GA4 Attribution Model Changed in April 2026 — Here’s What It Did to Your Conversion Data

April 2026 was one of those months where Google quietly shuffled the deck in GA4. Suddenly, the default attribution model was different, the lookback window for acquisition conversions shrank from 90 days to just 30, and two familiar reporting sections were merged into one. All of this happened with barely a whisper of warning. On the surface, nothing looked broken, but for anyone not paying close attention, it was as if the ground shifted beneath their feet without so much as a tremor.

I remember getting a message from a client in May, a bit puzzled, asking why Search conversions had suddenly dropped by 18 percent since April, even though the sales team was still seeing the same steady flow of leads. We hadn’t changed a thing in the campaigns. Budgets, targeting, creative, all untouched. The only thing that had changed was how GA4 was assigning credit behind the scenes. If you weren’t looking for it, you’d swear the campaigns were underperforming, when really, it was the reporting that had changed, not the campaigns.

This is the most operationally dangerous kind of update. One that produces numbers that look like campaign signals when they are actually measurement signals. Before any budget gets cut or any bid strategy gets revised on data that spans the April boundary, it is worth knowing exactly what moved and why.

Where the GA4 Attribution Model Change 2026 Actually Started

The April update recalibrated Google’s data-driven attribution model, meaning any historical comparison of attributed conversions across the April boundary will show discrepancies that reflect the model change, not campaign performance.

It also reset some properties to data-driven attribution even if they had been manually set to another model. An account that had been deliberately running on last-click attribution may have been silently switched to DDA in April with no obvious notification in the interface. That one change was enough to redistribute credit across channels. Channels contributing to upper-funnel interactions started receiving more credit under the updated DDA logic. Last-touch channels received less. A Search campaign that appeared to drop while Display improved in April is worth ruling this out before assuming one channel suddenly became better than the other. This is before drawing any conclusions about either channel.

Why B2B Advertisers Are Seeing Fewer Conversions Without Losing Any Real Buyers

The default attribution lookback window for acquisition conversion events was changed from 90 days to 30 days. Other conversion events kept their 90-day window. Acquisition events did not.

For anyone running campaigns with a longer consideration cycle, this is probably the change most B2B teams will feel. A buyer who first touched a campaign 45 days before converting would have been attributed under the old window. Under the new 30-day default, that same buyer does not appear in the attribution data at all. The conversion happened. The measurement just stopped looking far enough back to see it.

These settings also flow into Google Ads. If GA4 key events are imported into Google Ads, the shortened window changes what Smart Bidding is optimising toward. For B2B accounts where the sales cycle runs longer than a month, restoring the lookback window to 60 or 90 days in GA4 Admin is not a cosmetic fix. It is the difference between data that reflects how buyers actually behave and data that reflects a default Google chose for a different account type.

What Happened to the Attribution Reports You Used to Navigate Separately

The Advertising section of GA4 was reorganised. The Attribution paths report and the Model comparison report were merged into a single Attribution report with tabbed views. Journey analysis, formerly Conversion paths, now defaults to showing the top ten conversion paths only.

Accounts that relied on Conversion paths for multi-touch journey analysis are now looking at a condensed view by default. Paths that used to appear in the top fifteen are not gone, they are just not surfaced without adjusting the report manually. The data hasn’t disappeared. GA4 is just showing you less of it by default.

“The most dangerous thing about this change is that it looks like a performance shift. Campaigns that appear to be underperforming may simply be losing attributed credit under the new model — not actual conversions. We’ve seen accounts where Search campaigns lost 15–20% of attributed conversions in April with nothing else changing. Before you cut budget on a channel that ‘stopped working’, check whether the attribution model changed underneath it.”

— Vishal Singh, Performance Marketing Specialist

What DDA Does to Channel Credit When Conversion Volume Is Low

Data-driven attribution requires at least 400 conversions for the specific key event and 20,000 total conversions across all events within the lookback window to activate properly. If those thresholds are not met, GA4 silently falls back to last-click without flagging it anywhere visible.

For most B2B lead generation accounts, those thresholds are not being hit. The April recalibration applied DDA logic to accounts that did not have the volume to support it, or triggered the silent fallback to last-click without the account team knowing the switch had occurred. Either way, you end up looking at channel swings that have more to do with attribution than with what the campaigns actually did. An account generating 25 qualified leads a week does not have the signal for DDA to produce reliable outputs. If those conversions are feeding Google Ads, Smart Bidding is making decisions from a much noisier picture than it had before April.

The Right One Between Platform-Native Reporting vs an Independent Attribution Layer

Platform-native reporting is where GA4 and Google Ads both apply the model currently active in the account. After a mid-period recalibration, comparing any metric period over period in native reporting without accounting for the model change produces a number that blends performance shifts with measurement shifts. From inside GA4 alone, it’s almost impossible to tell where the performance change ends, and the measurement change begins.

Independent attribution layer is a third-party tool or CRM-based revenue attribution that does not depend on GA4’s model and gives a stable baseline unaffected by platform-side changes. Cross-referencing native reporting against an independent source is the only reliable way to know whether a movement in April data represents something that happened in the campaigns or something that happened in the measurement.

For day-to-day optimisation, GA4 is perfectly fine. But if you’re comparing performance before and after April or making budget decisions from that data, I’d want a second source of truth before making any big calls.

The Questions We Get Asked Every Time This Comes Up