GA4 changed how attribution works, and most lead generation businesses never updated how they think about it. Universal Analytics defaulted to last-click. GA4 defaults to data-driven attribution — a model that distributes conversion credit across multiple touchpoints using machine learning. For businesses that run Google Ads and care about which channels actually drive leads, this shift has real implications for what the reports say and which channels look productive.
This guide covers the GA4 attribution models, which one fits different types of lead generation businesses, and how to configure your setup so the data you act on is actually telling you something useful.
The GA4 Attribution Models
GA4 offers several attribution models. Here is what each one does and when it makes sense:
Data-Driven Attribution (GA4 Default)
Data-driven attribution (DDA) uses machine learning to assign credit to touchpoints based on their actual contribution to conversions, not a fixed rule. If organic search tends to introduce visitors who later convert after seeing a paid ad, DDA gives organic some credit for those conversions rather than awarding everything to the paid click.
In theory, DDA is the most accurate model. In practice, it requires volume — Google recommends at least 400 conversions over 30 days per conversion action for DDA to work well. For low-volume lead generation businesses (fewer than 400 form submissions or calls per month), DDA falls back to last-click anyway. If your lead volume is below this threshold, DDA looks sophisticated but behaves like last-click.
Last Click
All conversion credit goes to the final touchpoint before the conversion. Simple, auditable, predictable. The problem with last-click is that it systematically undercredits awareness channels. A prospect who found you through organic search, visited twice through direct, then converted after clicking a branded paid ad shows the branded paid ad driving 100% of conversions — even if they never would have heard of you without the organic post that introduced them.
For businesses that run single-channel campaigns or that just want a clear, consistent number for Google Ads performance, last-click works. For businesses evaluating multi-channel mix, it distorts the picture.
First Click
All credit goes to the first touchpoint. The mirror image of last-click, first-click is useful for understanding which channels introduce new prospects. It overcredits awareness channels and undercredits conversion-driving channels. Rarely used as a primary model; sometimes useful as a second lens when evaluating which channels are generating top-of-funnel reach.
Linear
Credit is distributed equally across every touchpoint in the conversion path. A prospect with four touchpoints gives each one 25% of the conversion. Linear is easy to explain and avoids extreme crediting at either end of the funnel. It also treats all touchpoints as equally valuable, which is rarely true — a brand awareness display impression at day one is not the same contribution as a branded search click on the day of conversion.
Position-Based (40-20-40)
40% of credit goes to the first touchpoint, 40% to the last, and the remaining 20% is divided evenly across any middle touchpoints. This model reflects the belief that both introduction (first touch) and conversion (last touch) are the most important moments, with middle touchpoints playing a supporting role. A reasonable compromise for businesses that want to value both acquisition channels and conversion channels.
Time Decay
Touchpoints closer in time to the conversion receive more credit. A touchpoint that happened yesterday gets more credit than one that happened two weeks ago. Time decay makes intuitive sense for short sales cycles — if a prospect researched your service and converted within 48 hours, the recent touchpoints genuinely drove the decision. For long B2B sales cycles where an initial blog post read six months ago actually started the relationship, time decay dramatically undercredits early-stage content.
Which Model Fits Your Business
High-volume lead gen with Google Ads as primary channel
If you run Google Ads and generate more than 400 conversions per month, data-driven attribution is worth using for campaign optimization. The model’s ability to learn from actual conversion paths and adjust bid optimization signals accordingly produces better bidding than last-click over time. Use DDA in GA4 and configure Google Ads to import GA4 conversions or use Google Ads’ own DDA model (which has access to more signal from the Google ecosystem).
Low-volume lead gen with Google Ads
If you generate fewer than 400 conversions per month, data-driven falls back to last-click. Use last-click explicitly so you know what you are reporting against. The honest answer for many local service businesses is that last-click is the practical model — volume is too low for machine learning to improve on a simple rule.
Multi-channel mix with SEO + paid
If organic search and paid search both run, and you want to understand how they interact, position-based (40-20-40) or linear give you a better picture than last-click. SEO regularly introduces prospects who convert through paid; last-click attribution makes paid look like the hero and SEO look unproductive. This is why SEO budgets get cut — the model does not credit introduction. A position-based model shows both channels contributing to the same conversion path.
Long sales cycle B2B
First-click or linear models work better than last-click or time decay for long B2B sales cycles. If a prospect reads your content in month one and converts in month six, last-click gives all credit to the bottom-of-funnel action that closed them. First-click or linear preserves credit for the content that started the relationship. For B2B lead generation where the buying journey spans multiple months, first-click is often the most honest lens for evaluating which channels generate new pipeline.
Configuring Attribution in GA4
GA4 lets you set attribution at the property level. Here is where to find and adjust it:
- Go to GA4 Admin > Property Settings > Attribution Settings
- Under “Reporting attribution model,” choose the model you want applied to GA4 conversion reports
- Under “Lookback window,” set how far back in the conversion path GA4 should credit touchpoints (default is 30 days for acquisition channels, 3 days for engagement; for longer sales cycles, extend the acquisition window to 60 or 90 days)
One important nuance: the attribution model you set in GA4 affects GA4 reports only. Google Ads uses its own attribution model for bidding optimization, set separately within the Google Ads account under Measurement > Attribution. If you import GA4 conversions into Google Ads, the Google Ads model applies for bidding purposes regardless of your GA4 setting. Keep both aligned if you want reporting and bidding to reflect the same logic.
The Lookback Window Problem
Attribution settings that get changed without changing the lookback window produce misleading data. The lookback window controls how far back from a conversion GA4 will look for touchpoints to credit. The default 30-day window is too short for service businesses with longer consideration periods.
A homeowner searching for a roofing contractor might research for 45 to 60 days before contacting anyone. A business owner evaluating a marketing agency might consume content for 90 days before submitting a contact form. If the lookback window is 30 days, any touchpoint from day 31 onward gets no credit — and the channels that generate initial awareness (organic blog posts, display campaigns, referral traffic) systematically disappear from the attribution picture.
For most lead generation businesses, a 60-day lookback window is more accurate than the default. For B2B businesses with long sales cycles, 90 days is appropriate. Set this in the same Attribution Settings panel in GA4 Admin.
Google Ads Attribution vs GA4 Attribution
One source of confusion in lead generation attribution is the difference between what Google Ads reports and what GA4 reports. They count conversions differently and can show very different numbers for the same campaigns:
- Google Ads counts a conversion when a conversion action fires, attributed to the ad click that preceded it. A conversion action fires once, but can be counted multiple times if a user converts multiple times within the lookback window.
- GA4 counts sessions and users, and reports conversions based on your configured attribution model applied to GA4-tracked events.
The result: Google Ads reports and GA4 reports almost never match on conversion count, even for the same time period. This is not a tracking error. It is the natural result of two systems counting different things with different methodologies. Using both as directional signals rather than precise counts, and not trying to reconcile them to the same number, is the right frame.
What Good Attribution Actually Looks Like
No single attribution model tells the complete story. The businesses that make good channel decisions do not pick one model and treat it as truth — they use multiple views:
- Last-click for Google Ads optimization (it is what bidding algorithms most reliably use)
- First-click or linear for understanding which channels generate new pipeline
- Offline conversion data (deal-level revenue from the CRM) uploaded back into Google and Meta to close the loop between lead attribution and actual closed revenue
The attribution model question matters most at the top — what do you tell Google Ads to optimize toward? For most lead generation businesses running Google Ads, the honest answer is: use last-click if you are below 400 conversions per month, and use data-driven if you are above it. Set the lookback window to 60 days. Import CRM-verified conversions as offline conversions when you can. Everything else is secondary to having clean, complete conversion data flowing into the system in the first place.
For a complete guide to getting attribution data from your website into Google Ads and your CRM without custom development: Sales Provenance handles UTM capture, form-to-CRM routing, and offline conversion upload as an integrated layer. Related reading: Google Ads conversion tracking for WordPress, UTM parameter tracking on WordPress, and closed-loop reporting for WordPress.