GA4 has powerful attribution built right in — multi-touch models, conversion path reports, adjustable lookback windows. But most people never configure it. They leave the defaults untouched and wonder why their channel data looks wrong. This guide walks you through setting up attribution in GA4 properly, step by step.

I’ve configured attribution settings across 150+ GA4 properties. The pattern is always the same: teams either don’t know these settings exist, or they change one thing and miss three others. Five steps, done correctly, and you’ll have attribution data you can actually trust for budget decisions.

Understanding GA4’s Attribution Settings

Before touching anything, you need to understand what GA4 is doing behind the scenes. Attribution is how GA4 assigns credit for conversions to the channels, campaigns, and touchpoints that drove them.

GA4 handles attribution differently from Universal Analytics. There’s no separate “Attribution” product — it’s baked into the platform. Three things control how attribution works in your property:

• Attribution model — the algorithm that distributes credit across touchpoints
• Lookback windows — how far back GA4 looks to find touchpoints that contributed to a conversion
• Key events (conversions) — which actions GA4 actually attributes credit for

Change any one of these and your reports shift. Change all three deliberately, and you get a clear picture of what’s actually driving results. Let’s configure each one.

Step 1 — Choose Your Attribution Model

GA4 used to offer multiple attribution models — linear, time decay, position-based. Google retired those in late 2023. Now you have two options:

• Data-driven attribution (DDA) — Google’s machine learning model that distributes credit based on actual conversion patterns in your data
• Last click — 100% of credit goes to the final touchpoint before conversion

Here’s how to set it:

1. Open your GA4 property
2. Go to Admin → Attribution settings (under Data display)
3. Under Reporting attribution model, select your preferred model
4. Click Save

Which one should you pick? Data-driven attribution is the right choice for most properties. It uses your actual data to determine which channels contribute most to conversions. Last click is simpler but gives a distorted picture — it ignores every touchpoint except the final one.

Data-driven attribution needs sufficient conversion volume to work properly. If your property gets fewer than 300 conversions per month, you might see DDA behave similarly to last click because there isn’t enough data for the algorithm to learn patterns.

One critical detail: changing the attribution model is retroactive. It recalculates historical data in your reports. So if you switch from last click to data-driven, your past reports will change too. This is actually helpful — you get consistent reporting without a sudden break in data.

For a deeper look at how different models compare and when to use each one, read my complete guide to attribution modeling.

Step 2 — Set Lookback Windows

Lookback windows define how far back GA4 searches for touchpoints when assigning attribution credit. If a user clicked your ad 45 days ago and converted today, does that ad get credit? Depends on your lookback window.

GA4 has two separate lookback windows, and this is where people get confused:

Window Type	What It Covers	Options	Default
Acquisition conversion events	First-time user conversions (first_open, first_visit)	30 days	30 days
All other conversion events	Every conversion after the first	30, 60, or 90 days	90 days

To configure lookback windows:

1. Go to Admin → Attribution settings
2. Under Lookback window, you’ll see both settings
3. Adjust the “All other conversion events” window based on your sales cycle
4. Click Save

How to choose the right window:

• Short sales cycle (e-commerce, impulse purchases): 30 days is usually enough
• Medium sales cycle (SaaS, lead gen): 60 days captures the research phase
• Long sales cycle (B2B enterprise, real estate): stick with 90 days

A shorter window means touchpoints from further back don’t get credit. A longer window means more touchpoints compete for credit. Neither is objectively “better” — it depends on how long your customers actually take to decide.

Check your own data first. Go to Advertising → Conversion paths and look at the average days to conversion. If 95% of your conversions happen within 14 days, a 90-day lookback window adds noise without useful signal.

Step 3 — Mark Your Key Events as Conversions

Attribution only works on key events (what GA4 now calls conversions). If you haven’t marked the right events as conversions, you’re attributing credit for the wrong actions.

GA4 marks purchase as a key event by default. Everything else — form submissions, sign-ups, downloads — you need to set manually.

Here’s how:

1. Go to Admin → Events
2. Find the event you want to mark as a conversion
3. Toggle the Mark as key event switch to on

If the event doesn’t exist yet, you need to create it first. See my guide on how to set up event tracking in GA4 for the full process.

What should be a key event? Only actions that represent real business value:

• E-commerce: purchase, add_to_cart, begin_checkout
• SaaS: sign_up, trial_start, subscription_activated
• Lead gen: generate_lead, form_submit, phone_call
• Content: newsletter_signup, resource_download

Don’t mark too many events as key events. I’ve seen properties with 20+ conversions — at that point, your attribution data becomes meaningless because every session triggers multiple conversions. Stick to 3-5 events that represent genuine business outcomes.

For deeper analysis of what happens after conversion, check out my guide on conversion analysis.

Step 4 — Read Attribution Reports

Once your settings are configured, it’s time to use the reports. GA4’s attribution reports live in the Advertising section (left sidebar). If you don’t see “Advertising” in the navigation, you need at least one active key event with conversion data.

There are three key reports to know:

1. Model comparison

This report shows conversions and revenue credited to each channel under different attribution models side by side. Use it to see how credit shifts between data-driven and last click. If a channel gets significantly more credit under DDA than last click, it’s an important upper-funnel contributor that last click ignores.

2. Conversion paths

This shows the actual sequence of touchpoints users follow before converting. You’ll see early touchpoints (awareness), mid touchpoints (consideration), and late touchpoints (decision). This is where you discover patterns like “users who see a display ad first and then search for us convert at 2x the rate.”

3. Advertising snapshot

A high-level overview showing which channels are driving the most conversions. Less detailed than the other two but useful for quick executive summaries.

When reading these reports, always check the date range and the conversion event filter at the top. If you have multiple key events, the report aggregates them by default. Filter to a single conversion type for cleaner insights.

Want to understand the difference between first-click and last-click models in more detail? I break that down in my first click vs last click attribution comparison.

Step 5 — Build Custom Attribution Reports in Explorations

The standard Advertising reports give you a good overview, but Explorations let you dig deeper. You can build custom attribution reports that answer specific questions about your conversion paths.

Here’s how to create a useful custom attribution exploration:

1. Go to Explore in the left sidebar
2. Click Blank to start fresh (or use the Path exploration template)
3. Add dimensions: Session source/medium, Session campaign, Session default channel group
4. Add metrics: Conversions, Total revenue, Sessions
5. Set the technique to Free form
6. Drag your dimensions to Rows and metrics to Values

Pro tip: Use the Path exploration technique to visualize the actual journey users take. Set the starting point as a specific channel (like paid search) and the ending point as your key event. This reveals which channels work well together in the conversion path.

You can also use Segment overlap to compare users from different channels. Create segments for your top three channels and see how many users interact with multiple channels before converting. If there’s significant overlap, that’s a strong signal that multi-touch attribution matters for your business.

Attribution Settings vs Google Ads Attribution — They’re Separate

This catches people off guard. GA4 attribution settings and Google Ads attribution settings are completely independent. Changing one does not affect the other.

Setting	GA4 Attribution	Google Ads Attribution
Where to configure	GA4 Admin → Attribution settings	Google Ads → Tools → Attribution
What it affects	GA4 reports only	Google Ads reports and bidding
Models available	Data-driven, Last click	Data-driven, Last click
Scope	All channels (organic, paid, social, etc.)	Google Ads clicks and impressions only
Used for bidding?	No (reporting only)	Yes — directly feeds Smart Bidding

If you’re running Google Ads, configure both. Your GA4 reports will show how all your channels work together. Your Google Ads attribution will optimize your ad spend. They should use the same model (data-driven) for consistency, but they’ll still show different numbers because they count differently.

GA4 looks at the full cross-channel journey. Google Ads only looks at ad interactions. A conversion attributed to “Paid Search” in GA4 might be attributed differently in Google Ads if the user had multiple ad clicks.

Advanced: Cross-Channel vs Ads-Preferred Attribution

In your GA4 attribution settings, you’ll notice an option for the attribution scope: Cross-channel or Ads-preferred. This controls which channels are eligible to receive attribution credit.

• Cross-channel (recommended): All channels compete for credit — organic search, paid search, social, direct, email, referral. This gives you the most honest view of what’s working
• Ads-preferred: Google Ads touchpoints get credit whenever they’re in the conversion path, even if other channels were more influential. Use this only if you need to maximize credit to paid channels for internal reporting to stakeholders who manage ad budgets

My recommendation: Use cross-channel. Ads-preferred inflates the perceived value of paid channels at the expense of organic and direct traffic. It’s not “wrong” — but it creates a bias that can lead to over-investing in paid and under-investing in organic.

The one exception: if your organization’s ad team and organic team operate independently with separate budgets, ads-preferred helps the ad team see their specific impact more clearly. Just know that it’s painting a partial picture.

Common Attribution GA4 Setup Mistakes

After auditing hundreds of GA4 properties, these are the mistakes I see over and over:

1. Never changing default settings

The defaults are reasonable but generic. A 90-day lookback window for an e-commerce site selling $20 products is overkill. A 30-day window for B2B enterprise deals is too short. Match your settings to your actual sales cycle.

2. Too many key events

When everything is a conversion, nothing is. If you mark page_view, scroll, and every button click as key events, your attribution reports drown in noise. Keep it to 3-5 genuine business outcomes.

3. Ignoring UTM parameters

Attribution relies on GA4 knowing where traffic comes from. If your email campaigns, social posts, and partner links don’t have proper utm_source, utm_medium, and utm_campaign parameters, GA4 lumps them into “Direct” or “(not set).” Garbage in, garbage out.

4. Confusing GA4 attribution with Google Ads attribution

As covered above, these are separate systems. I’ve seen teams change their GA4 model and expect their Smart Bidding to respond. It won’t. Configure both independently.

5. Not checking data quality first

Attribution is only as good as your underlying data. Before trusting your attribution reports, verify that your tracking is clean — events fire correctly, UTM parameters are consistent, and there’s no duplicate tracking inflating numbers. Start with solid event tracking before layering attribution analysis on top.

6. Using ads-preferred when you need cross-channel insights

Ads-preferred makes your paid campaigns look better than they might actually be. If you’re making holistic marketing decisions across all channels, cross-channel attribution gives you the unbiased view you need.

FAQ

What is the best attribution model in GA4?

Data-driven attribution is the best choice for most properties. It uses machine learning to analyze your actual conversion patterns and distribute credit based on real data rather than arbitrary rules. The only reason to use last click is if you need simplicity or have very low conversion volume (under 300 per month).

Does changing the attribution model affect historical data?

Yes. When you switch attribution models in GA4, it recalculates historical reports retroactively. This means past data will shift to reflect the new model. It’s not rewriting raw data — the underlying events stay the same — but the credit distribution in reports changes. This is actually a benefit since you get consistent data across all time periods.

Why can’t I see the Advertising section in GA4?

The Advertising section only appears when GA4 has sufficient conversion data. You need at least one active key event that has received conversions. If you’ve just set up your property or haven’t marked any events as key events, the section won’t show. Mark your key events, wait for data to accumulate (usually 24-48 hours), and it should appear.

What lookback window should I use for e-commerce?

For most e-commerce businesses, 30 days is sufficient. Check your Conversion paths report to see how long customers actually take to convert. If 90% of purchases happen within 7 days, a 30-day window captures nearly all relevant touchpoints without introducing noise from old interactions that didn’t really influence the purchase.

Can I use different attribution models for different reports?

Not exactly. The attribution model you set in Admin applies to all standard reports. However, the Model comparison report in the Advertising section lets you compare how two models would distribute credit side by side. This gives you the practical ability to see data under different models without switching your property-level setting.

Wrapping Up

Setting up attribution in GA4 comes down to five steps: choose your model (data-driven for most), set lookback windows that match your sales cycle, mark the right key events, learn to read the attribution reports, and build custom explorations for deeper analysis. Layer on awareness of the GA4 vs Google Ads distinction and the cross-channel vs ads-preferred choice, and you’ve got a solid attribution setup.

The biggest mistake isn’t choosing the wrong model — it’s not configuring attribution at all. Default settings are generic. Your business isn’t. Take 15 minutes to match your GA4 attribution setup to your actual customer journey, and your channel investment decisions get significantly sharper.

Next step: read my complete guide to attribution modeling to understand the theory behind the models, or jump into first click vs last click attribution if you want to see how different models change your data in practice.