Understanding Attribution Windows: Why Meta, Google and Your CRM Never Show the Same Numbers

Meta reports 52 leads, Google Ads 38, your CRM receives 61, and added up, the platforms claim more leads than there were inquiries. Each number is correct on its own; every platform just counts differently. The most important reason is the attribution window.

This guide explains how attribution windows work, which settings Meta, Google Ads, TikTok and LinkedIn use, why the numbers never match your CRM and how to reconcile them properly. It's written for performance marketers, agencies and lead generation businesses.

What an attribution window is

An attribution window defines how long after an interaction with an ad a conversion is still credited to that ad. If the lead happens inside the window, it counts for the ad. If it happens afterwards, it doesn't, even if the ad started the journey.

Platforms distinguish between different kinds of interaction:

• Click: the person clicked the ad and landed on your website, app or lead form. The strongest signal.
• Engage-through or engaged view: the person interacted with the ad without clicking a link, for example a like, a save or a video view of a few seconds.
• View: the person only saw the ad. Conversions from this are called view-through conversions.

Each category has its own window. "7-day click, 1-day view" means a lead counts if it happens up to 7 days after a click or up to 1 day after seeing the ad.

The platforms' windows

As of October 2026, sources: Meta for Business, Google Ads Help, TikTok Ads Help. Options change regularly, so check your settings directly in your account.

A few specifics:

• Meta sets the window per ad set, and it also affects optimization. Since March 2026, only link clicks count as clicks for website conversions. Likes, saves and shares count as engage-through. What this means for your numbers is covered in our article on Meta attribution changes.
• Google Ads sets the window per conversion action. For imported CRM conversions, the click window is especially important because deals often close weeks after the click.
• Every platform only counts its own ads. No platform knows that the same person saw an ad from another network before.

Why platform numbers don't match your CRM

The differences almost always come down to the same five causes.

1. Double counting across platforms

Someone sees an Instagram ad, searches for your brand on Google the next day, clicks a search ad and signs up. Meta counts the lead as a view-through conversion, Google Ads as a click conversion. In the CRM, it exists once. That's why the sum of platform numbers is almost always higher than the number of real leads.

2. Different windows

By default, Google Ads counts 30 days after the click, Meta 7 days. A lead that comes 12 days after the click counts on Google but not on Meta. If you compare the platforms, you're comparing different time spans.

3. Different dates

By default, Google Ads records a conversion on the date of the click, not the date of the lead. Meta assigns it to the date of the ad interaction. Your CRM uses the date the lead was created. A lead from November 3 after a click on October 28 shows up in October in Google Ads and in November in the CRM. In monthly reports, this causes differences, especially early in the month, when later conversions still raise the previous month's numbers.

4. Conversions no platform sees

Leads from referrals, organic search or direct visits don't appear in any ad account. Neither do leads whose click ID gets lost along the way, for example through redirects or iframes. How to keep click IDs reliably is covered in the guide on UTMs and click IDs.

5. Modeled conversions

When users don't allow tracking, Meta and Google partly add estimated conversions to their numbers. These don't exist as individual records in your CRM.

How to reconcile platform numbers

1. Define one source of truth

   For the question "How many leads and customers do we have?", the CRM is the reference, not the ad account. Platform numbers answer a different question: which leads does the platform credit to itself?

2. Align windows

   If you want to compare platforms, use the same click window everywhere, for example 7 days. In Meta, Compare attribution settings helps; in Google Ads, set a conversion window per action.

3. Separate click from view

   Report click conversions separately from view-through and engage-through conversions. Only clicks can be compared directly with website tracking and the CRM.

4. Attribute leads via your own website

   Capture UTMs and click IDs on every visit and attach them to the lead. That way, each lead is credited to exactly one campaign, using first-click attribution or last-click attribution.

5. Report by lead date

   For reports to management or clients, use the date the lead was created. Platform data by click date is better suited to optimizing individual campaigns.

6. Send real conversions back

   Send leads and deals from the CRM to the platforms via the Conversion API. Then the platform optimizes for real customers, not just the conversions it sees itself.

An example from lead generation

An agency manages a real estate agent with a monthly budget of €4,000, split between Meta and Google Ads.

Meta Ads (7-day click, 1-day engage, 1-day view):  52 leads
Google Ads (30-day click):                         38 leads
Platform total:                                    90 leads
Leads in CRM:                                      61 leads

With your own tracking, you see: 31 leads came via a click on Meta, 22 via a click on Google Ads, 8 via organic search and direct visits. The remaining 21 Meta leads were view-through and engage-through conversions that were credited to other channels, mostly Google search. Google Ads counts more than 22 because it uses 30 days and also claims leads that last came through a Meta ad but had clicked a search ad at some point before.

Cost per lead according to Meta  = €2,400 ÷ 52 ≈ €46
Cost per lead by click           = €2,400 ÷ 31 ≈ €77

Which number is right? Both, they just answer different questions. Meta says how many leads the campaign was somehow involved in. Click attribution says how many leads came directly through the ad. For budget decisions, the second number is the more cautious one. Whether ads work without a click can only be proven properly with control group tests, such as Meta's Conversion Lift studies.

Choosing a window that fits your customer journey

No window is right for everyone. What matters is how long your prospects typically take between click and lead:

• Short journey (webinar sign-up, lead magnet, free consultation): most leads come the same or next day. 7-day click works well.
• Medium journey (coaching, trades, insurance): leads often come within two weeks. 7-day click captures most of them, some fall outside.
• Long journey (real estate, B2B software, high-ticket services): weeks often pass between first click and inquiry. A short window underestimates campaigns that work early in the journey.

You can't see how long your journey really is in the ad account, only by connecting the first visit to the later lead. That requires tracking that recognizes visits over days or weeks.

Common mistakes

• Adding up platform numbers: the total is almost always higher than the number of real leads.
• Comparing platforms with different windows: 30-day Google against 7-day Meta isn't a fair comparison.
• Reading view conversions as clicks: it makes your cost per lead look better than it is.
• Closing monthly reports early in the month: Google Ads still adds later conversions to the click date retroactively.
• Windows shorter than the journey: campaigns that work early look worse than they are.
• Only looking at forms: a lead isn't a customer. How to measure all the way to revenue is covered in the guide to real ROAS in lead generation.

How LeadMetrics helps

LeadMetrics counts every lead once, from the perspective of your website and CRM:

• Your own attribution: lead tracking captures the traffic source on every visit via UTMs, click IDs and its own ad parameters. Every lead is credited to exactly one campaign and ad by first click or last click.
• Long journeys: leads are matched to their session via session ID, email or phone number up to 90 days back. Sessions last 90 days with consent and 1 day without.
• Journey per lead: the lead view shows the journey timeline and click history, so you see when the person came through which ad.
• All platforms in one dashboard: costs and metrics from Meta, Google Ads, TikTok, LinkedIn and Microsoft Ads (beta) next to your real leads and deals in dashboards.
• Conversion API: leads and won deals are sent back server-side to Meta, Google Ads, TikTok and LinkedIn.

LeadMetrics doesn't see pure view or engage-through conversions, because no website tracking can. In return, you get a way of counting that doesn't change when a platform changes its rules.