A marketing director pulls up three dashboards on Monday morning – Google Ads, Meta Ads Manager, and HubSpot – and each one claims credit for a different set of conversions from the same campaign. This is the attribution window problem, and it’s probably the single most common reason ad platforms and CRM data never seem to agree, no matter how carefully the tracking was set up.
An attribution window is the timeframe a platform uses to decide whether a conversion counts as coming from a specific ad interaction. Google Ads might use a 30-day click / 1-day view window by default. Meta defaults to 7-day click / 1-day view. HubSpot, depending on how it’s configured, might use last-touch attribution with no time limit at all, or a 90-day lookback on its own contact timeline. Three systems, three different rulebooks, and each one is technically “correct” by its own definition.
Why attribution windows cause reporting conflicts
Each platform can only see its own slice of the customer journey. Meta doesn’t know a user later clicked a Google ad three days after seeing an Instagram Story. Google doesn’t know that same user also opened four nurture emails from HubSpot before filling out a form. So each platform attributes the conversion to itself, using whatever touchpoint falls inside its own window – and none of them are lying, they just have blinders on.
A SaaS company running a $12,000/month Meta budget alongside $18,000/month on Google Ads will frequently see this: Meta reports 340 conversions for the month, Google reports 290, and the CRM shows 410 total closed-won opportunities tagged to paid channels. Add those platform numbers together and you get 630 – nearly 50% more than what actually happened. That gap isn’t fraud or a tracking bug in most cases. It’s duplicate attribution from overlapping windows counting the same buyer twice, sometimes three times.
The default windows on major platforms
Google Ads defaults to a 30-day click-through and 1-day view-through window, though this is adjustable per conversion action in the “Attribution settings” panel inside Google Ads (not GA4, a common confusion point). Meta Ads Manager moved to 7-day click / 1-day view as its default in 2021 after Apple’s iOS 14.5 update forced platforms to shrink windows and rely more on modeled data. LinkedIn Campaign Manager uses a flat 30-day window with no view-through option for most objectives. TikTok Ads defaults to 7-day click / 1-day view, matching Meta’s post-iOS14 posture.
None of these were designed to work together. They were each optimized to make the individual platform’s ROAS look defensible, not to produce a unified truth across a media mix.
How to reconcile the numbers without losing your mind
A revenue operations lead trying to reconcile platform-reported conversions with actual pipeline typically starts by picking one system of record – usually the CRM, since that’s where a deal either closes or doesn’t – and treats every platform’s self-reported number as directional, not final.
From there, the practical fix isn’t to force all platforms onto identical windows (some won’t allow it, and shorter windows understate long sales cycles anyway). It’s to build a first-touch or multi-touch model inside the CRM or a dedicated attribution layer like HubSpot’s multi-touch revenue attribution, Triple Whale, or Northbeam, and let platform dashboards answer a narrower question: “is this specific channel’s spend trending toward efficiency,” not “how many total conversions did we get.”
A few steps that actually move the needle:
1. Standardize on UTM parameters before touching attribution windows. If utm_source and utm_campaign aren’t consistent across every ad platform and landing page, no attribution model – long window or short – will produce clean data.
2. Extend the CRM lookback period to match your real sales cycle. A B2B company with a 60-day average sales cycle using a 30-day attribution window in Google Ads will systematically undercount influence from early-funnel content and top-of-funnel ads.
3. Separate “last-click efficiency” from “total pipeline influence.” Use platform-native reporting for day-to-day bid optimization, and use CRM-based multi-touch data for budget allocation decisions made monthly or quarterly.
4. Audit for duplicate conversion counting quarterly. Pull closed-won deals from the CRM and manually trace which ones show up as conversions in more than one ad platform for the same date range.
The myth worth retiring
A persistent misconception is that switching every platform to the same attribution window – say, forcing everything to 30-day click – will make the numbers match. It won’t. Even with identical windows, Meta’s algorithm still attributes based on its own pixel data and modeled conversions (increasingly common since Apple’s App Tracking Transparency rollout in April 2021), while Google attributes based on Google’s own signals. The underlying data sources remain disconnected even when the time window is aligned. The fix is structural – a shared source of truth outside the ad platforms – not cosmetic.
Common mistakes in attribution reporting
The most frequent error is reporting platform-native conversion counts directly to leadership without reconciling against CRM pipeline, which inflates perceived ROAS and eventually erodes trust when a board asks why “600 conversions” only produced 80 actual customers. A second mistake is changing attribution windows mid-quarter to make a struggling channel look better – it creates a discontinuity that makes trend analysis useless for months afterward. A third is ignoring view-through conversions entirely because they feel “soft,” when in reality a 1-day view-through on a $40 CPA campaign can represent a meaningful share of assisted conversions, particularly for brand-awareness-heavy channels like connected TV or YouTube. Teams evaluating channel performance sometimes lean too heavily on top-line conversion counts instead of examining the metrics that actually separate a profitable campaign from one quietly bleeding budget – something covered in more depth in The Hidden Metrics That Separate Profitable Ads From Losers.
FAQ
Why does my Google Ads conversion count not match my CRM’s closed-won deals?
Google Ads counts a conversion the moment a form is submitted or a defined action happens within its attribution window, while the CRM only marks a deal closed-won after the full sales cycle completes – which can be weeks or months later and may not happen at all if the deal falls through.
Should I use first-click, last-click, or multi-touch attribution?
For sales cycles under 14 days with a single dominant channel, last-click is often good enough. For B2B companies with multiple touchpoints and cycles over 30 days, multi-touch or data-driven attribution inside the CRM gives a far more accurate picture of what’s actually influencing revenue.
Can I make attribution windows match exactly across platforms?
Not fully. You can align the time window (e.g., 30-day click on both Google and Meta), but the underlying data collection methods still differ, especially post-iOS 14.5, so some discrepancy will always remain – the goal is minimizing it, not eliminating it.
Reconciling attribution data isn’t a one-time setup task – it’s a recurring discipline. The teams that stay sane are the ones who stop expecting platform dashboards to agree and instead build one central view, refreshed monthly, that treats CRM pipeline as the final word.