Data-Driven vs Last Click Attribution in Google Ads: Which Should You Use?
When setting up conversion tracking in Google Ads, advertisers are now faced with a much simpler — yet more controversial — choice regarding attribution models. Google has removed most attribution options, leaving advertisers with just two primary models:
- Data-Driven Attribution
- Last Click Attribution
For many advertisers, this change has created confusion around which model is best for measuring campaign performance accurately. Understanding how these attribution models work is essential if you want to optimise your campaigns, scale effectively, and avoid making poor budgeting decisions.
In this article, we’ll break down both attribution models, explain how Google’s data-driven system actually works, and help you decide which option makes the most sense for your business.
What Is Attribution in Google Ads?
Attribution is the process of assigning credit for a conversion to the ads, keywords, and touchpoints that influenced a user before they completed an action.
For example, a customer may:
- Search for a generic keyword
- Click an ad
- Leave the website
- Search again later
- Click another ad
- Finally convert
The question becomes:
Which interaction should receive credit for the conversion?
That’s exactly what attribution models are designed to answer.
Why Google Removed Other Attribution Models
Google previously offered several attribution models, including:
- Linear attribution
- Position-based attribution
- Time decay attribution
- First click attribution
However, Google has gradually removed these options in favour of simplifying attribution around two core approaches: data-driven and last click.
The reasoning behind this shift is tied to increasing privacy restrictions and the growing difficulty of accurately tracking user journeys across devices and sessions. Google believes its machine-learning-based data-driven model performs better than the older rule-based systems.
Whether advertisers fully agree with that is another question entirely.
What Is Data-Driven Attribution?
Data-driven attribution (DDA) uses machine learning and statistical modelling to determine how much credit each interaction deserves in a conversion journey.
Rather than assigning 100% of the credit to a single click, Google distributes conversion value across multiple touchpoints.
For example, imagine a user:
- Searches for a generic keyword
- Later clicks a remarketing ad
- Finally searches your brand name and converts
Instead of giving all credit to the final brand search, data-driven attribution may assign:
- 30% credit to the generic keyword
- 20% credit to the remarketing interaction
- 50% credit to the branded keyword
This provides a more balanced view of how users move through your sales funnel.
Why Google Recommends Data-Driven Attribution
Whenever you create conversions in Google Ads, Google strongly encourages advertisers to use data-driven attribution.
The reason is simple:
Modern customer journeys are rarely linear.
Users interact with multiple ads, devices, searches, and sessions before converting. Data-driven attribution attempts to reflect this complexity more accurately than traditional models.
This can help advertisers:
- Understand upper-funnel performance
- Identify assisting keywords
- Optimise campaigns more effectively
- Avoid overvaluing branded traffic
- Scale campaigns with better data
Why You See Decimal Conversions in Google Ads
One of the most confusing things about data-driven attribution is seeing fractional conversions in reports.
For example:
- 1.3 conversions
- 0.7 conversions
- 2.4 conversions
Obviously, there’s no such thing as 0.4 of a lead or sale.
These decimals appear because Google is splitting conversion credit across multiple touchpoints. If two keywords both contributed to a conversion, Google may assign partial credit to each.
This is perfectly normal within a data-driven model.
How Accurate Is Data-Driven Attribution?
This is where things become important.
Despite the sophisticated technology behind it, data-driven attribution is not an exact science.
Google does not know with complete certainty which interaction caused a conversion. Instead, it uses predictive modelling and probability analysis to estimate the influence of different touchpoints.
In essence, it’s intelligent statistical guesswork.
That doesn’t necessarily make it bad — in fact, it’s often extremely useful — but advertisers should understand that attribution is increasingly based on modelling rather than perfect tracking.
How Google’s Data-Driven Attribution Actually Works
Google’s system uses both converting and non-converting users to build predictive models.
It analyses:
- Historical ad performance
- User interaction patterns
- Click behaviour
- Impression data
- Conversion likelihood
Google’s own documentation explains that its algorithm uses an adaptation of survival analysis, a statistical technique commonly used in medical and clinical research.
Essentially, Google compares groups of users:
- Users exposed to certain ads
- Users not exposed to those ads
The platform then measures the statistical increase in conversion probability caused by those ad interactions.
A Simple Example of Data-Driven Attribution
Imagine someone searching for a new mobile phone.
Their journey might look like this:
- “Best tech gifts”
- “Top rated phones”
- “Google Pixel 4”
Google may determine this journey has a 3% probability of conversion.
Now imagine another user follows the same journey but never sees the “Google Pixel 4” ad.
In this case, the conversion probability may fall to 2%.
Google interprets this as the Pixel ad increasing conversion likelihood by 50%.
That additional uplift helps Google decide how much attribution credit the keyword deserves.
This is the foundation of data-driven attribution.
What Is Last Click Attribution?
Last click attribution is much simpler.
Whichever interaction immediately precedes the conversion receives 100% of the credit.
Using the earlier example:
- First ad click = 0% credit
- Second ad click = 0% credit
- Final converting click = 100% credit
This model focuses entirely on the action that directly generated the conversion.
The Problem With Last Click Attribution
While last click attribution is straightforward, it can create serious optimisation problems.
The main issue is that it ignores the earlier stages of the customer journey.
For example:
A user may first discover your business through a non-branded keyword such as:
- “best accounting software”
- “local web designer”
- “kitchen renovation company”
They later search your brand name and convert.
With last click attribution:
- Your branded keyword gets all the credit
- The discovery keyword gets none
This can lead advertisers to make poor decisions by:
- Overinvesting in branded traffic
- Underfunding top-of-funnel campaigns
- Misunderstanding customer acquisition paths
- Limiting account scalability
Over time, this can severely restrict growth.
When Last Click Attribution Can Still Work
Despite its limitations, there are situations where last click attribution may still be useful.
Specifically:
Small Campaigns With Low Volume
If your campaigns have:
- Very small budgets
- Low impression volume
- Few conversions
- Limited traffic
Then data-driven attribution may struggle to produce meaningful insights.
Why?
Because Google’s machine learning requires sufficient data to build accurate predictive models.
Without enough conversion volume, the system has very little information to work with.
Why Low-Budget Campaigns Behave Differently
Smaller campaigns often don’t generate enough repeat interactions.
In many cases:
- A user clicks once
- Either converts or disappears
Low budgets make it difficult for Google to repeatedly serve ads to the same users across multiple touchpoints.
As a result, the customer journey becomes much shorter and simpler.
In these situations, last click attribution may actually provide cleaner and more reliable reporting.
Which Attribution Model Should You Use?
For most advertisers, data-driven attribution is usually the better choice.
It provides:
- Better visibility into the full customer journey
- Improved optimisation opportunities
- More scalable campaign insights
- Stronger upper-funnel reporting
However, last click attribution can still make sense if:
- Your campaigns are extremely small
- Conversion volume is limited
- Your budget restricts repeat exposure
- User journeys are very short
The key is understanding the limitations of each model rather than blindly accepting Google’s recommendations.
The Bigger Challenge Facing Digital Advertisers
Attribution is becoming increasingly difficult across the entire digital marketing industry.
Privacy restrictions, tracking limitations, and machine-learning automation are pushing advertisers further into “black box” systems where platforms control more of the decision-making process.
As advertisers, we are gradually receiving:
- Less visibility
- Less raw data
- Less transparency
Understanding how attribution models work is therefore more important than ever.
Even if the systems are imperfect, knowing how Google interprets your conversions helps you make smarter optimisation decisions and avoid misleading performance analysis.
Final Thoughts
Google’s shift towards data-driven attribution reflects the broader evolution of digital advertising.
For most businesses, data-driven attribution offers a more complete understanding of how campaigns influence conversions across multiple touchpoints. However, it’s important to remember that these systems rely heavily on modelling and probability rather than perfect certainty.
Meanwhile, last click attribution still has a place for smaller advertisers with limited data and simple customer journeys.
The best approach is not simply choosing the “recommended” option — it’s understanding how attribution impacts your optimisation strategy, budgeting decisions, and long-term campaign growth.
