Imagine this:
Someone discovers your business through a Google ad.
A few days later, they see your social media post.
Then they visit your website directly.
Later, they click an email.
Finally, they make a purchase.
So, which marketing channel deserves credit?
The Google ad? Social media? Email? The final website visit?
This is exactly the problem that attribution modeling tries to solve.
A marketing attribution model gives marketers a framework for deciding how much credit different touchpoints receive when a customer converts. The challenge is that different models can tell very different stories about the same customer journey.
In this guide, we’re going to look at attribution models compared side by side, with a particular focus on three familiar approaches:
- Last-click attribution
- Linear attribution
- Time-decay attribution
We’ll also look at where data-driven attribution fits today, because the attribution landscape has changed.
What Is Attribution Modeling?
Attribution modeling is the process of assigning conversion credit to marketing interactions that happen before a customer completes an important action.
That action could be:
- Making a purchase
- Submitting a lead form
- Booking a demo
- Signing up for a service
- Starting a free trial
- Downloading something
- Completing another defined conversion
The basic idea sounds simple.
A customer interacts with several marketing channels, and eventually converts. An attribution model decides how that conversion credit should be distributed across those interactions.
For example, imagine this customer journey:
Google Ad → Blog Post → Social Media → Email → Purchase
With last-click attribution, the email gets all the credit.
With linear attribution, all four marketing touchpoints share the credit equally.
With time-decay attribution, the email receives the most credit because it happened closest to the purchase.
Same customer.
Same conversion.
Very different attribution reports.
That’s why understanding attribution models compared is important when you’re making decisions about ad spend, campaigns, and marketing channels.
Last-Click Attribution: Give the Final Touchpoint All the Credit
Last-click attribution is probably the easiest attribution model to understand.
The rule is simple:
The final qualifying touchpoint before conversion receives 100% of the credit.
For example:
Google Ad → Blog → Social → Email → Purchase
Under last-click attribution:
- Google Ad = 0%
- Blog = 0%
- Social = 0%
- Email = 100%
The model basically asks:
“What did the customer interact with immediately before converting?”
That final interaction gets the credit.
Why marketers use it
The biggest advantage of last-click attribution is simplicity.
It’s easy to understand, easy to report, and easy to explain to a team.
If your goal is to understand what finally pushed someone over the line, the model provides a straightforward answer.
But there is an obvious limitation.
It ignores everything that happened earlier.
A customer may have discovered your company through a paid social campaign, read three blog posts, watched a video, joined your email list, and then converted after clicking an email.
Last-click attribution makes the email look like the entire reason for the conversion.
That’s not necessarily what happened.
Google Ads still supports last-click attribution, while Google has retired several older rule-based models in its current Google Ads and Analytics products. (Google Help)
Internal link opportunity:
Link Google Ads or ad performance to your previous article, What Is Quality Score, Really?, where readers can learn how ad quality and auction performance work.
Linear Attribution: Everyone Gets an Equal Share
Now let’s make things a little more democratic.
Linear attribution doesn’t pick one touchpoint and give it everything.
Instead, it distributes conversion credit equally across the tracked touchpoints in the conversion path.
Suppose the journey looks like this:
Paid Social → Google Search → Blog → Email → Purchase
There are four marketing touchpoints.
A linear model would assign:
- Paid Social = 25%
- Google Search = 25%
- Blog = 25%
- Email = 25%
The logic is:
Every touchpoint played a role, so every touchpoint gets an equal share.
This can be useful when you don’t want to assume that one particular interaction was responsible for the conversion.
The problem with linear attribution
Equal credit doesn’t necessarily mean equal influence.
Imagine the customer:
- Saw a social ad for two seconds.
- Read your detailed product comparison.
- Attended a webinar.
- Clicked an email.
- Purchased.
A linear model treats all four interactions equally.
But were they really equally important?
Maybe the webinar was the interaction that convinced the customer.
Maybe the comparison article did most of the research work.
Maybe the social ad simply introduced the brand.
Linear attribution doesn’t know.
It follows its rule.
And that’s the important thing to remember when attribution models are compared: every model makes assumptions.
Time-Decay Attribution: More Credit to Recent Touchpoints
Time-decay attribution takes a different approach.
Instead of giving every interaction equal credit, it gives more credit to touchpoints that happened closer to the conversion.
Imagine this customer journey:
Social Ad → Blog → Google Search → Email → Purchase
The social ad happened first.
The email happened immediately before the purchase.
Under time-decay attribution:
- Social Ad = smaller share
- Blog = slightly more
- Google Search = more
- Email = largest share
The exact distribution depends on the implementation and decay settings.
The underlying idea is:
The closer an interaction is to conversion, the more influence it may have had on the final decision.
This can make sense for businesses with relatively short sales cycles where recent interactions genuinely matter.
For example, imagine someone is looking for concert tickets.
They see an ad Monday.
They visit the website Tuesday.
They compare ticket options Wednesday.
They receive an email Thursday.
They buy Thursday evening.
A time-decay approach gives more weight to those later interactions because they happened closer to the purchase.
But there is a catch
Time-decay can undervalue awareness.
That first interaction might have been the reason the customer entered the journey in the first place.
Without that original discovery, the later interactions might never have happened.
So the model can still leave early-stage marketing looking less valuable than it actually was.
Attribution Models Compared at a Glance
Here’s the simple version:
| Attribution Model | How Credit Is Assigned | Main Focus | Potential Blind Spot |
| Last-Click | 100% to final touchpoint | Conversion/closing | Ignores earlier interactions |
| Linear | Equal credit across touchpoints | Full journey | Assumes every interaction is equally valuable |
| Time-Decay | More credit to recent interactions | Recent influence | Can undervalue awareness |
| Data-Driven | Based on observed data and modeling | Actual conversion patterns | Requires sufficient quality data |
The important thing is that these aren’t simply different ways of displaying the same answer.
They’re different ways of interpreting the customer journey.
Here’s Where It Gets Interesting: Data-Driven Attribution
If you’ve been reading about attribution recently, you’ve probably noticed another term appearing everywhere:
Data-driven attribution.
Unlike fixed rule-based models, data-driven attribution uses account data and machine learning to estimate the contribution of interactions across conversion paths.
Google says its data-driven attribution evaluates converting and non-converting paths and uses account data to determine how interactions contribute to key events. (Google Help)
That is a major difference.
Instead of saying:
“The last interaction gets 100%.”
or:
“Every interaction gets 25%.”
the model attempts to learn from actual conversion patterns.
Google currently uses data-driven attribution as the default for most conversion actions in Google Ads, while last-click remains available. (Google Help)
And here’s an important update if you’re reading older attribution guides:
Google Analytics and Google Ads no longer offer first-click, linear, time-decay, and position-based attribution models in their current standard attribution settings. Google Analytics says those models were removed as of November 2023. (Google Help)
That doesn’t mean linear or time-decay are useless concepts.
They are still extremely useful for understanding how different attribution assumptions work, and similar models remain available in some other marketing and advertising systems. Google Search Ads 360, for example, documents linear and time-decay models for certain use cases. (Google Help)
Why Attribution Models Can Tell Different Stories
Let’s say your customer journey contains five touchpoints:
Social → Search → Blog → Retargeting → Email → Purchase
Now imagine looking at the same conversion through different models.
Last-click
Email gets 100%.
Linear
Every touchpoint gets an equal share.
Time-decay
Email and retargeting receive more credit than the earlier social interaction.
Data-driven
The system analyzes historical patterns and assigns credit based on the observed contribution of interactions.
This is why you shouldn’t look at one attribution report and assume it represents an absolute truth.
An attribution model is a measurement framework.
Change the framework, and the way credit is distributed can change.
Attribution and Your Marketing Budget
This becomes especially important when you’re deciding where to put your next dollar.
Imagine your last-click report says:
- Email = 40 conversions
- Paid Search = 30 conversions
- Social = 5 conversions
You might conclude that email and search are doing almost all the work.
But what if social is responsible for introducing many of those customers to your brand?
Last-click attribution may not show that contribution.
This is one reason marketers use multiple attribution views when evaluating cross-channel marketing.
Instead of asking only:
“Which channel got the final click?”
you can ask:
“Which channels appeared throughout the customer journey?”
and:
“How does the picture change when we use a different attribution model?”
Google’s Model comparison reporting is designed to let marketers compare attribution approaches and examine differences in metrics such as conversion value and cost per conversion. (Google Help)
Try Attribution Modeling Instead of Just Reading About It
This is where Play2Lrn makes the concept much easier to understand.
Instead of staring at a static table of percentages, open the Attribution Modeller and experiment with the same customer journey.
The Play2Lrn simulation uses a fixed journey involving paid social, retargeting, video, email, and search. You can switch between first-click, last-click, linear, U-shaped, and time-decay approaches and watch the same order value move between touchpoints. (Play2Lrn)
[Try the Attribution Modeller →]
Move the controls.
Change the number of journey steps.
Adjust the time-decay setting.
Then watch what happens to the assigned credit.
That’s the part that tends to make attribution finally click.
How Attribution Connects With the Other Marketing Concepts
Attribution doesn’t exist in isolation.
It connects with many of the concepts you’ve already explored.
For example, your search intent determines what a person may be looking for at different stages of the journey.
Internal link: The 4 Types of Search Intent (And Why Google Cares More Than You Do)
Your advertising performance is also affected by things such as ad relevance and Quality Score.
Internal link: What Is Quality Score, Really? (Drag the Sliders and Watch Your CPC Change)
And once your ads enter an auction, the auction itself determines how ads compete for visibility.
Internal link: How a Second-Price Auction Actually Decides What You Pay
On the organic side, your customer’s discovery journey can also involve search visibility and SEO.
Internal link: Every On-Page, Off-Page, and Technical SEO Factor, on One Screen
Think of these topics as different pieces of the same marketing system:
Search → Discovery → Advertising → Interaction → Conversion → Attribution
That’s where the bigger picture starts to emerge.
So, Which Attribution Model Should You Use?
There isn’t one universal answer.
The useful question isn’t:
“Which attribution model is the best?”
Instead, ask:
“Which model gives me the information I need for the decision I’m trying to make?”
If you’re trying to understand the final interaction before conversion, last-click provides a simple view.
If you’re studying the idea of distributing credit evenly across a customer journey, linear attribution is easy to understand.
If recent interactions are particularly important to your analysis, time-decay provides a recency-focused framework.
If you’re working within Google Ads or Google Analytics today, you’ll also need to account for the current availability of data-driven and last-click models rather than assuming older rule-based models are still selectable. (Google Help)
And whatever model you use, remember that attribution depends on the quality of your tracking.
If important touchpoints aren’t captured, the model can’t give them credit.
Garbage in, garbage out still applies.
Final Takeaway
Attribution models compared side by side can look very different, but each model is answering a slightly different question.
Last-click asks:
“What happened immediately before the conversion?”
Linear asks:
“How can we distribute credit equally across the journey?”
Time-decay asks:
“Which interactions happened closest to the conversion?”
Data-driven attribution asks a more complex question:
“What does our conversion data tell us about the contribution of different interactions?”
The key is to understand the assumptions behind each approach before using the numbers to make marketing decisions.
And rather than memorizing the formulas, play with them.
Open the Play2Lrn Attribution Modeller, change the journey, switch the model, and watch the same conversion value move around.
Once you see the credit shift in real time, attribution modeling becomes much easier to understand.
Don’t just read attribution. Play with it.
FAQ
What is attribution modeling?
Attribution modeling is a method of assigning conversion credit to marketing touchpoints that occur during a customer’s journey. Different models distribute that credit in different ways.
What is last-click attribution?
Last-click attribution assigns 100% of the conversion credit to the final qualifying touchpoint before the conversion.
How does linear attribution work?
Linear attribution distributes conversion credit equally across the tracked touchpoints in a conversion path.
How does time-decay attribution work?
Time-decay attribution gives greater weight to touchpoints that occur closer to the conversion and less weight to earlier interactions.
Are linear and time-decay attribution still available in Google Analytics?
No. Google Analytics says first-click, linear, time-decay, and position-based attribution models are no longer available as of November 2023. Current Google Analytics attribution includes data-driven and last-click options for relevant reporting contexts. (Google Help)
Why should marketers compare attribution models?
Comparing models can show how different credit-assignment rules change the reported contribution of marketing touchpoints. This can provide additional context when analyzing campaigns, channels, conversions, and ad spend. (Google Help)