Google Ads Attribution: Complete Guide to Conversion Tracking

Attribution determines which ad interactions get credit for conversions. It directly impacts how you optimize and where you spend.

The Attribution Challenge

Modern customer journeys are complex:

1. User searches on phone, clicks ad
2. Later browses on desktop
3. Sees remarketing ad
4. Clicks branded search and converts

Which touchpoint deserves credit?

Attribution Impact

Different attribution models give different answers:

• Last-click: Brand search gets 100%
• First-click: Initial phone search gets 100%
• Linear: Each touchpoint splits equally
• Data-driven: ML determines credit

Why This Matters

If you use last-click attribution:

• Brand campaigns look amazing
• Prospecting campaigns look expensive
• You may underinvest in top-of-funnel

If you use data-driven:

• Credit is distributed more accurately
• Top-of-funnel gets appropriate credit
• Optimization decisions are better informed

Key Attribution Concepts

Conversion window: How long after click/view to count conversion (1-90 days)

View-through conversions: Conversions after ad view (no click)

Cross-device tracking: Connecting journeys across devices

Assisted conversions: Touchpoints that helped but didn't get last-click credit

Google Ads offers several attribution models for search and shopping campaigns.

Data-Driven Attribution (DDA)

How it works:

• Machine learning analyzes your conversion paths
• Compares paths that convert vs. don't convert
• Assigns credit based on actual impact
• Adjusts based on your specific data

Requirements:

• 300 conversions in 30 days (lowered from previous requirements)
• Available for most accounts now

Best for:

• Most accounts that meet the threshold
• Complex customer journeys
• Multi-campaign accounts

Last Click

How it works:

• 100% credit to the last clicked ad/keyword

Pros:

• Simple to understand
• Clear cause-and-effect
• Easy to optimize against

Cons:

• Ignores top-of-funnel contribution
• Overvalues brand/retargeting
• Undervalues prospecting

Best for:

• Simple, short purchase cycles
• Direct-response campaigns
• When you need simplicity

First Click

How it works:

• 100% credit to the first clicked ad/keyword

Pros:

• Values customer acquisition
• Shows how awareness starts

Cons:

• Ignores closing touchpoints
• May overvalue broad keywords
• Rarely reflects actual journey value

Best for:

• Understanding awareness campaigns
• Acquisition-focused businesses

Linear

How it works:

• Equal credit to all touchpoints

Pros:

• Recognizes all contributors
• Simple to understand

Cons:

• May give too much credit to weak touchpoints
• Doesn't reflect actual impact differences

Best for:

• When all touchpoints are genuinely equal
• Long consideration cycles

Time Decay

How it works:

• More credit to touchpoints closer to conversion
• Earlier touchpoints get less credit

Pros:

• Values closing while acknowledging assists
• More realistic than last-click

Cons:

• May still undervalue top-of-funnel
• Fixed decay pattern may not fit your business

Best for:

• When closing matters most but assists matter too
• Shorter sales cycles

Position-Based

How it works:

• 40% to first click
• 40% to last click
• 20% split among middle touchpoints

Pros:

• Values acquisition and closing
• Acknowledges middle touchpoints

Cons:

• Fixed percentages may not reflect reality
• 40/40/20 may not suit your business

Best for:

• When first and last touch both matter
• Multi-step journeys

Select the right model based on your business and goals.

Decision Framework

Question 1: Do you meet DDA requirements?

• Yes → Consider DDA first
• No → Choose a rules-based model

Question 2: How complex is your customer journey?

• Simple (1-2 touchpoints) → Last click may be fine
• Complex (3+ touchpoints) → DDA or position-based
• Very long cycles → Time decay or DDA

Question 3: What's your campaign mix?

• Brand only → Last click is acceptable
• Brand + non-brand → DDA or position-based
• Full funnel → DDA strongly recommended

Question 4: What are you optimizing for?

• Immediate conversions → Last click
• Customer acquisition → First click or DDA
• Balanced view → DDA or position-based

Model Comparison by Scenario

Scenario: Ecommerce with brand + prospecting

• Last click: Brand looks great, prospecting looks expensive
• DDA: Prospecting gets more credit, brand gets less
• Recommendation: DDA

Scenario: Lead gen with short cycle

• Last click: Probably accurate
• DDA: May reveal unexpected assists
• Recommendation: Either, test DDA

Scenario: B2B with long sales cycle

• Last click: Miss entire early journey
• DDA: Better reflects multi-touch reality
• Recommendation: DDA or time decay

Transitioning Attribution Models

Before changing:

1. Document current performance metrics
2. Understand how credit will shift
3. Prepare stakeholders for changes
4. Plan optimization adjustments

When you change:

• Expect metrics to shift
• Brand campaigns typically lose credit
• Prospecting campaigns typically gain credit
• Total conversions stay the same

After changing:

• Don't make immediate optimization decisions
• Let data stabilize (2-4 weeks)
• Compare performance directionally
• Adjust targets based on new model

Conversion windows define how long after an interaction a conversion counts.

Types of Conversion Windows

Click-through window:

• Time between click and conversion
• Options: 1-90 days
• Counts conversions after ad clicks

View-through window:

• Time between ad view (no click) and conversion
• Options: 1-30 days
• Counts conversions after ad impressions

Engaged-view window (Video):

• Time after watching 10+ seconds of video
• Options: 1-30 days
• Middle ground between view and click

Choosing Click-Through Windows

Business Type	Recommended Window	Reasoning
Impulse purchases	1-7 days	Short decision cycle
Ecommerce (general)	7-30 days	Research and comparison
High-ticket items	30-60 days	Long consideration
B2B	30-90 days	Complex buying process
Subscriptions	14-30 days	Moderate consideration

Choosing View-Through Windows

More conservative approach:

• 1 day: Very conservative, only immediate impact
• 3 days: Reasonable for most
• 7 days: Standard default
• 30 days: Maximum, may overcount

Recommendation: Start conservative (1-3 days), expand if needed.

Window Impact on Data

Longer windows:

• More conversions counted
• Better for long sales cycles
• May overcount for short cycles
• Harder to connect specific actions

Shorter windows:

• Fewer conversions counted
• More confident attribution
• May undercount assisted conversions
• Better for quick decisions

Testing Window Length

Use Conversion Path reports to understand:

1. Actual time between first touch and conversion
2. Typical number of touchpoints
3. How many conversions happen after various time periods

Set windows based on actual customer behavior, not arbitrary choices.

Understanding how Google tracks users across devices.

How Cross-Device Works

Signed-in users:

• Google links activity via Google account
• Most accurate connection
• Requires user to be signed in

Modeled conversions:

• Google estimates cross-device for non-signed-in users
• Uses patterns from signed-in users
• Less accurate but extends coverage

What's Tracked

When cross-device is enabled:

• User clicks ad on mobile
• Later converts on desktop
• Conversion attributed to mobile click

This matters because:

• Many users research on mobile
• Many users convert on desktop
• Without cross-device, mobile looks bad

Cross-Device Data in Reports

Find cross-device data in:

• Conversions > Cross-device conversions
• Attribution > Device paths
• Segment any report by device

Interpreting cross-device data:

• "Cross-device conversions" column shows these specifically
• Compare to same-device conversions
• Understand true device performance

Cross-Device Optimization

If mobile has high cross-device conversions:

• Don't reduce mobile bids based on direct CVR alone
• Consider total value including cross-device
• Mobile may be crucial for starting journeys

If desktop has low cross-device but high direct:

• Desktop is likely the closing device
• Ensure desktop experience is optimized
• Consider different messaging for each device

Privacy and Cross-Device

Limitations:

• Requires Google sign-in for deterministic matching
• Cookie restrictions affect modeling
• Opt-outs reduce data
• Privacy regulations vary by region

Implications:

• Reported cross-device is likely undercount
• True cross-device impact is probably higher
• Use directional insights, not exact numbers

Google provides several tools for understanding attribution.

Attribution Reports in Google Ads

Location: Tools & Settings > Measurement > Attribution

Key reports:

Overview:

• High-level attribution summary
• Model comparison
• Top converting paths

Model Comparison:

• See how conversions differ by model
• Compare DDA to last-click
• Identify undervalued campaigns

Conversion Paths:

• See actual customer journeys
• Understand touchpoint sequences
• Find common patterns

Path Metrics:

• Average path length
• Average time to conversion
• Top path types

Top Paths Report

Shows most common conversion paths:

• Which campaigns appear in paths
• Order of touchpoints
• Frequency of each path

Use this to:

• Understand customer journey
• Identify key campaigns in paths
• Find unexpected contributors

Assisted Conversions

Campaigns can be:

• Last-click converters (closed the deal)
• Assisters (helped but didn't close)
• Both (sometimes last, sometimes assist)

Assisted conversion metrics:

• Assisted conversions count
• Assisted conversion value
• Assisted/Last-click ratio

High assist ratio means:

• Campaign helps other campaigns convert
• Cutting it may hurt overall performance
• Consider value beyond direct conversions

Time Lag Report

Shows days between first touch and conversion:

• 0 days: Immediate conversions
• 1-7 days: Short consideration
• 8-30 days: Medium consideration
• 30+ days: Long consideration

Use this to:

• Set appropriate conversion windows
• Understand your sales cycle
• Time follow-up campaigns

Path Length Report

Shows number of interactions before conversion:

• 1 interaction: Direct path
• 2-3 interactions: Short path
• 4+ interactions: Long path

Use this to:

• Understand journey complexity
• Choose appropriate attribution model
• Evaluate campaign contributions

Use attribution insights to make better optimization decisions.

Identifying Undervalued Campaigns

Signs a campaign is undervalued:

• High assisted conversions
• Frequently in early path position
• Low last-click but high assists
• DDA gives significantly more credit than last-click

What to do:

• Don't cut based on last-click alone
• Consider full-funnel contribution
• Test increasing budget and bids
• Monitor total account performance

Identifying Overvalued Campaigns

Signs a campaign may be overvalued:

• High last-click, zero assists
• Brand campaigns with inflated numbers
• Getting credit for inevitable conversions

What to do:

• Test reducing budget slightly
• Monitor incremental impact
• Consider brand/non-brand separation
• Don't blindly cut—test incrementally

Budget Allocation with Attribution

Traditional (last-click based):

• All budget to high last-click ROAS
• Starve top-of-funnel
• Short-term thinking

Attribution-informed:

• Consider assist value
• Fund top-of-funnel appropriately
• Balance acquisition and closing

Example reallocation:

Campaign	Last-Click ROAS	DDA ROAS	Action
Brand	15:1	8:1	Slight reduce
Non-brand	3:1	5:1	Increase
Prospecting	1:1	3:1	Increase
Remarketing	10:1	6:1	Maintain

Target Setting with Attribution

When using DDA, targets may need adjustment:

If moving from last-click to DDA:

• Brand campaign targets may need lowering
• Prospecting targets may need raising
• Overall targets should remain similar

Calculate new targets:

1. Compare credit by model
2. Adjust proportionally
3. Test and refine

Testing Attribution Impact

Experiment approach:

1. Select test campaigns
2. Compare performance under different models
3. Make controlled budget changes
4. Measure total account impact
5. Adjust strategy based on learnings

Step-by-step implementation guide.

Setup Checklist

Conversion tracking:

• All conversions properly tracked
• Conversion values accurate
• No duplicate conversions
• Enhanced conversions enabled if eligible

Attribution model:

• Current model documented
• Data requirements for DDA checked
• Appropriate model selected
• Stakeholders informed of model

Conversion windows:

• Click-through window set appropriately
• View-through window configured
• Windows match actual customer journey
• Documented reasoning for choices

Cross-device:

• Cross-device tracking enabled
• Understanding of limitations
• Device-specific strategies considered

Regular Review Process

Weekly:

• Check conversion counts are normal
• Monitor any tracking issues
• Review real-time data

Monthly:

• Review attribution reports
• Compare model performance
• Identify undervalued/overvalued campaigns
• Consider optimization actions

Quarterly:

• Full attribution audit
• Model appropriateness review
• Window length review
• Strategy adjustment

Stakeholder Communication

Explain to stakeholders:

• How attribution affects reported metrics
• Why metrics may differ from other platforms
• What attribution model is used and why
• How to interpret cross-device data

Common questions to address:

• "Why do Facebook and Google show different numbers?"
• "Why did campaign performance change when we switched models?"
• "Which numbers are correct?"

Documentation to maintain:

• Attribution model in use
• Conversion windows and reasoning
• Date of any changes
• Impact of changes

Troubleshooting

Sudden conversion drop:

• Check tracking implementation
• Review for duplicate removal
• Check consent/privacy changes
• Verify conversion window catches conversions

Metrics don't match GA:

• Different attribution models
• Different conversion windows
• Cross-device differences
• Session vs. user-based counting

DDA not available:

• Insufficient conversion volume
• Check 30-day conversion count
• Consider alternative models
• Build volume before switching