RPV by Segment
A framework for slicing revenue per visitor by channel, device, vertical, and audience — with benchmark ranges to compare against and a read on what aggregate RPV hides.
RPV by Segment
Splitting revenue per visitor across traffic source, device, vertical, and audience to surface what aggregate RPV hides.
RPV by segment is the practice of decomposing revenue per visitor — total revenue divided by total sessions — into the cuts that actually drive operational decisions: traffic source, device, vertical, and audience. Aggregate RPV is a single number that blends a €0.40 organic-social session with a €6.20 returning-email session into one mean that describes neither.
The framework treats segmentation as the unit of analysis. Each axis answers a different question: channels tell you where margin lives, devices tell you where the experience breaks, verticals calibrate ambition, and audiences tell you who's worth paying to reacquire. Read together, they convert a vanity number into a budgeting input.
Most stores stop at aggregate revenue per visitor and a topline conversion rate. That's enough to spot a trend, not enough to act. When RPV drops 8% week over week, the question isn't "is it down" — it's which segment moved, and whether the move is a traffic-mix shift or a real experience problem.
The four cuts below — source, device, vertical, audience — are the ones that change weekly budgets, sprint priorities, and creative briefs. Each section gives you the typical range to expect and what a gap usually means before you start testing.
Cut 1: RPV by traffic source
Traffic source is the highest-signal cut because it maps directly to spend. A Meta prospecting visitor and a branded-search visitor are not the same person at the same stage — they should not be evaluated against the same RPV target. Most Shopify stores in the €1–15M band see a 4–8x spread between their worst and best channel.
Branded search and email/SMS typically lead, paid social prospecting trails, and organic social sits in the middle with high variance. The actionable read is per-channel: a paid-search RPV under €1.20 with a €0.80 CPC means you're losing money on every click, regardless of how healthy aggregate RPV looks. See the dedicated breakdown in RPV by Traffic Source for the per-channel diagnostic flow.
Cut 2: RPV by device
Mobile usually carries 65–75% of sessions and 45–55% of revenue. That gap is the device tax — and it's where most experimentation budget should go, because the fix is on-site, not in media. If your mobile RPV is more than 40% below desktop, you have a checkout or PDP problem, not a traffic problem.
Tablet is often ignored and shouldn't be. On apparel and home, tablet RPV frequently beats both — older, higher-AOV shoppers, longer evening sessions. Worth segmenting separately rather than rolling it into mobile.
The mobile RPV gap is the highest-leverage segment
If mobile sessions outnumber desktop 3:1 but mobile RPV is half, closing 20% of that gap moves total revenue more than any channel reallocation. Audit cart drawer, address autofill, Apple Pay placement, and PDP image loading before you touch ad budgets.
Cut 3: RPV by vertical and audience
Vertical sets the ceiling. Beauty and supplements run RPV €3–6 on subscription-heavy assortments; apparel runs €2.50–5 with seasonal swings; furniture and home runs €5–12 but with 5x lower conversion rate. Without the vertical calibration, you'll either over-celebrate a beauty RPV that's actually mid-market, or panic about a furniture RPV that's perfectly healthy. The full ranges are in RPV Benchmarks by Industry.
Audience is where Revenue Per Visitor stops being a marketing metric and becomes a retention metric. Returning customers typically run 2–4x the RPV of first-time visitors. If your returning-visitor RPV is under 2x new, your post-purchase flow, lifecycle email, and on-site personalisation are leaking value — that's a CRM problem dressed up as a CRO one.
Typical RPV ranges by segment cut (DTC Shopify, €1–15M band)
Frequently asked questions
RPV (revenue per visitor) is total revenue divided by sessions. Unlike conversion rate, it captures AOV changes too — so a 4% CR with a €120 AOV beats a 5% CR with a €80 AOV. Segmenting it isolates where the value actually comes from, because a single aggregate hides 4–8x spreads between channels and a 40%+ mobile-desktop gap.
On Shopify stores in the €1–15M band, branded search and email/SMS typically run €5–7, non-brand paid search €2–3, paid social prospecting €0.80–1.50, and organic social €1–2.50. Vertical shifts these meaningfully — see the per-channel breakdown for diagnostic ranges.
Weekly at the channel level during active media spend, monthly at the device and audience level, and quarterly at the vertical-benchmark level. Channel mix moves daily with campaign changes; device behaviour moves with site releases; vertical benchmarks barely move at all.
Almost always a UX issue, not a traffic issue. The usual suspects: slow PDP image load, cart drawer friction, address form without autofill, Apple Pay or Shop Pay buried below the fold, and discount-code fields that distract from checkout. Audit those before assuming the traffic is lower-intent.
Yes for the aggregate, but always report new vs returning side by side. Blending them hides the real picture: a healthy aggregate often masks weak new-visitor RPV propped up by a small, loyal returning cohort. Set acquisition targets against new-visitor RPV only.
AOV is revenue per order — it ignores everyone who didn't buy. RPV is revenue per session, so it includes the 95%+ who left without ordering. Improving AOV without improving RPV usually means you raised prices and lost conversions; improving RPV means you actually moved the funnel.
Around 1,000 sessions per segment per week as a floor for directional reads, 5,000+ for confident week-on-week comparisons. Below that, daily RPV swings ±30% are just noise. For low-volume channels, aggregate to 28-day rolling windows instead of weekly.
Per-channel RPV minus per-channel CAC tells you contribution margin per visitor for each source. That's the number that should drive reallocation, not blended ROAS. A channel with €1.80 RPV and €1.20 effective CPC is profitable; one with €3.50 RPV and €4.00 CPC isn't, regardless of how the blended view looks.
Yes, and you should — it's the fifth cut most teams add once the first four are stable. Landing-page RPV exposes which PDPs and collection pages convert which traffic, and it's how you find the mismatches (e.g. paid social driving to a generic homepage instead of the offer page).
GA4 handles source, device, and basic audience cuts natively if your channel grouping is clean. Vertical benchmarking needs an external reference. Metricuno imports historical GA4 on day one so you get all four segment views without waiting 30 days for cold-start data — useful when you're auditing a new account or rebuilding broken UTMs.
Track CAC, channels, and funnel conversion in one place
Metricuno connects ad spend, funnel events, and revenue so you can see CAC by channel, cohort, and campaign — without stitching together five tools.