See where profitable demand really starts
Connect Meta, Google and TikTok and Comver joins ad spend to Shopify revenue as three revenue truths — what the platform reports, what Comver attributes to real Shopify orders, and total Shopify revenue — never blended into one number. Fifteen traffic sources ranked by the revenue they actually create, wasted spend flagged by transparent rules, and every budget decision proposed to you, not made for you.
- Three revenue truths, never blended
- Wasted spend is rule-based, not an AI guess
- You approve, edit or dismiss — every decision logged
AI Commerce Ads · Traffic sources
Every measurable path into the store, ranked by Shopify revenue · last 30 days
Platform-reported revenue
$64,000
What Meta, Google and TikTok claim
Comver-attributed revenue
$50,300
Shopify orders tied to paid touchpoints · gap −21%
Total Shopify revenue
$148,200
Blended MER 6.9× · 15 measured sources
Estimated wasted spend
$3,840
Rule-based, not an AI guess
Overlapping rules are not double-counted.
Scale a stable winner by 15%
Google Ads · Vitamin C — high-intent search · 5.1× ROAS in the selected period; budget efficiency stayed inside the 8% stability band for seven days.
15
Measured traffic sources
Meta, Google and TikTok Ads beside Google Organic, Direct, Email, Instagram, Bing, Pinterest, YouTube, Reddit, LinkedIn, other referrals — and ChatGPT and Perplexity
3
Revenue truths, kept apart
Platform-reported · Comver-attributed · total Shopify revenue (Blended MER)
4
Guardrails behind wasted spend
Below target ROAS, above the CAC guardrail, spend without purchases, fatigued creatives — overlaps never double-counted
0–100
Traffic quality per source
Green from 70, amber from 50, red below — with conversion, AOV and bounce beside it
Beta. Live figures appear once an ad platform is connected; Comver never shows demo numbers outside an explicitly labelled demo mode.
Three revenue truths, clearly separated
Ad platforms report the conversion value they claim. Comver counts the Shopify orders that actually followed a paid touchpoint inside the attribution window. And the store has one total. Most dashboards blend those into a single ROAS; AI Commerce Ads shows all three side by side, so over-claiming is visible at a glance and every efficiency number says which truth it was built on.
- Platform-reported revenue — conversion value divided by platform spend gives Platform ROAS
- Comver-attributed revenue — Shopify orders tied to paid touchpoints give Comver attributed ROAS, CAC and new-customer revenue
- Total Shopify revenue — all revenue divided by all ad spend gives Blended MER; the Platform → Comver gap is printed, not hidden
Attribution bridge
Three revenue truths, clearly separated
Conversion value the platforms report
Shopify orders tied to paid touchpoints inside the window
Every Shopify order, paid or not
Platform → Comver gap
−21%
Over-claiming, visible at a glance
Blended MER
6.9×
All Shopify revenue ÷ ad spend
Paid source efficiency
Comver ROAS per platform against the portfolio line
Fifteen paths into the store, ranked by Shopify revenue
Paid is only part of the picture. The Traffic sources tab measures every path a shopper takes into your store — paid search and social, organic search, organic social, owned email, direct, referrals and AI discovery from ChatGPT and Perplexity — each with sessions, Shopify revenue and share, conversion rate, AOV, bounce, a 0–100 traffic-quality score, a seven-period trend and a status. Pick any slice of the donut and Copilot writes the brief for that source.
- Direct and assisted Shopify revenue per source — a source can influence revenue even when it is not the final click
- A traffic-quality map of conversion against bounce, with “Efficient traffic” and “Conversion leakage” quadrants sized by revenue
- Revenue mix by acquisition system: paid search, paid social, organic search, organic social, AI discovery, owned, direct, referral
Traffic source command center
See where profitable demand really starts
Paid, organic, owned, referral and AI discovery traffic measured against Shopify revenue.
Shopify revenue
$148,200
Meta Ads selected
Comver Copilot · Meta Ads
Scaling opportunityStrong scale potential, with one broad campaign wasting budget.
Revenue
$24,600
16.6% share
Sessions
18,420
13.9% share
Conversion
2.4%
AOV $58
Traffic quality
74 / 100
Bounce 41%
Wasted spend is rule-based, not an AI guess
Comver puts a number on the budget you should protect — and shows the arithmetic. Four deterministic guardrails run across every paid source: spend below the target ROAS, spend above the CAC guardrail, spend that produced no purchases, and spend still attached to creatives whose fatigue score has crossed the line. Where rules overlap, the spend is counted once. The Copilot surfaces the total as “Budget to protect” and explains it in words; the numbers themselves come from rules, never from a language model.
- Four transparent guardrails — target ROAS, CAC guardrail, no-purchase spend and creative fatigue — with overlaps not double-counted
- “Analyze sources” re-runs the guardrail analysis across all paid sources on demand
- The disclosure travels with every result: numeric recommendations are calculated from deterministic campaign rules; language is explanatory only
Estimated wasted spend
$3,840
Rule-based, not an AI guess
Last 8 periods · 18% of paid spend
Spend below target ROAS, above target CAC, without purchases, or attached to fatigued creatives. Overlapping rules are not double-counted.
Budget to protect
Transparent rule-based estimateProposed to you, never decided for you
Every recommendation follows the same evidence chain — source → traffic quality → Shopify outcome → action → impact — and arrives as a card with a diagnosis, three evidence lines, the recommended action, the estimated revenue or protected spend and a confidence score. Open the decision review to see the current value, the proposed change and the confidence in three steps, edit the proposed value if you disagree, and confirm. The decision is written to the audit log and appears in Decision history, where it stays after every refresh.
- Current value → Proposed change → Confidence, plus an “Expected business impact” box that names its assumption
- Edit the proposed value — a 15% budget step can become 10%, a pause can become a 20% reduction
- Every write action to a platform carries an explicit confirmation state; nothing is auto-optimised behind your back
Stop inefficient spend and protect CAC
Fourteen days of spend without a matching Shopify order pattern; CAC sits well above the guardrail while the retargeting campaign is under-funded.
Evidence
- 1.3× Comver ROAS — below the 2.5× threshold
- CAC $71 versus the guardrail
- PDP conversion trails the blended store rate
Recommended action
Pause the campaign and move the recoverable budget to the strongest retargeting campaign.
Comver decision review
Edit proposed change
01
Current value
Active
02
Proposed change
Paused
03
Confidence
88%
Edited proposed value
Expected business impact
Fourteen-day budget protection at a 20% step. Estimated protected spend: $1,060.
Eight kinds of reviewable, reversible change
Each recommendation type is emitted by the same deterministic engine, with its own evidence lines and impact explanation — “based on the last seven-day marginal ROAS and a 15% budget step”, “fourteen-day avoided spend at the current daily budget”, “assumes a conservative 12% relative PDP conversion improvement”.
Scale budget
Increase a stable winner gradually — the engine proposes a step, you approve or trim it.
Reduce budget
Reduce exposure above the CAC guardrail while a new offer is tested.
Pause or resume
Stop inefficient spend and move the recoverable budget to the strongest retargeting campaign.
Shift budget
Move a share of inefficient prospecting budget toward high-intent demand — without increasing total spend.
Improve landing page
Post-click “intent leakage” — a high CTR with a low CVR — routes traffic to a PDP that matches the ad’s promise.
Replace creative
A negative CTR slope with rising frequency asks for fresh variants before efficiency erodes further.
Turn a winner into a test cell
Keep the proven benefit and offer constant and propose new openings from the winning hook.
Cap for inventory risk
When stock cover falls under the safety threshold, cap growth and shift demand toward products that can ship.
Four tabs, one set of filters
Overview · Traffic sources · Campaigns · Creative studio share a header of filters — date range, market and traffic source — that survives every tab switch. Connection state is reported per platform, never averaged into a single light.
Every campaign ranked by Shopify revenue
Attributed revenue, ad spend, blended ROAS and new customers on top; below, each campaign with its platform, objective, status — Scaling, Healthy, Learning, Attention, Creative fatigue, Critical, Paused — spend, revenue, ROAS, orders and a ten-day trend, beside the share of spend it concentrates.
AI agents
AI Commerce Ads
Live platform dataAttributed revenue
$50,300
10 campaigns
Ad spend
$21,400
Selected period
Blended ROAS
2.35×
Revenue ÷ spend
New customers
612
727 orders
Campaign performance
Revenue, efficiency and operational status in one scan.
Budget concentration
Share of spend by campaign
Spend distribution
From connection to the first protected dollar
- 1
Connect your ad platforms
Open Ads in the AI agents section and link Meta, Google and TikTok Ads from Manage connections. Credentials stay server-side; connection state is shown per platform.
- 2
Set the lens
Choose the date range, market and traffic source once — the filters follow you across Overview, Traffic sources, Campaigns and Creative studio.
- 3
Comver reads and joins
Spend, impressions, clicks and platform-reported revenue are read through one adapter layer per platform, joined to Shopify orders and sessions, and split into three separate truths beside every non-paid source.
- 4
Decide, then measure
Approve, edit or dismiss each recommendation. The decision goes to the audit log; KPI deltas, event annotations and source trends track what happened next.
Frequently asked questions
Meta Ads, Google Ads and TikTok Ads, each through its own typed adapter behind one unified data layer. Connection status is reported per platform — never averaged — and credentials live server-side only. The OAuth connectors are rolling out with the Beta; until your account is connected the workspace shows a clear “No ad platforms connected” state rather than any placeholder numbers.
Comver keeps three numbers separate: platform-reported revenue, Comver-attributed Shopify revenue (Shopify orders tied to paid touchpoints inside the attribution window) and total Shopify revenue, which yields Blended MER. The Attribution bridge shows all three and prints the Platform → Comver gap so over-claiming is visible at a glance.
No. The Traffic sources tab measures 15 sources — Meta, Google and TikTok Ads alongside Google Organic, Bing, Direct, Email, Instagram, Pinterest, YouTube, LinkedIn, Reddit, other referrals, and AI discovery sources such as ChatGPT and Perplexity — each with sessions, Shopify revenue, revenue share, CVR, AOV, bounce, a 0–100 traffic-quality score, trend and status.
It is a rule-based estimate, not an AI guess: spend below the target ROAS, above the CAC guardrail, spend without purchases, and spend attached to fatigued creatives — with overlapping rules not double-counted. The Copilot surfaces the same figure as “Budget to protect” and explains it in words; the number itself comes from deterministic rules.
No. Every recommendation is proposed with a diagnosis, evidence, current vs proposed value, estimated impact and a confidence score. You approve, edit the proposed value, or dismiss; every decision is written to the audit log (Decision history) and every write action to a platform carries an explicit confirmation state. Nothing is auto-optimised behind your back.
Scale budget, reduce budget, pause or resume a campaign, shift budget between channels, improve a landing page when post-click intent leaks, replace a fatigued creative, turn a winning hook into the next test cell, and cap spend when inventory cover is at risk. Each carries three evidence lines and an impact explanation that names its assumption.
It ranks your creatives by Shopify-attributed revenue and ROAS, tracks CTR fatigue over 14 days and tags each asset Winner, Healthy, Watch or Fatigued, with filters for winners, fatigue and AI-generated assets and a Creative Copilot summary of the current winning pattern. Formats it understands include AI UGC video, UGC script, static ad, carousel, product animation and ad copy pack. It is a performance library today; creative generation is on the roadmap.
AI Commerce Ads is in Beta. The workspace, the three-truths attribution model, traffic-source intelligence, the rule-based recommendation engine and the audit trail are built; live figures appear once your ad accounts are connected. Comver never substitutes demo numbers outside an explicitly labelled demo mode.
Find out where profitable demand really starts
Install Comver, connect Meta, Google and TikTok, and read the three revenue truths side by side. In Beta today — every number counted from your own store, never demo data.
14-day free trial · Live on Shopify in minutes · Checkout stays in Shopify