Attribution Windows: Why Meta and GA4 Disagree

Every month, the same meeting plays out in Malaysian businesses. The agency shows a Meta Ads Manager report with 120 purchases. The owner opens Google Analytics 4 and finds 55 purchases from paid social. Someone asks which number is real, and nobody has a clear answer.

The gap is not usually a tracking fault or an agency hiding something. It comes from attribution windows: the rules each platform uses to decide which ad or visit gets credit for a sale, and for how long after the interaction. Meta and GA4 were built to answer different questions, so their windows, models and even their calendars differ.

This guide from IZI Digital Marketing explains each window, sizes a normal gap and helps you decide which number to use for which decision. It supports our guide to choosing a Facebook ads agency in Malaysia, because how an agency explains this gap tells you a lot about its reporting. The video below covers the basics.

PART 1 · DIAGNOSE

What Is an Attribution Window?

IN BRIEFAn attribution window is the time limit a platform uses to link a conversion back to an ad or visit. If a customer buys inside the window, the ad gets credit; outside it, the ad gets nothing. Windows change what counts as a result, so read them before you judge any of the Meta Ads metrics that actually matter.

Think of the window as a stopwatch that starts when someone interacts with your ad. Three things decide what the stopwatch records:

• The type of interaction. A link click, a like or save, or simply seeing the ad. Each can have its own window.
• The length of the window. One day, seven days or up to 90 days, depending on the platform and setting.
• The rule for sharing credit. One touchpoint takes all the credit, or several channels share it.

Meta’s guide to attribution models and settings explains that the window tells the model how far back to look for touchpoints. GA4 calls the same idea a key event lookback window. Same concept, different defaults, and that is where the disagreement starts.

BENCHMARK BRIEFING 1 OF 4

Meta vs GA4 Attribution Settings, Side by Side

IN BRIEFMeta’s default counts 7-day link clicks, 1-day engagements and 1-day views, all credited to Meta. GA4’s default looks back 90 days, sees only website visits and shares credit across channels with a data-driven model. Before comparing them, confirm your Meta Pixel and Conversions API setup is sound.

The three highlighted rows explain most of the gap. GA4 cannot see a view or a like, because nothing reaches your website until someone clicks a link. Google’s attribution settings page confirms the 90-day and 30-day lookback defaults, and notes that first click, linear, time decay and position-based models were removed in November 2023. The last row matters too: a sale on Monday from a Friday click sits in different weeks in each tool.

PART 2 · DIAGNOSE

Why Does Meta Report More Conversions Than GA4?

IN BRIEFMeta usually reports more because it counts views and engagements GA4 cannot see, and keeps full credit when a buyer later returns through Google search or a direct visit. Tracking gaps on GA4’s side widen it further. Our guide to GA4 reporting metrics that matter shows where to look.

When the numbers differ, check these six causes before you blame anyone:

• View-through conversions. Someone saw your ad, did not click, and bought within a day. Meta counts it; GA4 has no record of the ad.
• Engage-through conversions. A like, save or short video view followed by a purchase within a day. Again, invisible to GA4.
• Shared credit in GA4. A buyer clicks your Meta ad, leaves, then returns through Google search to buy. GA4 splits or moves the credit; Meta keeps it all.
• Missing or broken UTM tags. Without them, GA4 may file Meta traffic as referral or unassigned instead of paid social.
• Consent and blockers. Cookie refusals, ad blockers and in-app browsers can stop GA4 from recording the session at all.
• Cross-device journeys. A click on a phone and a purchase on a laptop look like two strangers to GA4 unless the user is signed in.

BENCHMARK BRIEFING 2 OF 4

How Big Is a Normal Gap Between Meta and GA4?

IN BRIEFIn our illustrative model, for every 100 purchases Meta reports under its default setting, GA4 credits roughly 46 to paid social. The rest is explained by views, engagements, shared credit and tracking loss. If your gap is far wider, audit the tracking; our guide to conversion tracking with GA4 covers the basics.

Read the chart as a checklist, not a target. Only the bottom three rows are problems you can fix; the others are the platforms working as designed. A low-priced impulse product usually shows more view-through credit, while a considered purchase shows more credit moving to Google search. If you run lead forms inside Meta, GA4 sees even less, which is one of the trade-offs in our comparison of Facebook lead ads vs website conversions.

PART 3 · DESIGN

Which Number Should You Trust for Each Decision?

IN BRIEFUse Meta’s numbers to compare ads inside Meta, GA4 to compare channels against each other, and your own sales records to judge whether marketing pays. Mixing them in one report causes most arguments. Our guide to what to expect from agency marketing reports shows how to lay this out.

The question is not “which tool is right” but “right for what”. Each source is best at one job, and poor at the others:

This split also stops a common trap. Adding Meta’s and Google’s reported conversions together double counts any sale both platforms claim. If the total is higher than your real orders, that is why. Our breakdown of Google Ads vs Meta Ads for cheaper leads uses the same one-source-per-question approach.

BENCHMARK BRIEFING 3 OF 4

How Have Meta and GA4 Attribution Rules Changed?

IN BRIEFFour rule changes since 2023 reshaped the gap. GA4 dropped four attribution models. Meta added engagement credit, removed longer view windows from its API and now counts only link clicks as clicks. Year-on-year comparisons across these dates mislead, a point worth raising when you audit a Meta ads account.

The March 2026 change matters most for owners. Before it, a like followed by a purchase six days later counted as a click-through sale. Now that sale only counts if it happens within a day, under engage-through. Many accounts saw click-through results drop with no real change in sales. If your agency’s report shows a sudden dip around March 2026, ask whether it is a rule change before anyone changes the campaign.

BENCHMARK BRIEFING 4 OF 4

How Does the Window Setting Change Reported Results?

IN BRIEFIn our model, the same campaign reports anywhere from 58 to 100 purchases depending only on Meta’s window setting, and 46 in GA4. Lead campaigns show a smaller spread. Knowing this range stops you rewarding or firing an agency for a settings choice, a point to settle when you keep ownership of your Meta Business Manager.

The online store shows a wider spread because shoppers browse, leave and come back. Leads tend to convert faster, so short windows capture more of them. Meta’s attribution setting also changes how ads are delivered, not just how they are reported, so switching mid-campaign is not a free reporting tweak. Settings choices also interact with creative format; our guide to video vs static ads on Meta in Malaysia covers why video tends to earn more engage-through credit.

PART 4 · DEPLOY

How to Reconcile Meta and GA4 Reports Step by Step

IN BRIEFYou cannot make Meta and GA4 match, but you can make the gap stable and explainable. Tag every ad, check events, align dates and time zones, then track the ratio weekly. If GA4 itself is shaky, weigh a GA4 setup service vs doing it yourself first.

Work through these steps in order. Each removes one source of noise:

1. Tag every ad with UTM parameters. Use a consistent source (facebook or instagram) and a paid medium such as paid_social or cpc, so GA4 files the traffic under paid social.
2. Check pixel and Conversions API events. Confirm purchases or leads fire once, with deduplication between browser and server events.
3. Use Compare Attribution Settings in Ads Manager. See how many results come from click, engage-through and view without changing live settings.
4. Align date ranges and time zones. Check both tools use Malaysia time, and compare weeks or months, not single days.
5. Track the ratio, not the totals. Record GA4 paid social conversions divided by Meta conversions each week. A steady ratio means tracking is healthy.
6. Tie back to real orders. Compare both against your order system or CRM monthly, and investigate if the ratio suddenly shifts.

Step 5 is the one most teams skip. A ratio that holds at around 0.45 for months tells you more than any single report. When it jumps to 0.2, something broke: a missing UTM on a new campaign, a checkout change or a consent banner update. If lead quality rather than volume is the worry, pair this with our guide to fixing junk leads from ads.

PART 5 · DRIVE

What to Ask an Agency About Attribution

IN BRIEFA capable agency names its attribution setting, explains the gap with GA4 cause by cause and reports against your real sales. If it only shows Meta’s highest number, be cautious. Use these questions when you compare any Facebook ads agency in Malaysia.

Ask these before signing, and listen for specific answers:

• Which attribution setting will you report on, and why? A good answer ties the window to how long your customers take to decide.
• Will you show click, engage-through and view results separately? A single blended total hides how much credit rests on views.
• How will you explain the gap with GA4? Look for a cause-by-cause answer, not “Meta is more accurate”.
• What will you compare against our real orders or leads? The best agencies ask for your CRM or order data from week one.
• Who owns the pixel, ad account and GA4 property? You should, with the agency given partner or user access.

Put these questions into a written digital marketing RFP so every agency answers the same way. Online stores should also ask how catalogue reporting works, since Meta catalogue ads depend on the same pixel events.

CONCLUSION

Attribution Windows: What to Decide Next

Meta and GA4 will not agree, and they are not meant to. Meta counts every sale it touched within its windows, including views and engagements. GA4 counts website visits and shares credit across channels over a longer lookback. Since March 2026, Meta’s clicks are closer to GA4’s, but views and engagements still keep the two apart.

Your next step depends on where you stand. If the gap is stable, agree one source of truth per decision and move on. If it swings from week to week, check UTMs, events and consent settings first. For how we set up Meta measurement inside full campaigns, see our Meta ads services.

FAQ

Frequently Asked Questions

1. What are attribution windows?

They are time limits for giving an ad credit. It depends on the platform, but a conversion only counts if it happens within a set number of days after someone clicks, engages with or views the ad.

2. What is Meta’s default attribution window?

Seven-day click, one-day engage-through and one-day view. It depends on your campaign goal, but this is the default for ad sets optimising for website conversions since Meta’s March 2026 update.

3. Why does GA4 show fewer conversions than Meta?

Because GA4 cannot see views or engagements. It depends on your tracking too, but GA4 also shares credit with other channels and loses some sessions to missing UTMs, consent refusals and ad blockers.

4. Can I make Meta and GA4 numbers match?

No, and you should not try. It depends on what you need, but the realistic goal is a stable, explainable gap, checked monthly against your real orders or leads.

5. Should I change Meta’s attribution setting to 1-day click?

Only for quick decisions. It depends on your sales cycle; free offers and simple leads suit 1-day click, while considered purchases usually suit the 7-day default. The setting also changes delivery, not just reporting.

6. What is the GA4 lookback window?

Ninety days by default for most key events. It depends on the event type; first visits use 30 days, and you can shorten other key events to 30 or 60 days in GA4’s attribution settings.