What is Conversion Rate Optimisation (CRO) ?
Turn the traffic already arriving into revenue, without spending another dollar to acquire it.
Conversion Rate Optimisation is the disciplined practice of increasing the proportion of visitors to a website or app who complete a defined commercial action, whether a purchase, a demo request, or another step carrying measurable revenue value. The conversion rate is the number of visitors completing that action divided by total visitors, and the practice treats that ratio as a variable to be raised deliberately rather than left to chance.
The discipline formed as digital analytics made visitor behaviour measurable at scale, and matured once testing tools made it possible to show two versions of a page to different visitor segments and measure which produced more of the desired action. That capability turned page design from internal opinion into a testable hypothesis, and it remains the mechanism underneath every serious programme: a change is proposed, a portion of traffic is exposed to it under controlled conditions, and the outcome is measured against a held-out control rather than assumed.
A mature programme runs through diagnosis, prioritisation, controlled testing, and implementation as a continuous cycle rather than a single redesign project. Diagnosis identifies where and why visitors fail to convert, using analytics, session recordings, and structured review. Prioritisation ranks hypotheses by estimated revenue impact against the effort to test them. Testing validates or rejects each hypothesis under statistical discipline before it reaches the live experience, and the cycle repeats against the next-ranked hypothesis.
The practice differs between transactional commerce, where the conversion event is typically a purchase decided within a single session, and considered B2B purchases, where the tracked event is an earlier step such as a qualified lead, evaluated afterward by a committee over an extended cycle. The mechanics of testing carry across both, but the definition of a conversion, and the weight a single page carries in the outcome, differ substantially.
Why It Matters
Conversion Rate Optimisation matters because it is the rare growth lever that raises the return on every acquisition channel simultaneously, turning traffic a company has already paid to attract into more revenue, without a corresponding request for a larger budget to reach the same volume of buyers, and it is one of the few disciplines that compounds every time it runs.
Conversion Rate Optimisation is the rare growth lever that improves the return on every acquisition channel at once, because a higher conversion rate multiplies the value extracted from traffic a company has already paid to attract. Where new-channel expansion carries rising acquisition costs, a validated conversion improvement compounds against the existing media plan immediately, lifting revenue per visitor across paid, organic, and referral traffic without a matching increase in spend. This multiplying effect is why conversion work outranks incremental acquisition spend for a CFO.
A disciplined testing programme surfaces friction that no single function owns, because a conversion barrier can originate in page design, pricing presentation, sales handoff, or technical performance, and the test result implicates whichever cause the data supports rather than whichever team is asked. This forces marketing, product, and sales into a shared view of where buyers actually stall, replacing departmental opinion with a governed record of what changed and what it moved. Organisations that treat it this way gain a standing diagnostic instrument, not a one-time redesign.
The recurring failure is calling a result before the sample and duration justify it, so a test stopped at the first sign of a lift is stopped on noise rather than signal, and the reversal shows up months later as an unexplained dip nobody can trace back to its cause. A second failure runs several overlapping tests against the same traffic, so interaction effects contaminate every result and no single change can be credited with the outcome. Programmes that avoid both failures pre-register sample size and duration before launch and sequence tests deliberately rather than running them concurrently for convenience.
In B2B specifically, the conversion event is rarely a purchase, and is more often a demo request or a qualified form fill, evaluated by a buying committee averaging six to ten stakeholders working across weeks rather than a single visitor deciding in one session. Reading the discipline in isolation from that committee dynamic overstates what a landing page change can achieve, because a page rarely moves a decision that sales and product must still close. The strongest programmes size each test against the specific stage of the buying journey it actually influences.
How It Works
A programme opens with diagnosis: analytics, session recordings, and heuristic review identify where visitors drop off and why, producing a ranked hypothesis list rather than a single guess. Each hypothesis is prioritised by estimated revenue impact against implementation effort, commonly through a framework such as PIE or ICE, so the test queue reflects commercial stakes rather than whichever change is easiest to ship. A test is then designed with a pre-registered sample size, minimum detectable effect, and stopping rule, then run as an A/B or multivariate split against a held-out control, and read only once that pre-registered threshold is reached, never at the first sign of a lift, since early movement in either direction is routinely noise rather than a genuine effect worth acting upon. Winning variants are implemented, monitored past launch for a reversal, and the finding is logged in a shared record so the next hypothesis builds on accumulated evidence rather than restarting the diagnosis from a blank page each quarter, keeping the queue from running dry between planning cycles.
Advisory Insight
Conversion Rate Optimisation is where Dromley's Growth Marketing practice converts strategy into compounding revenue. Organisations that run tests without advisory support optimise isolated pages in place of the full buying path, mistake short-lived statistical noise for a genuine lift, and roll changes into production that later reverse under a longer sample, eroding confidence in the programme itself and, eventually, executive appetite to fund it. A senior-led engagement starts by mapping the funnel to the buying committee's real decision points, prioritises the tests carrying the largest revenue exposure first, and installs the statistical discipline that turns isolated wins into a governed, repeatable system the organisation can keep running on its own.
Common Misconceptions
MYTH
A high-traffic page with a low conversion rate mainly needs more visitors sent to it, not a rebuilt or reworked page.
REALITY
More traffic multiplies an existing conversion problem rather than solving it, since the same proportion of visitors still fails to act. Fixing the page raises the yield of every visitor already arriving, compounding across every channel that sends traffic there.
MYTH
A test that shows a promising lift after only a few days of results can safely be rolled out as a permanent gain.
REALITY
Early leads are frequently noise, not signal, and reversing a change after rollout costs more credibility than waiting. A result only qualifies once the pre-registered sample size and duration are reached, holding through a full weekly cycle.
MYTH
Conversion Rate Optimisation is fundamentally an e-commerce discipline, built entirely around the checkout page.
REALITY
In B2B the conversion event is a demo request or a qualified form fill, evaluated by a multi-stakeholder committee over weeks, not a single visitor completing a purchase in one session. The methodology transfers; the event and the buying pattern behind it do not.
MYTH
Running several tests on the same set of pages at once always accelerates the overall pace of improvement achieved.
REALITY
Concurrent tests on shared traffic contaminate each other through interaction effects, making it impossible to credit any one change for the result. Sequencing tests deliberately produces slower but attributable gains that compound; simultaneous tests often produce neither.
Sources & Further Reading
Digital Marketing Strategy to Increase Sales Conversion on E-commerce Platforms
The Impact of Website Usability and Mobile Optimization on Customer Satisfaction and Sales Conversion Rates in E-commerce Businesses in Indonesia
A Click Conversion Rate Model of E-Commerce Platforms Aiming at Effective Data Sparse
Beyond Clickstream: Web Analytics and Session Duration in E-commerce Conversion
related
Growth MarketingDecks are easy. Decisions are not.
Bring us the real question. We’ll come back with how we’d approach it. Not a brochure. A starting point.
Take the Next Step