At a glance
Percentage of sessions that result in a purchase event. Conversion rate is the single highest-leverage metric in ecommerce. A 0.5 percentage-point improvement on 100K sessions equals 500 extra orders, at typical AOV £100-£250, that’s £50K-£125K in additional revenue per period. Average ecommerce conversion is 2-3%; top performers exceed 5%. The gap between average and top-tier represents the biggest single revenue lever a merchant can pull.
Calculation
Worked example
A UK-based BC store, conversion rate reading on Wednesday 15 May 26.
What the conversion reading is telling us:
- Site-wide conversion at 1.94% is below the 2% typical ecommerce baseline. Not at alert state (alert at <1%) but in the investigation zone. The vsP delta of -6.3% indicates a recent deterioration, something changed in the last 30 days.
-
Mobile vs desktop gap is severe. Desktop converts at 3.11% vs mobile at 1.50%, a 2x ratio. For a store with 70% mobile traffic, mobile conversion is the highest-leverage fix point. Cross-reference
psi_mobile_score_compareandcrux_mobile_pass; mobile conversion gap usually traces to mobile site speed or mobile checkout friction. -
Channel-by-channel reading:
- Email at 5.94% confirms the merchant has a healthy own-audience asset. Maintain.
- Paid search at 2.63% is strong, paid traffic landing on intent-aligned pages converts well.
- Organic at 2.10% is acceptable; investing in PDP optimisation could lift this significantly.
- Direct at 1.49% is concerning, direct visitors typically convert above-baseline. Investigate brand confusion or returning-user friction.
- AI at 3.36% is excellent given small volume. Cross-reference
ai_conversion.
-
Likely causes for site-wide -6.3% conversion change:
- Site speed regression: cross-reference
psi_perf_score_summaryandcrux_lcp_p75for the prior 30 days. - Checkout flow change: any recent app installs, theme updates, or checkout customisations? Test in incognito.
- Product mix shift: if low-converting categories now dominate traffic mix, headline conversion falls.
- Pricing change: any recent price increases on bestsellers?
- Trust signal change: was a payment method removed, a security badge swapped, or a review feed broken?
- Site speed regression: cross-reference
-
Recommended response, in priority order:
- Day 1: Run a checkout flow test in incognito on mobile and desktop. Check every step for errors.
- Day 1: Audit site speed (compare 30 days vs 60 days ago).
- Day 1-2: Audit any release deploys, app installs, or theme changes in the last 30 days.
- Day 2-3: Review pricing changes and trust signals.
- Day 7: Re-measure conversion; confirm trend reversal.
- Read the rate. Below 1% alert; below 2% investigate.
- Decompose by device, mobile-vs-desktop gap is the most common gap.
- Decompose by channel, identify which channels are driving the headline.
- Cross-reference site speed + checkout + pricing for upstream causes.
- Confirm with vsP delta, declining trend signals recent regression.
Sibling cards merchants should reference together
Reconciling against the vendor’s own dashboard
Where to look in GA4: Reports → Engagement → Conversions; or Reports → Monetization for ecommerce-specific conversion. Why our number may differ:
Quick rule: match conversion event definition first.
Known limitations / merchant FAQs
Q: Mobile converts at 1.5%, desktop at 3.1%. Should I just optimise mobile? Yes, that’s the highest-leverage fix point given mobile is likely 70%+ of your sessions. Mobile gap is usually 2 things: speed (CWV failures) and checkout friction (small touch targets, awkward form fields, slow payment flow). Cross-referencecrux_mobile_pass and psi_mobile_score_compare. A 1 percentage point mobile CR lift on a 75K-session-per-month mobile audience is 750 extra orders.
Q: Our CR dropped from 2.5% to 1.8%, what’s the most common cause?
Site speed regression (apps installed, theme update). Run psi_perf_score_summary for the 30 days vs 60 days ago. The second most common: checkout flow change (new payment app, address validation step). Test the checkout in incognito on mobile and desktop.
Q: Is 2% a good conversion rate?
Average. Top-quartile ecommerce sits at 3-5%; top-decile above 5%. Use category-specific benchmarks: fashion 1.5-2.5%, beauty 2.5-3.5%, supplements 3-5%, B2B 0.5-1.5%. Generic averages mislead.
Q: How does AI traffic affect conversion rate?
AI traffic typically converts 1.5-3x site-wide rate. As AI traffic grows, headline CR will rise gradually. Use the per-channel decomposition rather than headline if AI is becoming a meaningful share.
Q: Does this card include returning users or just first-time?
Both. The card is session-based (every session counts). For first-time-vs-returning split, GA4’s user reports give the breakdown, typically first-time CR is 0.5-1.5% and returning user CR is 5-10%.
Q: We have a high refund rate. Are refunded purchases counted as conversions here?
By default yes (the purchase event fired; the conversion happened). Configurable to exclude refunded if the merchant prefers. Note that refund rate is its own card (refund_rate); double-counting between the two is a real concern.
Q: Should I optimise for conversion or for AOV?
Both, but convert first. CR improvements scale with all traffic; AOV improvements scale with conversions. A 10% CR improvement is typically more impactful than a 10% AOV improvement because it compounds across more sessions. But after CR is healthy (3%+), AOV becomes the next lever.
Q: How do I get from 2% to 5%?
A 12-24 month investment cycle for most stores. The major levers: (1) mobile site speed (LCP, INP, CLS to all-pass), (2) checkout flow (one-page or short-step), (3) product page quality (rich descriptions, multiple images, reviews, social proof), (4) trust signals (badges, return policy, support channels), (5) personalisation (post-cart abandonment, post-purchase). Each lever produces 0.3-0.8 percentage point gains; layered they compound to 3+.