The Advertorial + Shopify Playbook
Run Taboola traffic to an advertorial, follow every reader across domains to your Shopify checkout, and retarget the ones who mattered on Meta and Google. One Tracking Tag, zero lost signals.
Advertorials sell. The funnel behind them is a tracking nightmare: native ad, then an advertorial on one domain, then a Shopify checkout on another. Browsers scope first-party data to each domain by design, so the click ID collected on the advertorial becomes invisible the moment the reader crosses to the store. By the time a sale lands, Taboola can’t see what drove it, Meta and Google never meet the readers who almost bought, and the ROAS your platforms report won’t match the revenue in your Shopify dashboard.
This playbook closes the loop. One Tracking Tag on each domain, stitched by Cross-Domain Tracking, feeds every conversion server-side to Taboola, Meta, and Google. Audiences built from real intent, not pageviews.
The Reader's Journey, From Native Ad to Attributed Sale
Follow one reader from the Taboola click, across the domain hop, through every step in your Shopify store, to a purchase that lands back in the campaign that earned it. Each step feeds your bidding, audiences, and reporting automatically.
Steps marked with the AnyTrack node run automatically. No extra code, no manual tagging.
Example Scenario
Think of it as being the affiliate of your own store. You build a top-10 product review site that promotes your own products, run ads to it, and let the editorial content do the selling. You still want exactly what any affiliate wants: to open your analytics and see how every conversion is attributed to the ad that earned it, and to make sure your ad platforms receive the real conversions they generated, not the fraction that survives the domain hop.
Concretely: you’re promoting a DTC sleep supplement. Taboola drives readers to a “5 Reasons You Wake Up Tired” advertorial on thesleepreview.com, a separate root domain that keeps the editorial feel; its CTA links to your Shopify store at yourbrand.com. Two different domains is the whole point: the browser won’t carry identity across them, so the tracking has to. You need three things to happen:
- Taboola gets purchase postbacks, so its bidding learns from buyers instead of bounce-prone clicks.
- Meta and Google get the conversion signals to retarget engaged readers across feeds, Search, and YouTube.
- Your Campaign Report ties Taboola campaigns to Shopify revenue, attributed through the whole funnel.
Here’s the part that makes it work: every event on the store, from a product view to the final purchase, travels two ways at once. It flows back to the advertorial property and out to the ad platforms connected there, so Taboola, Meta, and Google can credit the campaign that drove the sale. And the store’s own Tracking Tag sends it through each platform’s Conversions API, so your Shopify property reports the same revenue and builds its own audiences. One event, counted once, working on both sides of the domain hop.
Each domain is its own AnyTrack property, with its own Tracking Tag and its own configuration. That separation is what makes the two-domain funnel measurable:
| Configuration | Advertorial property | Shopify store property |
|---|---|---|
| Domain | thesleepreview.com | yourbrand.com |
| Runs your ads | Yes, readers land here | No, checkout only |
| Tracking Tag | In the site <head> | In the theme, plus the 5 Shopify webhooks |
| Cross-Domain Tracking | Store domain added (yourbrand.com) | Off |
| Sync Conversions | Off, it receives the synced conversions | On, syncs conversions to the advertorial property |
| Taboola Pixel ID | Connected | Optional |
| Meta Pixel ID | Connected | Optional, own CAPI |
| Google Ads conversion | Connected | Optional, own CAPI |
The advertorial property is where your ad pixels live, because that’s where the ads land. The store property collects the orders and syncs them back, so the campaign that paid for the click gets credited for the sale.
- An AnyTrack account on Personal or above (the Taboola server-side integration requires it)
- Admin access to both domains, the advertorial site and the Shopify store
- Active Taboola, Meta, and Google Ads accounts
- At least one Google Ads conversion action of type Website with manual setup
Step 1: Set Up a Property for Each Domain
Cross-domain tracking works between two AnyTrack properties, one per domain.
- Create a property for the advertorial domain and paste the Tracking Tag before
</head>on its pages. - Create a property for the store and complete the Shopify setup: Tag in the theme, plus the five webhook notifications that collect checkout events server-side.
Step 2: Connect the Domains
In the advertorial property, open Property Settings → Cross-Domain Tracking and add the store’s root domain. In the store property, toggle Sync Conversions. From now on, every advertorial link pointing at the store is AutoTagged with the click ID, and store conversions flow back to the advertorial property.
Add only the destination domain
Never add the property’s own domain: self-referral traffic corrupts tracking. Hover an advertorial CTA to confirm the atclid or atrefid parameter.
Step 3: Connect Taboola
In the advertorial property, open Tracking Pixels, select Taboola, add your Taboola Pixel ID, and toggle Server-Side Events. AnyTrack loads the Taboola pixel for you, sends engagement events client-side, and returns conversions server-to-server, so Taboola’s bidding optimizes on actual purchases. Once events arrive, create your conversions in Taboola from the events AnyTrack sends.
Attach the UTM template
Add AnyTrack’s pre-built UTM template to your Taboola campaigns: the click ID carries identity across domains, the UTMs credit the campaign.
Step 4: Connect Meta and Google Ads
Connect both platforms under Tracking Pixels, next to Taboola.
- Connect Facebook Ads: AnyTrack loads your Meta Pixel and sends conversions through the Conversions API, deduplicated automatically.
- Connect Google Ads, turn on server-side events, and map AnyTrack events to the conversion actions you created in Google Ads. AnyTrack automatically formats and sends the customer data Enhanced Conversions requires.
Standard events map automatically on the Meta side: Purchase lands as Purchase, with hashed customer data from checkout attached.
Step 5: Turn Events Into Audiences
Not every reader is worth the same bid, and the events you track become your audience seeds.
- Your warm pool is every reader who clicked through to the store. AnyTrack tracks
OutboundClickautomatically on each advertorial CTA. - Cart abandoners and buyers arrive from Shopify, tied to the same journey:
ViewContent,AddToCart,InitiateCheckout,AddPaymentInfo, andPurchase. - A deeper intent signal, like a 75% scroll on the advertorial, is one custom event away. The snippet below tracks it as
EngagedReader; remember custom event names need manual mapping to your ad platforms.
Paste the snippet on your advertorial pages, anywhere after the Tracking Tag:
<script>
// Fire "EngagedReader" once when the visitor scrolls 75% down the advertorial.
AnyTrack(function () { // callback = waits until the AnyTrack Tag is loaded
var fired = false; // guard so the event fires only once per pageview
function onScroll() {
if (fired) return;
// How far down the document the user has scrolled, as a 0-1 ratio
var scrolled = window.scrollY + window.innerHeight;
var docHeight = document.documentElement.scrollHeight;
var depth = scrolled / docHeight;
if (depth >= 0.75) { // 75% threshold
fired = true;
AnyTrack('trigger', 'EngagedReader'); // custom event, no revenue/attributes
window.removeEventListener('scroll', onScroll); // stop listening once fired
}
}
window.addEventListener('scroll', onScroll, { passive: true });
onScroll(); // handle short pages already past 75% on load
});
</script>
Step 6: Launch the Full-Funnel Campaign Stack
The tracking is the hard part, and it’s done. What’s left is four short builds in your ads managers, each one a few minutes of clicking, because the audiences and conversions they need are already flowing.
Top of funnel: Taboola prospecting
Fills the funnel with new readers at discovery prices.
- Destination: the advertorial on your content domain, with AnyTrack’s pre-built UTM tracking template attached to the campaign. The template is what ties every downstream purchase back to this campaign in your Campaign Report.
- Budget: start at a daily level you can hold steady for a week or two of learning, then scale winners in steps of about 20%. Big jumps reset what conversion bidding has learned.
- Audience: broad. Conversion bidding learns from the purchase postbacks AnyTrack sends, so let it find your readers instead of over-narrowing. Test a handful of headline variations against one strong article.
- Event tracking: optimizes on the
Purchaseconversions AnyTrack returns server-side; every click generates the sessions andOutboundClickevents that seed your warm pool downstream. - Conversion objective: conversions, optimized toward the
Purchaseconversion you created in Taboola from the events AnyTrack sends. If purchase volume is too thin for bidding to learn from, create the conversion on a mid-funnel event AnyTrack already maps for you (start_checkoutoradd_to_cart), then switch tomake_purchaseonce sales accumulate.
Middle of funnel: retarget warm readers on Meta
Brings back the readers who clicked through to the store but didn’t buy.
- Destination: the product page the advertorial sells, with the UTM tracking template from Tracking Pixels → your Meta Pixel → Ads Integrations pasted at the campaign level.
- Budget: roughly 10 to 15% of your Taboola spend to start. The audience size caps how much this campaign can usefully spend, so raise it as the warm pool grows.
- Audience: a website custom audience built from the
OutboundClickandEngagedReaderevents AnyTrack sends, excluding purchasers. A 14-day window is a good starting point. - Event tracking: fed by
OutboundClickandEngagedReader; generates theViewContent,AddToCart, andPurchaseevents that move readers into the next stage. - Conversion objective: Sales, optimized for the
Purchaseevent AnyTrack sends through the Conversions API.
Creative: the advertorial’s strongest proof points, restated as direct-response ads. These readers already know the story; remind them why it mattered.
Bottom of funnel: close cart abandoners
Converts the highest-intent segment you have.
- Destination: the product they left behind, via catalog ads on Meta, with the same campaign-level UTM tracking template attached.
- Budget: small and capped by the audience size. This pool is narrow, so watch frequency and let the audience, not the budget, decide the ceiling.
- Audience:
AddToCartwithoutPurchase, on a short window (7 days), purchasers excluded. Mirror it on Google as a remarketing list for your search and display campaigns. - Event tracking: fed by the
AddToCartandInitiateCheckoutevents arriving from Shopify; generates thePurchaseevents that close the loop and reseed the lookalike below. - Conversion objective: Sales, optimized for
Purchase.
Creative: the product they left behind, plus whatever genuine urgency you have: shipping cutoffs, low stock, a first-order offer.
Scale: lookalikes from real buyers
Finds the next thousand customers who look like your best ones.
- Destination: the same product pages as your middle of funnel, each platform’s UTM tracking template attached.
- Budget: treat it as prospecting: start near your Taboola daily level and scale whatever holds your target CPA. On Google, scaling is a budget decision, because Performance Max uses the enriched conversion signals AnyTrack already sends as its optimization input.
- Audience: on Meta, a Lookalike Audience seeded from your purchaser custom audience, starting at 1% and widening as it proves out. Keep purchasers excluded everywhere so you stop paying to re-win them.
- Event tracking: optimizes on
Purchase; every sale it generates enriches the purchaser seed it was built from. - Conversion objective: Sales on Meta, optimized for
Purchase; on Google, your mapped purchase conversion action driving Performance Max.
The Optimization Loop
Two loops keep this stack improving, and the faster one needs nothing from you. It runs on every event, not just sales: with Sync Conversions live, each funnel stage the store tracks (ViewContent, AddToCart, InitiateCheckout, AddPaymentInfo, Purchase) travels back to the advertorial property, and every platform connected there receives every stage. Each arrival does its own quiet work. A reader who adds to cart joins your abandoner retargeting on Meta and Google. A checkout hands Taboola’s conversion bidding a mid-funnel lesson while purchase volume is still building. A completed order retires its buyer from retargeting across the stack, adds them to the purchaser seed your lookalike draws from, and shows Taboola, Meta, and Google one more example of a converting reader. AnyTrack sends the signals in real time; each platform works them into bidding and audiences at its own pace (Taboola surfaces server-side conversions within 30 to 45 minutes, and Google attributes within hours). None of it needs a slot on your calendar.
The slower loop is yours, and its job is to feed the fast one better raw material. One pass a week:
- Measure. Start in the Campaign Report and read revenue per campaign: attributed revenue for Taboola prospecting (its spend lives in Taboola’s own dashboard), then spend, revenue, and ROAS side by side for the Meta and Google campaigns. While you’re on Meta, check Event Match Quality; it should climb as checkout events carry hashed customer data.
- Improve. Give budget to the campaigns whose CPA sits under your target and take it from the ones running over. Move each campaign’s optimization event down-funnel as its volume matures, so the Taboola campaign that learned on
start_checkoutgraduates tomake_purchase. Tighten the abandoner window if the audience has gone stale, stretch it if the pool is starving, and kill any advertorial headline that draws clicks but noOutboundClick. - Expand. Every synced event is an audience seed, so cut finer tiers as volume grows:
AddPaymentInfowithoutPurchasebecomes your hottest retargeting ad set, andViewContentwithoutAddToCartgets warmed separately. When the purchaser audience reaches a workable size, seed the 1% lookalike. When the middle of the funnel converts reliably, raise Taboola’s budget to feed it. Each campaign you add generates more events, and every audience in the stack gets sharper for it.
Between your passes, three platforms keep tuning themselves on the signals you wired. At each pass, you decide what they tune toward next. A month of that and the ad, the advertorial, and the checkout stop behaving like three systems and start compounding like one.
Wrapping Up
Here’s what you shipped:
- A customer journey stitched across ad, advertorial, and checkout, with no lost hops.
- Taboola bidding on real purchases; Meta and Google retargeting real intent.
- A TOF, MOF, BOF, and lookalike campaign stack, each stage executable straight from its task card.
- A weekly loop that reads one report and feeds every improvement back into bidding.
Less guessing, more ROAS.
Explore Related Playbooks
Turn Taboola Traffic Into High-Intent Google Ads Audiences
Native traffic is cheap, but most readers leave without buying, and Google never meets them. This playbook streams their on-site intent into GA4 audiences your Google Ads campaigns can retarget across Search, YouTube, and Display.
Build TikTok Custom Audiences from High-Value Buyers
TikTok can find more of your best customers, but only if it knows who they are. This playbook sends full order values server-side, so you can build value-based audiences and lookalikes from the buyers who move your AOV.
Send Shopify Purchases to the Meta Conversions API
Meta reports fewer conversions than Shopify, and your campaigns scale on a number that isn't real. This playbook sends every checkout event server-side, deduplicated by AnyTrack before it reaches Meta.

