What Is Conversion Rate Optimization (CRO)?

Conversion rate optimization (CRO) is the practice of systematically increasing the percentage of visitors who take a desired action — signing up, starting a trial, purchasing, or submitting a lead. The key word is systematically: CRO is a repeatable loop grounded in behavioral data, not a one-off redesign or a designer's taste.

The definition, unpacked

A conversion is any action you've defined as valuable: a trial start, a purchase, a demo booking, a newsletter signup. Your conversion rate is conversions divided by unique visitors over a window. CRO is everything you do to move that rate up on purpose and measurably.

That distinguishes it from two things it's often confused with:

• It's not just A/B testing. Testing is one technique inside CRO used to validate a change. Plenty of high-value CRO is fixing things that are obviously broken — a CTA below the fold, a confusing headline — where a test would only slow you down.
• It's not a redesign. A redesign changes everything at once, so you learn nothing about why the number moved. CRO changes things deliberately and attributes the result.

The CRO loop (a methodology you can run weekly)

The whole practice reduces to a four-step loop you repeat:

1. Observe. Look at how visitors actually behave — where they land, how far they scroll, what they click, and where they leave. This is the step most teams skip, and it's where the leverage is.
2. Hypothesize. Turn an observation into a falsifiable statement: "Most visitors never reach the pricing section because the hero is too tall, so moving the CTA above the fold will raise signups."
3. Change. Ship the smallest change that tests the hypothesis. One change at a time, so the result is attributable.
4. Measure. Compare the conversion rate before and after (or run an A/B test if you have the traffic). Keep the win, revert the loss, and feed what you learned back into step 1.

The discipline is in doing this continuously against your own baseline — not in any single clever idea.

A worked example

Say your signup page converts 2.1%. You observe in your analytics that the median visitor scrolls only 40% of the page, and your primary CTA sits at 70% depth. That's an observation (step 1).

Hypothesis: the CTA is below where most people ever look, so a duplicate CTA at 30% depth will lift conversions (step 2). You add one button (step 3). Two weeks later the page converts 2.9% on comparable traffic (step 4). You keep it, and your new baseline is 2.9% — then you go looking for the next leak.

That's CRO. Not a guess that the button should be green; a behavior signal, a hypothesis, a change, and a measured outcome.

Grounding step 1 in real behavior (HogQL)

The "observe" step is concrete if you have analytics. Here's a PostHog HogQL query that surfaces the average maximum scroll depth on your signup page over 30 days — the exact signal from the example above:

SELECT
  round(avg(properties.$prev_pageview_max_scroll_percentage) * 100, 1)
    AS avg_max_scroll_pct,
  count() AS sessions
FROM events
WHERE event = '$pageleave'
  AND properties.$pathname = '/signup'
  AND timestamp > now() - INTERVAL 30 DAY

Illustrative sample output:

avg_max_scroll_pct	sessions
41.3	2,880

Using $pageleave (not $pageview) counts bounced single-page sessions too, so the number reflects reality rather than only your most engaged visitors. An average max scroll of ~41% tells you anything below the fold is being seen by a minority — a textbook CRO opportunity.

Where CRO fits

SEO and ads grow the top of your funnel; CRO improves the rate at which that traffic converts. The two compound: doubling traffic and doubling conversion rate quadruples results. For solo founders especially, CRO is often the cheaper lever — you already paid to acquire the visitors; CRO is about not wasting them.

If you want a reference point for where your rate stands, see what counts as a good SaaS landing page conversion rate. Then run the loop.