate actions and quality criteria separately. This helps avoid optimising for many inexpensive contacts your sales team cannot use. Document the definition in the measurement plan and agree it with the responsible teams.

Two examples from the decision journey.

Illustrative B2B example: a form requires a phone number although the first exchange can happen by email. The hypothesis is that making the number optional reduces friction. First establish whether sales actually need it immediately. After a change, examine submission quality as well as quantity.

Illustrative shop example: delivery terms only appear at checkout. A possible change presents relevant information earlier on the product page. Whether this creates more commercially useful orders needs evaluation. Both examples are hypotheses, not claimed client results.

How does CRO work in practice?

Start with the goal, audience and verified measurement. Then examine specific journeys: what does the ad promise, which questions does the page answer, and where does input become difficult? Data reveals patterns; observation, conversations and technical checks help explain them.

For each change, document the observation, hypothesis, expected effect and evaluation method. Prioritise by importance, evidence and effort. Implementation is followed by functional checks and evaluation. Record what was not supported too, creating a traceable learning process rather than a shifting list of preferences.

How is conversion rate calculated?

First define the denominator: sessions, users and a specific group of eligible visits are different bases. Calculation example: 30 sessions containing a confirmed enquiry out of 1,000 considered sessions yield a session-based rate of 3%. This is not an industry benchmark.

Check duplicate events, test orders, failures and consent-related gaps. In Analytics, an event must represent the intended action. Google’s documentation explains the distinction between general pageviews and specifically defined key events. Traffic mix, seasonality and offer changes also inform interpretation.

What are the limits of CRO?

A better interface cannot replace a relevant offer, reliable fulfilment or sales follow-up. Before-and-after comparisons alone do not establish causation. Controlled experiments need a suitable question and sufficient data; with low volume, qualitative reviews are often the more useful next step.

Usability remains a quality requirement in its own right. W3C recommends clear form labels and understandable error feedback, among other measures. Such foundations should not need a revenue experiment to justify them.

Sources and further documentation

Google Analytics: Create or modify key events

W3C WAI: Forms Tutorial

Explore the topic

Conversion optimisation

Tracking & analytics