Google Ads Attribution Models: Which One Should You Use for Ecommerce?

Attribution determines which touchpoints get credit for conversions. Choose wrong, and you'll kill profitable campaigns or pour budget into channels that don't actually drive sales.

Google Ads offers six attribution models. Each one tells a different story about how your marketing drives revenue. This guide breaks down which model works best for DTC brands based on your customer journey complexity and conversion volume.

Why Attribution Matters for Ecommerce

Most customers don't convert on first touch. They: - See a Meta ad → don't click - Search your brand name → visit site → leave - Click a retargeting ad → add to cart → abandon - Search product category → click Shopping ad → purchase

Question: Which campaign gets credit for the sale?

Your attribution model answers that question—and directly impacts how Google's algorithms optimize your campaigns.

The 6 Google Ads Attribution Models

1. Last Click (Default)

How it works: 100% credit to the final ad clicked before conversion. Example:

- Day 1: User clicks Facebook ad → visits site → leaves - Day 3: User clicks Google Search ad → purchases

Credit distribution: Google Search ad gets 100% credit. Facebook gets 0%. Best for:

- Brands with simple, short sales cycles (1-3 days) - High-intent products where last touch drives decision - Campaigns focused purely on bottom-funnel conversions

Limitations:

- Ignores all earlier touchpoints - Under-values awareness and consideration channels - Favors retargeting and brand search over prospecting

2. First Click

How it works: 100% credit to the first ad clicked in the conversion path. Example:

- Day 1: User clicks Display ad → visits site → leaves - Day 5: User clicks Shopping ad → purchases

Credit distribution: Display ad gets 100% credit. Shopping gets 0%. Best for:

- Awareness campaigns where you want to measure top-of-funnel impact - Testing new acquisition channels - Understanding which platforms introduce new customers

Limitations:

- Ignores what actually closed the sale - Not useful for optimizing conversion-focused campaigns - Rarely recommended for ecommerce

3. Linear

How it works: Equal credit distributed across all touchpoints in the conversion path. Example:

- Day 1: Display ad → 33.3% credit - Day 3: YouTube ad → 33.3% credit - Day 5: Search ad → 33.3% credit

Best for:

- Longer sales cycles (7+ days) - Brands running full-funnel campaigns across multiple platforms - Understanding contribution of every touchpoint

Limitations:

- Treats all touchpoints equally (early awareness = final click) - Can dilute focus on high-impact moments in customer journey

4. Time Decay

How it works: More credit to touchpoints closer to conversion. Typically uses 7-day half-life (touchpoints 7 days before conversion get half the credit of those 1 day before). Example:

- Day 1: Display ad → 10% credit - Day 5: YouTube ad → 30% credit - Day 7: Search ad → 60% credit

Best for:

- Moderate sales cycles (5-14 days) - Balancing awareness and conversion touchpoints - Seasonal campaigns where urgency increases near conversion

Limitations:

- Still somewhat arbitrary in credit distribution - May undervalue early touchpoints that introduced brand

5. Position-Based (U-Shaped)

How it works: 40% credit to first touch, 40% to last touch, 20% distributed evenly across middle touchpoints. Example:

- Day 1: Display ad → 40% credit - Day 3: YouTube ad → 10% credit - Day 5: Retargeting ad → 10% credit - Day 7: Search ad → 40% credit

Best for:

- Multi-channel strategies valuing both awareness and conversion - Brands wanting to balance prospecting and retargeting - Complex customer journeys with multiple mid-funnel touches

Limitations:

- Arbitrary weighting (why 40/20/40 vs. other ratios?) - Doesn't adapt to your actual customer behavior

6. Data-Driven Attribution (DDA)

How it works: Google's machine learning analyzes your actual conversion paths and assigns credit based on what statistically drives conversions. Requirements:

- 3,000+ ad interactions + 300+ conversions in 30 days for Search/Shopping - 2,000+ ad interactions + 400+ conversions in 30 days for Display/YouTube

Example:

Google analyzes thousands of conversion paths and determines: - Display ad → 15% credit (introduces brand, moderate correlation) - YouTube ad → 25% credit (strong engagement signal) - Shopping ad → 60% credit (highest conversion probability)

Credit distribution adapts as behavior changes.

Best for:

- High-volume accounts meeting minimum thresholds - Multi-channel campaigns needing accurate attribution - Brands serious about optimizing full-funnel performance

Limitations:

- Requires significant conversion volume - Black box (you can't see the exact algorithm) - Not available for accounts below thresholds

Which Model Should You Use?

For Most Ecommerce Brands: Data-Driven (If Qualified)

If you meet the volume requirements, data-driven attribution is the most accurate model because it's based on your actual data, not generic assumptions.

Advantages:

- Adapts to your specific customer journey - Gives more credit to touchpoints that statistically drive conversions - Improves automated bidding performance (Google's algorithms optimize better with accurate attribution) - Updates continuously as behavior changes

When DDA isn't available: Use position-based or linear as a bridge until you reach required volume.

For Low-Volume Accounts: Last Click or Position-Based

If conversion volume <300/month: Stick with last click.

- Simple, transparent, easy to understand - Works well with manual bidding strategies - Focuses optimization on bottom-funnel performance

If running multi-channel campaigns: Consider position-based.

- Values both awareness (first click) and conversion (last click) - Better than last click for brands investing in YouTube, Display, or other awareness channels - Still simple enough to explain to stakeholders

For Specific Use Cases

Testing new awareness channels (Display, YouTube, Discovery):

Use linear or position-based to measure their contribution without last-click bias.

Pure retargeting campaigns:

Last click works fine—you're only measuring bottom-funnel efficiency.

Long sales cycles (30+ days):

Time decay or data-driven to account for extended consideration periods.

How Attribution Affects Automated Bidding

Google's Smart Bidding (Target CPA, Target ROAS, Maximize Conversions) uses attribution data to optimize bids.

With last click:

- Algorithms optimize for final touchpoints - May under-bid on awareness campaigns - Favors retargeting and brand search

With data-driven:

- Algorithms optimize across full customer journey - More accurate value signals for each touchpoint - Better performance across all campaign types

Example: A Shopping campaign might look less efficient under last click (many assists, few last clicks), but data-driven reveals it's a critical mid-funnel touchpoint. With DDA, Google can bid more accurately on Shopping and improve overall ROAS.

Comparing Attribution Models: Real Scenario

Customer journey:

Clicks Display ad (Day 1)

Clicks YouTube ad (Day 3)

Searches brand name, clicks Search ad (Day 5)

Clicks Shopping ad, purchases $100 product (Day 7)

Credit distribution by model:

| Model | Display | YouTube | Brand Search | Shopping | |-------|---------|---------|--------------|----------| | Last Click | $0 | $0 | $0 | $100 | | First Click | $100 | $0 | $0 | $0 | | Linear | $25 | $25 | $25 | $25 | | Time Decay | $10 | $20 | $30 | $40 | | Position-Based | $40 | $10 | $10 | $40 | | Data-Driven* | $15 | $25 | $10 | $50 |

*Example allocation based on typical patterns—actual DDA varies by account.

Takeaway: Your attribution model fundamentally changes how you evaluate campaign performance and where you allocate budget.

Switching Attribution Models: What to Expect

Immediate Impacts

- Reported conversions change (same sales, different credit distribution) - CPA/ROAS shift by campaign (prospecting improves, retargeting may worsen) - Automated bidding adjusts (30-90 day learning period)

Performance Changes by Campaign Type

When switching from last click to data-driven: Search (Brand):

- Conversions decrease (loses credit to earlier touchpoints) - CPA appears higher - Reality: Brand search was over-credited under last click

Search (Non-Brand):

- Conversions often stay similar or increase slightly - Better reflects true acquisition value

Shopping:

- Conversions increase (gets credit for assists) - CPA improves - Budget allocation may increase

Display/YouTube:

- Conversions increase significantly (now credited for awareness) - Appears more efficient - Justifies higher budget allocation

Retargeting:

- Conversions decrease (shares credit with earlier touches) - CPA appears higher - Reality: Was over-credited for conversions initiated by other channels

Migration Best Practices

Don't panic when numbers shift. Total revenue doesn't change—only how credit is distributed. Run parallel reporting for 30 days:

- Keep last click as comparison view - Monitor data-driven as new model - Compare total account performance (should match)

Wait 30-60 days before major budget shifts:

- Let automated bidding adjust - Observe new performance patterns - Make incremental changes, not dramatic cuts

Attribution Windows and Lookback Periods

Attribution models work within defined time windows:

Default Google Ads windows:

- Search/Shopping: 30-day click, 1-day view - Display/YouTube: 30-day click, 1-day view-through - All campaigns: Engagement (clicks/views) to conversion

What this means:

- Conversions within 30 days of ad click are attributed - View-through conversions (saw ad, didn't click, converted within 1 day) count for Display/YouTube

Adjusting windows:

Longer sales cycles may need 60 or 90-day windows, but this requires manual configuration in Google Analytics 4.

Google Ads vs. Google Analytics 4 Attribution

Google Ads attribution:

- Only tracks Google Ads touchpoints - Used for bidding optimization within Google Ads - Attribution models set per account

Google Analytics 4 attribution:

- Tracks all traffic sources (Meta, TikTok, organic, email, etc.) - Cross-channel view of customer journey - Data-driven model default (no other options)

Key difference: GA4 shows how Google Ads works with other channels. Google Ads attribution only measures Google performance in isolation. For full picture: Use both. Google Ads attribution optimizes your campaigns; GA4 shows how all marketing works together. Learn more: Multi-Touch Attribution for DTC Brands

Common Attribution Mistakes

1. Never Changing from Last Click Default

Last click systematically under-values prospecting and awareness. If you're running multi-channel campaigns, you're misallocating budget.

Solution: Switch to position-based (immediate) or data-driven (once qualified).

2. Comparing Campaigns with Different Attribution Models

Campaign A on last click vs. Campaign B on data-driven creates apples-to-oranges comparisons.

Solution: Use same attribution model across all campaigns for fair comparison.

3. Ignoring View-Through Conversions

Display and YouTube ads often drive awareness without immediate clicks. View-through conversions capture delayed impact.

Solution: Monitor view-through alongside click-through conversions for full picture.

4. Switching Models Mid-Campaign

Changing attribution during a campaign causes performance fluctuations and confuses automated bidding.

Solution: Change attribution at campaign launch or during low-volume periods (not during Black Friday).

5. Expecting Attribution to Solve All Measurement Problems

Attribution models distribute credit, but they don't: - Fix broken conversion tracking - Account for offline sales - Measure brand lift or long-term value - Show individual-level customer journeys

Solution: Use attribution as one measurement tool alongside incrementality testing, surveys, and cohort analysis.

Advanced: Incrementality Testing

Attribution shows correlation (what touchpoints were present), not causation (what actually drove the sale).

Incrementality testing measures true causal impact by comparing:

- Test group: Exposed to campaign - Control group: Not exposed to campaign

Difference in conversion rate = true incremental impact Example:

- Hold out 10% of brand search traffic from ads - Measure how many still convert organically vs. paid group - Calculate true incremental value of brand search ads

When to use:

- Validating attribution model assumptions - Deciding whether to invest in upper-funnel channels - Measuring cannibalization between channels

How ATTN Approaches Attribution

At ATTN Agency, we use data-driven attribution wherever possible and complement it with cross-platform attribution tools (Triple Whale, Northbeam) for a complete view.

Our process:

Audit account to see if DDA requirements are met

If not, use position-based and build volume toward DDA eligibility

Set up parallel tracking in GA4 for cross-channel visibility

Run quarterly incrementality tests on brand search and retargeting

Build custom reports showing contribution of each platform to overall revenue

Real example: A supplement brand was cutting Display spend due to "poor last-click performance." We switched to data-driven attribution, revealing Display was driving 22% of total revenue through assists. We tripled Display budget, and overall ROAS improved by 18%.

Getting Started with Better Attribution

Step 1: Check If You Qualify for Data-Driven

- Navigate to Google Ads → Tools → Conversions → Attribution - See if data-driven is available - If not, monitor progress toward thresholds

Step 2: Choose Your Model

- Qualified for DDA: Switch to data-driven - Not qualified: Use position-based for multi-channel or last click for simplicity

Step 3: Communicate Changes to Stakeholders

Explain that conversion numbers will shift, but total revenue stays the same—credit is just distributed more accurately.

Step 4: Set Parallel Reporting (First 30 Days)

- Keep previous model as comparison column - Monitor differences by campaign - Ensure totals match

Step 5: Let Automated Bidding Adapt (30-60 Days)

- Don't make drastic budget changes immediately - Watch for stabilization in CPA/ROAS - Optimize once algorithms adjust

Step 6: Review Quarterly

- Attribution models should evolve as customer behavior changes - Revisit model choice as account grows - Run incrementality tests to validate assumptions

Conclusion

Attribution isn't about finding the "perfect" model—it's about choosing the one that best reflects your customer journey and helps you make smarter budget decisions.

Quick decision framework: Use data-driven if:

- You have 300+ conversions/month - Running multi-channel campaigns - Using automated bidding

Use position-based if:

- <300 conversions/month - Running awareness + conversion campaigns - Want to value first and last touch

Use last click if:

- <100 conversions/month - Primarily bottom-funnel campaigns (retargeting, brand search) - Need simple, transparent reporting

Most importantly: Any attribution model is better than ignoring the customer journey entirely. Start where you are, and improve as you grow.

Need help setting up attribution that aligns with your full marketing strategy? Work with ATTN Agency to build measurement systems that show true marketing impact. Related reading:

- Google Ads for Ecommerce: Complete Strategy Guide - Multi-Touch Attribution for DTC Brands