PRODUCT DOCS
Docs/Integrations/Hub

Cross-Channel Analytics for Ecommerce: Google, Meta and Amazon Ads Beside Store Sales

Cross-channel analytics in Luca AI reads Google Ads, Meta, Amazon Ads and TikTok Ads read-only and sets their spend against the revenue Shopify recorded, as MER, CAC and break-even ROAS.

Live5 min read

Cross-channel analytics in Luca AI starts from your store, not the platforms: Luca reads Google Ads, Meta, Amazon Ads and TikTok Ads read-only and sets total and per-platform spend against the revenue Shopify actually recorded, as MER, CAC and break-even ROAS. Where a brand wants orders assigned to ads, Luca's team sets up a multi-touch attribution model with it, and Luca never changes a budget, bid or campaign.

Out of the box, Luca compares what you spent and what each platform reports with what your store took, and it does not assign orders to ads.

At a glance

Answer in 20 seconds
Google AdsPaid search, Shopping, Performance Max and YouTube. Connects self-serve on the Google card, read-only
MetaFacebook and Instagram ads. Some accounts connect by OAuth; others through guided partner access with Luca's team
Amazon AdsSponsored ads by campaign, keyword and search term. Connects self-serve, read-only
TikTok AdsOn the TikTok card in the Marketing group, with spend and reach
RefreshEvery source refreshes once a day, overnight
What never changesYour budgets, bids and campaigns. Luca reads each ad account and writes to none of them
Built from Luca's code-checked product facts · as at 20 Nov 2026
01

What this area covers

Two ad platforms have their own pages today:

  • Google Ads reporting for Shopify: spend and results by campaign type, including Shopping and Performance Max, with search terms and impression share, beside your store revenue.
  • Meta Ads reporting: Facebook and Instagram spend, reach and Meta's reported conversions down to each ad and creative.

Amazon Ads and TikTok Ads will each get their own page. Until then, the rows above say what each one brings in. Your Amazon store itself connects separately, through Amazon Seller Central.

02

Multi-touch attribution

Some brands want more than spend set against revenue. They want orders assigned to ads, with the double counting between platforms taken out. For those brands, Luca's team sets up a multi-touch attribution model with each brand, case by case.

You agree the model with Luca's team, and it works from the store and ad accounts you have already connected. Once it is set up, Luca shows which ads produced your orders. It is not switched on by default.

Until then, the figures in this area are what each platform reports, set beside the revenue your store recorded. Ask your Luca contact if you want a model set up for your brand.

03

Which figure answers your question

Start from the question your founder or finance lead asked, then open the page that owns the figure.

The question you're askingWhere to read itWhat it counts
Does our total ad spend earn more than it costs?Marketing efficiency ratioAll store revenue divided by all ad spend, beside each platform's ROAS
What does a new customer cost us?Customer acquisition costBlended CAC and CAC per ad platform, week by week
What ROAS do we need to break even?Break-even ROASOne divided by your gross margin, set against your ad spend
How fast do new customers repay us?CAC paybackThe months it takes a new customer to pay back what they cost
How do ads compare with email and search?Marketing performance analysisAd spend beside Klaviyo, Search Console and GA4 results

A page on store ROAS against platform ROAS is planned for this area. For ROAS by channel and campaign in one job, the ROAS analysis use case walks through it.

04

What to check before you trust the numbers

  1. Don't add the platforms' revenue together. Each platform counts the conversions it claims, so one order can appear in Google and Meta at once. Summed platform revenue can then exceed what your store recorded.
  2. Read MER against store revenue instead. Treat each platform's ROAS as that platform's own claim, and judge total spend by MER, all store revenue divided by all ad spend. Our post on why platform ROAS can mislead you explains the gap.
  3. Check each card's last sync time. On Sources, every connected card shows when it last synced. A stale card means yesterday's spend is missing from today's figures.
  4. Reconnect an account whose access has expired. Ad platforms can expire Luca's access. Reconnect the source, and the next overnight refresh brings its data back.
05 · Common questions

Common questions

Can Luca's team set up attribution that ties orders to ads?

Yes, once Luca's team has set up a multi-touch attribution model with your brand. Without that model, Luca sets each platform's spend and reported results beside your store revenue, and it does not assign orders to ads.

Does Luca change bids or pause campaigns?

No. Luca reads your ad accounts and never changes a budget, bid or campaign, so every decision stays with you. The read-only access page explains how this works.

Do I connect Amazon Ads separately from Seller Central?

Yes. Amazon Ads is its own source, for sponsored ads spend and performance. Seller Central is a separate source for your Amazon sales, fees and stock, and it connects per region.

Does Luca support Pinterest, Snapchat or Microsoft Ads?

Not as standard sources. Luca's team can build a connector for them on request, at no extra fee. Connecting a source covers how to ask for one.

Why does Meta need Luca's team?

Some Meta accounts connect by OAuth, and others need guided partner access with Luca's team. If Meta asks you to get in touch, your Luca contact walks you through it. The Meta Ads page has the details.

Can I compare channels on cost per new customer?

Yes. Luca works out CAC for each ad platform as well as blended CAC across all of them. The CAC page shows how to read the two side by side.

Luca AI Product team · Last verified 20 November 2026

Something wrong or missing? Send the page link with your correction to the Luca AI team.

Your numbers, answered.

Ask Luca about your own store.

Browse the product docs →Book a demo