The CRO metrics that matter most are revenue per visitor, your primary conversion rate broken down by segment, and the funnel drop-off points that explain why that rate is what it is — everything else, including most of what dashboards show by default, is supporting detail. Tracking the wrong numbers is one of the quieter ways a CRO program stalls, because a team can look busy watching metrics that never actually explain what to fix next.
- The Metric Hierarchy: North Star, Primary, and Micro-Conversions
- Conversion Rate Isn't One Number
- Revenue-Level Metrics That Outrank Conversion Rate Alone
- Funnel and Drop-Off Metrics
- Statistical and Test-Health Metrics
- Metrics by Business Model
- Vanity Metrics to Stop Reporting
- Building a CRO Dashboard Stakeholders Actually Read
This article lays out the metric hierarchy we use at Salterra Digital Services and teach in the SEO University CRO track: which numbers deserve a dashboard, which belong in a weekly glance, and which ones — despite being popular — rarely deserve the attention they get.
The Metric Hierarchy: North Star, Primary, and Micro-Conversions
Not every number deserves equal weight, and confusing the levels is a common source of bad decisions. A north star metric is the single number the whole program is accountable to — often revenue per visitor or cost per acquired customer. A primary conversion metric is the specific action a page or funnel exists to drive: purchases, form submissions, bookings. Micro-conversions are the smaller steps along the way — add-to-cart, form starts, video views — that reveal where people stall before ever reaching the primary goal.
The hierarchy matters because micro-conversions are diagnostic, not the goal itself. A page can improve every micro-conversion in sight and still not move revenue, if the improvement pulled in a lower-intent audience or made an early step easier without fixing what happens later. Treat micro-conversions as clues, and treat the north star as the only number that ultimately decides whether a change was worth shipping.
Conversion Rate Isn't One Number
A single site-wide conversion rate hides more than it reveals. The number that matters is conversion rate segmented by traffic source, device, new versus returning visitor, and landing page — because averaging across all of those flattens exactly the differences that tell you where to focus.
- By device: mobile and desktop conversion rates frequently differ enough that a single blended number misrepresents both.
- By traffic source: organic, paid, and email traffic arrive with different intent levels, and mixing them obscures whether a page problem or a traffic-quality problem is driving a low rate.
- By new vs. returning: returning visitors typically convert at a different rate, and a shift in that ratio can move your blended average without any page actually changing.
Set up these segments in your analytics platform before you draw any conclusions from a headline conversion rate. A rate that looks flat overall can be masking a real win on one segment and a real loss on another, and only the segmented view shows you that.
Revenue-Level Metrics That Outrank Conversion Rate Alone
Conversion rate answers “how often,” not “how much it was worth,” which is why revenue-level metrics belong closer to the top of any serious CRO dashboard. Revenue per visitor (RPV) combines conversion rate and order value into one number, and it’s a better test-success metric than conversion rate alone, because a test can lift conversion rate while quietly dragging down average order value.
For ecommerce, track average order value (AOV) alongside conversion rate on every test, so a discount-driven lift doesn’t get celebrated as a pure win. For subscription or service businesses, customer lifetime value (LTV) and cost per acquisition (CPA) matter more than a single conversion event, since a change that attracts more but lower-retention customers can look like a win in month one and a loss by month six.
Where possible, tie test results back to actual revenue in your analytics or CRM rather than stopping at the conversion event itself. A “successful” lead form test that produces leads sales can’t close isn’t success by any definition that matters to the business.
Funnel and Drop-Off Metrics
Funnel metrics show you where visitors leave the path to conversion, and they’re usually the fastest route to a strong hypothesis. Build a funnel exploration — landing page to product or service page to cart or form to confirmation — and look for the step with the steepest percentage drop, not just the step with the most total visitors leaving.
Two funnel-adjacent metrics deserve their own tracking line: cart or form abandonment rate, which flags friction at the commitment moment, and time-to-conversion, which shows whether visitors are converting quickly or circling back multiple times before deciding — a pattern that often points to trust or price-comparison hesitation worth investigating qualitatively.
Field-level form analytics, where available, go a step deeper than page-level funnel data by showing exactly which form field causes the most abandonment, which turns a vague “forms underperform” observation into a specific, testable fix.
Statistical and Test-Health Metrics
A test result is only as trustworthy as the statistics behind it, and a mature CRO program tracks test health alongside test outcomes. Statistical significance (commonly a 95% confidence threshold) tells you whether an observed difference is likely real rather than noise, but it should always be read alongside sample size and minimum detectable effect — a significant result on too small a sample, or one that barely clears a trivial lift threshold, isn’t a reliable basis for a permanent change.
Track your program’s win rate (the percentage of tests that ship a validated improvement) and average lift per win over time. A declining win rate often signals weakening hypothesis quality or premature test calling, well before it shows up anywhere else in the metrics.
Metrics by Business Model
The right supporting metrics shift by business type, even though the hierarchy above applies universally.
- Ecommerce: conversion rate, RPV, AOV, cart abandonment rate, and product page add-to-cart rate as a leading micro-conversion indicator.
- Lead generation: form completion rate, cost per qualified lead (not just cost per lead), and lead-to-close rate fed back from sales to confirm lead quality held up.
- SaaS and subscriptions: trial-to-paid conversion, activation rate within the trial period, and early churn, since a signup win that doesn’t survive onboarding isn’t a real win.
- Local and service businesses: call and booking conversion rate, form-to-appointment show rate, and, where relevant, click-to-call rate from mobile search and map listings.
Vanity Metrics to Stop Reporting
Some numbers show up on dashboards by default and rarely deserve the attention they get. Pageviews and bounce rate alone say almost nothing about conversion intent without context. Click-through rate on a CTA, tracked in isolation, can improve while downstream conversion or revenue stays flat or drops, especially when a headline change attracts curiosity clicks from a worse-fit audience.
Reporting these numbers isn’t wrong — they have diagnostic value — but presenting them as headline program metrics creates a false sense of progress. If a monthly report leads with pageviews or CTR instead of revenue-level or segmented conversion metrics, that’s worth correcting before it shapes a stakeholder’s expectations incorrectly.
Building a CRO Dashboard Stakeholders Actually Read
A useful dashboard is short. Lead with the north star metric and its trend, follow with primary conversion rate segmented by the two or three cuts that matter most for your business, and include a running log of tests in flight, shipped, and their measured impact. Anything beyond that belongs in a deeper report, not the front page.
Update on a fixed cadence — weekly for the team running tests, monthly for stakeholders who need the summary. In the AI-search era, it’s also worth adding a light segment for AI-referred or assistant-referred traffic where your analytics can identify it, since that segment tends to arrive with different intent and is worth watching separately as its share of traffic grows.
Frequently Asked Questions
What's the single most important CRO metric to track?
If forced to pick one, revenue per visitor, because it combines conversion rate and order value into a number that can't be gamed by a change that helps one at the expense of the other. Most programs still need several supporting metrics beneath it.
Should I trust a test the moment it hits statistical significance?
Not by itself. Confirm the sample size matches what you calculated before launch and that the test has run through at least one full weekly cycle. Significance reached early, before the planned duration, is a common source of false winners.
How many metrics should a CRO dashboard actually show?
Fewer than most teams start with. A north star metric, two or three segmented conversion cuts, and a test log covers what most stakeholders need; additional detail belongs in a deeper report pulled on request, not the default view.
Is conversion rate a good metric to optimize for on its own?
It's useful but incomplete. Pair it with a revenue-level metric like RPV or AOV so a lift in conversion rate isn't accidentally masking a drop in order value or lead quality.
What metric should lead generation businesses track that ecommerce sites don't need?
Cost per qualified lead and lead-to-close rate, fed back from sales. Raw form completion rate can rise while lead quality drops, and only a metric tied to what sales actually closes reveals that gap.
How does SEO University teach metric selection differently from a generic analytics course?
We tie every metric back to the business outcome it's meant to prove, drawing on how Salterra Digital Services builds reporting for client CRO programs. A metric that doesn't change a decision isn't worth a permanent spot on the dashboard, regardless of how easy it is to pull.
Terry has 30+ years in software and SEO. He’s the founder of Salterra Digital Services and SEO Spring Training, host of the Roundtable SEO Mastermind, and lead instructor at SEO University — teaching the exact tactics his team uses on client work.
Continue learning
Ready to master this?
This guide is one lesson from the Conversion Rate Optimization course. Get every lesson, framework and checklist — plus the full 38-course catalog — inside SEO University.
Practitioner-focused training across the full digital marketing stack — from technical SEO to conversion optimization and the AI search era. By Salterra Digital Services, since 2011.