Facebook Ads for a Brand New Shopify Store With No Pixel Data
How to structure your first Meta campaigns on a Shopify store with zero pixel history, from event setup to realistic early budgets.
Written by Mantas Jurgutis — Founder, Adsify — builds the Google & Meta automation merchants use daily
Editorially reviewed by Adsify Editorial on February 5, 2026 — Reviewed against Shopify, Google Ads and Meta official documentation.
Why zero pixel history changes your strategy
A store with no historical pixel data has no signal for Meta's algorithm to match against, no lookalike source, and no proof yet that its checkout converts at any particular rate. This is normal for every store on day one, but it means your first two weeks of spend should be treated as data collection, not scaling. Expect higher CPA than the numbers you'll see once the pixel has 30 days of purchase history — a 2x difference between week 1 and week 5 CPA is common purely due to signal maturity, independent of any creative or targeting change.
Install order matters: pixel, then Conversions API, then test
In Shopify admin, go to Settings > Apps and sales channels, add the Meta and Instagram channel, connect your Business Manager, and it will auto-install both the Meta Pixel (via the storefront) and server-side Conversions API events keyed to the same event IDs, deduplicating automatically per Meta's documentation. Before spending a single euro, use Events Manager > Test Events to fire a test Purchase from a real checkout and confirm it appears with a high Event Match Quality score, ideally above 6.0 out of 10, which typically requires email, phone and external ID to be passing correctly.
Set up Aggregated Event Measurement before launch
With no purchase history, decide your 8 prioritized events in Events Manager > Aggregated Event Measurement before your first campaign goes live: Purchase, Add to Cart, Initiate Checkout, View Content in that order is the standard for ecommerce. Getting this wrong after you've launched campaigns means re-prioritizing events, which can reset some attribution modeling and cost you another few days of clean data collection.
Your first campaign objective
For a store with under roughly 50 total historical purchases (including organic, pre-ads), start with a single Advantage+ shopping campaign rather than a manual catalog sales campaign with multiple ad sets. A brand-new pixel has no audience signal to split against interest groups meaningfully, so manual segmentation just divides an already-thin signal into even thinner slices, each of which individually never reaches the 50-events-per-week threshold to exit learning phase.
Setting a realistic starting budget
Worked example: if your product's AOV is EUR 45 with 40% margin, that's EUR 18 of margin per order. To collect useful signal, plan for at least 15 to 20 purchases in the first week, which at an estimated EUR 25 CPA (typical for a new pixel before optimization) means a daily budget of roughly EUR 55 to EUR 70. If that budget is not sustainable relative to your margin, it is more honest to expect a longer data-collection runway at a lower daily spend than to launch underfunded and pause after three days, which wastes the little signal you did collect.
Using broad targeting as the default
Do not layer detailed interest targeting or exclude broad segments in your first campaign; Meta's own guidance for accounts with limited conversion history recommends the largest reasonable audience so the delivery system has room to find responsive users. For most Shopify stores this means a single country or a small group of similar countries, no interest narrowing, and no minimum age restriction beyond your product's legal requirements.
Creative volume compensates for signal scarcity
With no historical data telling Meta which creative style resonates, upload more creative variety than you would for an established account: at minimum 4 to 6 distinct concepts (not just color variants of the same ad), mixing UGC-style video, a founder or product demo video, and 2 to 3 static image ads with different value propositions. Meta's delivery system will allocate impressions toward whichever performs, but it needs genuine variety to have something meaningful to test against.
What Event Match Quality means for a new store
Event Match Quality (EMQ), visible in Events Manager next to each event, scores how well your Purchase and Add to Cart events can be matched to a real Facebook or Instagram user profile using hashed customer data. A new Shopify store using the standard Meta channel integration typically starts around 5 to 6 out of 10; improving it by ensuring checkout captures email at the start of checkout (not just at payment) and that phone number fields are enabled can lift EMQ toward 7 to 8, directly improving how many of your conversions Meta can attribute and optimize against.
Handling the first week of noisy data
Expect day-to-day CPA in week one to swing widely — a EUR 15 CPA on Tuesday and a EUR 60 CPA on Wednesday from the same campaign is not unusual with under 50 weekly events. Resist pausing or heavily editing the campaign based on any single day; instead track a 3-day rolling average and only make structural changes (creative swap, budget change) after at least 4 to 5 days of data.
Building your first real audience asset
Your first genuinely valuable audience is not a lookalike — you have no seed data yet — it's a website custom audience built from Pixel events (all visitors, Add to Cart, Initiate Checkout) over the trailing 30, 60 and 180 days, created in Business Settings > Audiences > Custom Audience > Website. These become usable for retargeting once you have roughly 100 to 1,000 people in each bucket, but even a starting audience of a few hundred is worth creating from day one since it accrues automatically.
When to introduce a lookalike audience
Once your Purchase custom audience (people who bought via web pixel or Conversions API) reaches at least 100 people, ideally closer to 300 for stability, you can create a 1% lookalike in Business Settings > Audiences > Lookalike. Before that threshold, lookalikes built from too-small seed audiences tend to be unstable and often underperform Meta's own broad or Advantage+ automated targeting for a store this early.
Avoiding the discount-code trap
New stores often launch with an aggressive first-order discount to drive early conversions and prove the pixel works faster. This is a reasonable short-term tactic, but be aware it distorts your true CPA and margin math — a 20% discount changes a EUR 45 AOV at 40% margin (EUR 18 margin) to roughly EUR 36 AOV at a genuinely lower margin, so any CPA target calculated before the discount needs to be revised down accordingly or you will look profitable in Ads Manager while losing money per order.
Setting a checkpoint before scaling
Give the first campaign a defined evaluation checkpoint — 14 days or a spend threshold like 3x your target CPA times 30 conversions, whichever comes first — rather than an open-ended run. At that checkpoint, look at cost per Purchase, EMQ score trend, and whether CPA is trending down week over week, which is the expected pattern as the pixel accumulates signal, before deciding whether to increase budget or diagnose creative and landing page issues.
How Adsify handles the cold-start problem
Adsify is built for exactly this scenario: it launches Advantage+ shopping campaigns directly from a store's existing Shopify catalog on day one, wiring up both the Web Pixel and server-side Conversions API events together so Event Match Quality starts as high as the integration allows rather than needing manual configuration, and it re-evaluates spend and creative signals every 6 hours through the early data-collection period.
Setting expectations with stakeholders
If you're running ads for a client or reporting to a founder, set the expectation explicitly before launch: the first two weeks of a new pixel's data are structurally noisier and more expensive than months three and beyond, independent of skill or budget. Framing this upfront avoids the common mistake of killing a viable account on day 5 because CPA looked 2x worse than a benchmark that assumed a mature pixel.