Your Meta ROAS Is Built on a Default Setting Nobody Told You About

The Number You Watch Every Morning

Every founder running Meta ads has a ROAS number they watch. It's the first thing they check when they log into Ads Manager. It drives the decisions that matter most - which campaigns to scale, which to kill, whether this whole paid acquisition thing is actually working.

That number is calculated using a setting most founders have never touched, probably don't know exists, and almost certainly haven't thought about since their first campaign launched.

It's called the attribution window. And right now, Meta's default is claiming credit for purchases that may have absolutely nothing to do with your ad. Not through any deception - the setting is documented and changeable. But no one walks you through it when you set up your account, and it defaults to generous, not conservative.

Understanding exactly what your ROAS number is counting - and what it's not - is the difference between optimizing against real performance and chasing a metric that flatters your campaigns more than your business deserves.

How Meta's Attribution Window Actually Works

When someone converts after interacting with your ad, Meta needs a rule for deciding whether to credit that ad. Attribution windows set those rules. They answer the question: how long after an ad interaction should we still count a resulting purchase as "caused by" this campaign?

Meta's current default attribution setting is: 7-day click, 1-day view.

That means two separate claims are happening simultaneously:

• If someone clicked your ad and purchased within 7 days, Meta counts it as a conversion from that ad
• If someone saw your ad - just an impression, no click - and purchased within 1 day, Meta also counts it as a conversion from that ad

Both sound plausible in isolation. Both have serious problems in practice. And because these rules are set to "on" by default, most founders have no idea how much of their reported ROAS is genuinely earned versus simply claimed.

The Click Window: Seven Days Is a Long Leash

The 7-day click window is the less controversial of the two defaults, but it still builds in a generous assumption about causality. Here's how that plays out: someone clicked your ad on Monday and bought on Sunday. Six days passed. In that window, they may have visited your website three more times, read two email newsletters, and clicked a Google Shopping ad before finally converting. Meta counts Sunday's purchase as its conversion.

Maybe it was. The original ad click may have been the catalyst that put you on their radar. Or maybe your email sequence closed the deal and Meta is claiming credit for a first touchpoint in a much longer journey. The 7-day window can't distinguish between those two scenarios. It just counts the purchase.

For high-consideration products - furniture, software, anything over $200, anything with a real comparison-shopping phase - the 7-day window inflates ROAS because the real path to purchase involves multiple touchpoints. Those touchpoints exist on your email list, on Google, in organic search. They all played a role. Every platform claims the same sale, and a 7-day click window is how Meta maximizes its claim.

For impulse purchases - low price point, immediate gratification, category where people decide in minutes - the 7-day window is probably overcounting too, but the distortion is smaller. Someone who clicks a $29 impulse buy ad typically doesn't wait five days to complete the purchase.

The View-Through Window: The Most Contested Setting in Paid Media

The 1-day view attribution is where things get genuinely contentious among anyone who has thought hard about paid media measurement.

Meta counts a conversion as "yours" if someone saw your ad - no click, just an impression - and then made a purchase within the next 24 hours. You never drove them to your site. They might have searched for your brand directly, clicked an email, or just opened a new tab from memory. Meta still claims the sale.

This is not Meta acting in bad faith. The logic behind view-through attribution has a real theoretical basis: advertising has always worked partly through exposure. Someone sees a billboard on their commute, doesn't pull over immediately, but searches for the product online later. View-through attribution tries to capture that effect for digital impressions. The problem is that Meta serves ads at scale to almost everyone in your target audience. A buyer who was already going to purchase - a repeat customer in your retargeting pool, someone who just clicked your email ten minutes ago, a person who searched your brand name yesterday - will see your ad in their feed before they convert. That does not mean the ad caused the conversion.

View-through attribution is most defensible for high-frequency, impulse-driven categories where ad exposure and purchase have a short, direct connection - and where you have minimal organic traffic that could explain the conversions independently. It's least defensible for established brands with substantial email lists, SEO traffic, or returning customers who were already in a purchase cycle before they happened to see an ad.

How to Find (and Read) Your Attribution Breakdown

In Meta Ads Manager, attribution settings live at the ad set level, inside the Conversion section when you create or edit an ad set. The field is labeled "Attribution Setting" and shows your current window with an option to change it. Most founders see this once during setup, leave it at the default, and never look again.

To see the breakdown on live campaigns without changing anything:

1. In Ads Manager, go to the Columns dropdown and select Customize Columns
2. Search for and add the attribution breakdown columns: 1-day click, 7-day click, 1-day view, and 7-day view
3. Apply and look at your data across a 30-day window

What you'll see is a breakdown of how many conversions each window type is claiming. The number in your default ROAS column is the total - it adds all of these together. The 1-day click column alone is your most conservative, most credibly causal number.

Matching Your Window to Your Purchase Cycle

The goal is not to find the attribution window that makes your numbers look best. It's to find the one that most accurately reflects how your customers actually buy. There's no universal correct answer - but there's a principled way to arrive at the right one for your business.

Start with your actual data. If you have Shopify or WooCommerce, look at the time-to-purchase distribution in your analytics: how many hours or days typically pass between a first session and a completed order? That number tells you the realistic window where an ad click could plausibly be the initiating event.

A rough framework:

• Sub-$50, impulse category, no real comparison phase: 1-day click captures most genuine conversions. The 7-day window is overcounting.
• $50-$200, some research involved: 7-day click is reasonable. Accept that other channels are participating in those multi-day journeys.
• $200+, or subscription, or B2B: 7-day click but treat it as an upper bound, not a clean attribution. Consider triangulating with MTA or incrementality testing to understand true channel contribution.

For view-through: if you're running conversion-objective campaigns, most practitioners move view-through to zero or keep it at 1-day only as a transparency floor. If you're running awareness campaigns where impression volume is genuinely the mechanism - video views, brand reach - view-through has more logical basis, but keep those campaigns separate so their attribution logic doesn't pollute your direct-response reporting.

One rule that applies regardless of your product: all campaigns you're comparing against each other must use the same attribution window. If one ad set is running 7-day click + 1-day view and another is running 1-day click only, you're comparing different measuring systems. Any conclusion you draw about which campaign is winning is potentially wrong.

Why This Gets Murkier After iOS 14

The attribution complexity doesn't stop at window settings. Apple's App Tracking Transparency framework, rolled out in 2021, broke the pixel's ability to directly observe conversions from iOS users - a significant share of Meta's audience, particularly for consumer products.

Meta responded with Aggregated Event Measurement (AEM) and statistical modeling. For iOS conversions that the pixel cannot directly observe, Meta uses modeled data to estimate what likely happened. These modeled conversions are included in your reported ROAS as if they were observed events.

This is not unique to Meta - every platform with significant iOS exposure does some degree of conversion modeling now. But it means your ROAS number is a blend of directly observed pixel events and statistical estimates. The balance depends on your audience demographic: a product skewing toward Android users has higher direct observation rates than one targeting iPhone-heavy demographics like affluent consumers or younger women.

This is part of why your pixel's Event Match Quality score matters so much. Better EMQ means the pixel can match more observed events to real user profiles before the data needs to be estimated. A high EMQ score reduces the modeling layer, which means a ROAS number that reflects more hard data and fewer probabilistic guesses.

The practical upshot: if your audience skews iOS and your EMQ is low, your ROAS number has two separate inflation sources working simultaneously - a generous attribution window on top of modeled conversion estimates. Neither one is fraud. Both of them make your numbers look better than the underlying reality warrants.

Your costs and your revenue are what they are. Attribution windows and modeled conversions don't change those numbers - they only change what Meta claims credit for. The gap between claimed and real is a business risk you're carrying without knowing it.

What to Do With the Real Numbers

When founders run the attribution breakdown for the first time and see what their ROAS looks like on 1-day click only, the reaction is usually somewhere between surprise and dismay. The number is lower. Sometimes the difference is modest - 10 to 15 percent. Sometimes it's more significant than that, especially for brands with large retargeting pools that generate a lot of view-through claims from audiences already deep in the funnel.

And that is just the attribution window layer. Underneath that number, there is another dimension: your ROAS is also a weighted average across every age group, placement, device, and region in your account. Once you fix the attribution window, the Breakdowns report shows you what's hiding inside that blended number.

The reframe: a lower, more accurate number is categorically better than a higher number that flatters your campaigns. You can make real scaling decisions with an honest number. You cannot build a profitable business by optimizing against a metric that's systematically overcounting your results.

Here's what to actually do:

1. Run the breakdown exercise. Pull 1-day click, 7-day click, and 1-day view as separate columns across your last 30 days. Note the gap between your default ROAS and your 1-day click ROAS. That gap is your "attribution premium" - the amount Meta is adding to your headline number via the view window and multi-day click claims.
2. Pick a window that fits your product. Apply it consistently going forward. Set it at the ad set level before campaigns launch, not retroactively after you check performance.
3. Use the same window across all campaigns you compare. This is the most immediately actionable step. Standardizing the window doesn't change your real performance - it makes your internal comparisons valid.
4. Cross-reference with your Shopify or CRM revenue. If Meta is claiming 400 conversions in a period when Shopify shows 280 total orders, you have a concrete signal that attribution overlap is significant. Understanding where that overlap lives across your channels is the next layer of the problem.

None of this requires changing how your campaigns run. You're not touching budgets, targeting, or creative. You're changing what measuring stick you hold them to. And if you're currently making decisions about campaign objectives, bid strategies, or which ad sets deserve more budget - those decisions get sharper the moment they're based on a number you actually trust.

The point of this exercise is not to demoralize you about Meta's reported numbers. It's to give you a foundation that holds up when the dashboard and the bank account are compared. Build on honest numbers, and every optimization decision you make from here gets more reliable.