Late Attribution in Meta Ads: Why Your ROAS Looks Worse on Day 1

The Disappearing Conversions Problem

You launch a campaign on Monday. By Tuesday morning, the ROAS looks terrible — half of what you expected. You pause it. On Friday, you check the account and realise the Monday campaign's ROAS has now climbed to your target, as conversions trickled in throughout the week.

You paused a winning campaign because you didn't account for late attribution.

This is one of the most expensive mistakes in Meta Ads, and it happens constantly because Ads Manager defaults to showing you today's data — not the full picture of conversions that will be credited to yesterday's spend.

What Attribution Windows Actually Mean

An attribution window defines the time period after an ad interaction during which a conversion is credited to that ad. Meta offers four main windows:

• 1-day click: A conversion is credited if it happens within 24 hours of someone clicking your ad.
• 7-day click: Credited if the conversion happens within 7 days of a click. This is Meta's default.
• 1-day view: Credited if someone saw (but didn't click) your ad, then converted within 24 hours.
• 7-day click + 1-day view: Credits both click-through and view-through conversions within their respective windows. This is the most commonly used combined setting.

The window you choose determines which conversions Ads Manager assigns to your campaigns. A sale that happens 5 days after someone clicks your ad counts in a 7-day click window, but not in a 1-day click window.

Why Day 1 ROAS Is Almost Always Understated

If you're running on a 7-day click window and you look at yesterday's campaign data today, you're seeing only a fraction of the conversions that will eventually be credited to that spend. The remaining conversions are still happening — users who clicked yesterday are still considering the purchase, still coming back to the site, still converting over the next 6 days.

This creates a predictable pattern: ROAS for any given day's spend rises throughout the week following that spend, then plateaus around day 7–8. If you evaluate a campaign on day 1, you're reading an incomplete book. You're making a "pause or scale" decision on partial data.

The magnitude of this lag varies by product:

• Impulse purchases under £30: Most conversions happen within 24–48 hours. Day 1 ROAS is fairly representative.
• Considered purchases £50–£200: Purchase cycles of 3–5 days are common. Day 1 ROAS understates true ROAS by 30–60%.
• High-ticket or subscription products £200+: Consideration cycles of 7–14+ days. Day 1 data is nearly meaningless as a decision signal.

The Attribution Window Comparison Trap

A subtler and more damaging error is comparing two campaigns that are running on different attribution windows.

If Campaign A is on 7-day click and Campaign B is on 1-day click, Campaign A will almost always report higher ROAS — not because it's performing better, but because it's crediting more conversions to itself. This makes Campaign A look like a winner and Campaign B look like a loser. You scale the wrong one.

Always standardise attribution windows before making any cross-campaign comparisons. In Ads Manager, use the "Attribution Setting" column to confirm each campaign's window. If they differ, you cannot compare them meaningfully.

The View-Through Attribution Controversy

View-through attribution (VTA) is the most contentious part of Meta's attribution model. When you include the 1-day view window, Meta credits a conversion to your ad if someone saw it without clicking, then purchased within 24 hours.

The concern: many of those view-through converters would have bought anyway. Someone who sees your brand awareness ad in the morning and buys via direct search in the afternoon gets attributed to the ad — but would they have bought without seeing it? In many cases, yes.

For direct-response conversion campaigns, most practitioners exclude view-through attribution (or use 1-day view at most) to avoid inflated attribution. For brand awareness objectives where you're deliberately trying to drive delayed consideration, view-through is more defensible.

Meta's Attribution Comparison tool (under Ads Reporting) shows you the same campaign data under multiple attribution window settings simultaneously. Use it to understand how much of your reported CPA is coming from each attribution type, and decide which window is most honest for your business model.

How to Choose the Right Attribution Window

Match the window to your typical purchase cycle:

• Same-day or next-day purchase products (food, low-ticket consumables): 1-day click is sufficient and gives you the cleanest data for fast optimisation.
• Most e-commerce (3–7 day consideration): 7-day click without view-through is the standard. It captures the majority of real purchase intent while avoiding view-through inflation.
• High-consideration or subscription products: 7-day click + 1-day view, but weight the view-through conversions sceptically in your analysis.

Once you've chosen a window, don't change it unless you have a strong reason. Changing the attribution window on a live campaign changes what it optimises for at the algorithm level, effectively resetting your learning phase.

Monitoring Across Attribution Windows

Keeping attribution windows consistent across all active campaigns — and catching when a new campaign accidentally launches with a different window — is the kind of hygiene detail that's easy to miss when you're managing a complex account.

AdEvolver monitors your account for structural anomalies, including attribution mismatches between campaigns, so you don't inadvertently compare incompatible ROAS figures and make the wrong scaling decision.

Get a free ad account audit with AdEvolver.