A customer sees your Instagram ad on Monday, searches your brand name on Wednesday, opens your newsletter on Friday, and buys the following week. Which of those moments made the sale?
An attribution model is the rule that answers that question. It decides how much credit each touchpoint gets for a conversion, and that credit shapes which campaigns look like winners and where your budget goes next. Pick a model that doesn’t match how your customers buy, and you’ll keep funding the channels that close sales while cutting the ones that start them.
Here’s how each model works, what it hides, what Google Ads and GA4 still offer today, and how to choose the right one for your business.
What Does a Marketing Attribution Model Do?
Every conversion path has touchpoints: an ad click, an organic visit, an email open, a referral. An attribution model is a set of rules for dividing the credit for one conversion across those touchpoints. Some give all of it to a single moment. Others spread it across the whole journey.
There’s a catch worth knowing before you compare models. Each ad platform applies its own model to the data it can see, and each one tends to give itself the credit. That’s why Google Ads, Meta, and GA4 can report more conversions combined than your business actually closed, a problem we break down in our guide to building a marketing attribution system that proves real ROI.
Single-Touch Attribution Models
Single-touch models give 100% of the credit to one touchpoint. They’re simple to read, which is why they’re still everywhere, but they ignore everything else that happened along the way.
First-Touch Attribution
First-touch attribution gives all the credit to the first interaction, such as the ad or post that introduced someone to your brand. It’s useful when your main goal is awareness and you want to know which channels bring in new people.
Its blind spot is everything after that first moment. The retargeting ad, the comparison article, and the sales call that actually closed the deal get zero credit.
Last-Touch Attribution
Last-touch (or last-click) attribution gives all the credit to the final interaction before the conversion. It’s the easiest model to measure, and for short, one-visit purchases, it’s often close enough.
Its blind spot is the opposite of first-touch. Branded search and email usually show up last, so they look like stars, while the prospecting campaigns that created the demand look like they did nothing. Teams that rely on last-touch alone often cut awareness spend and then wonder why branded searches drop a few months later.
Multi-Touch Attribution Models
Multi-touch attribution spreads credit across several touchpoints. These models give a fuller picture of longer journeys, but each one still uses a fixed rule that you choose, not one based on how your customers actually behave.
- Linear: every touchpoint gets an equal share. If someone had five interactions, each gets 20%. It’s fair but blunt, because a quick banner impression counts as much as a demo request.
- Time-decay: touchpoints closer to the conversion get more credit. It suits short sales cycles and promotions, where the last few days matter most.
- Position-based (U-shaped): the first and last touches get most of the credit, commonly 40% each, and the middle shares the remaining 20%. It values both the introduction and the close.
- W-shaped: a version of position-based that adds a third key moment, usually the point where someone becomes a lead. It’s popular in B2B, where the lead form matters as much as the first visit and the closed deal.
Data-Driven Attribution
Data-driven attribution doesn’t follow a fixed rule. It uses your own conversion data to compare the paths of people who converted with the paths of people who didn’t, then assigns credit based on how much each interaction actually changed the odds of a conversion (2).
In Google Ads, data-driven attribution is now the default for most conversion actions (2). Every conversion action is eligible, but Google recommends at least 200 conversions and 2,000 ad interactions within 30 days for the most reliable results (2). It’s also limited to what Google can see: clicks and video engagements on Search, Shopping, YouTube, Display, and Demand Gen ads, not your Meta campaigns, your emails, or what happens in your CRM.
Which Attribution Models Do Google Ads and GA4 Offer Today?
Fewer than they used to. In 2023, Google announced it was removing first-click, linear, time-decay, and position-based attribution from both Google Ads and GA4. Data-driven and last-click were kept, and conversion actions still using a removed model were switched to data-driven (1)(3).
So if you want to see how your campaigns look under first-touch or a U-shaped model, you won’t find that switch in Google’s tools anymore. The practical alternative is to run those models yourself on data you own, which is one of the biggest reasons to bring your ad, analytics, and CRM data into a central database.
Attribution Models Compared
| Model | How credit is split | Best for | Blind spot |
| First-touch | 100% to the first interaction | Measuring awareness and new-audience channels | Ignores everything that closed the sale |
| Last-touch | 100% to the final interaction | Short, one-visit purchases | Undervalues prospecting and awareness |
| Linear | Equal share to every touchpoint | Seeing every channel that played a part | Treats a minor touch like a major one |
| Time-decay | More credit to recent touchpoints | Short sales cycles and promotions | Undervalues early discovery |
| Position-based (U-shaped) | About 40% first, 40% last, 20% middle | Journeys where both the start and the close matter | Fixed weights that may not match reality |
| W-shaped | Weighted to first touch, lead creation, and close | B2B funnels with a clear lead stage | Needs reliable lead-stage data in your CRM |
| Data-driven | Based on your own conversion paths | Accounts with steady conversion volume | Only sees its own platform’s touchpoints |
How to Choose the Right Attribution Model
There’s no single correct model. The right one depends on how your customers buy. Start with these questions:
- How long is your sales cycle? For same-day purchases, last-touch or time-decay is usually enough. For journeys that take weeks or months, use a multi-touch or data-driven model.
- How many channels are involved? If most sales come from one or two channels, a simple model works. If people move between paid social, search, email, and events, you need a model that shares credit.
- Do you have a clear lead stage? If marketing hands leads to a sales team, a W-shaped model shows which campaigns create real pipeline, not just traffic.
- How much conversion data do you have? Data-driven models need volume. With only a handful of conversions a month, a rule-based model you understand is more trustworthy than an algorithm guessing from too little data.
- What decision are you making? Use first-touch to judge awareness spend and last-touch to judge closing tactics. Comparing two or three models side by side is often more useful than committing to one.
Why Your Data Matters More Than the Model
Every model on this page divides credit for the conversions it can see. If your tracking misses visits, your platforms double-count, or your forms don’t capture where leads came from, every model will give you a confident answer built on incomplete data.
The fix is to connect platform data to your CRM, so you can measure revenue and not just form fills, and keep it all in one database where you can apply whichever model you choose. We walk through that setup, step by step, in Marketing Attribution in 2027: Build a System That Proves Real ROI.
Key Takeaways
- An attribution model decides how credit for a conversion is split, and that split drives where your budget goes.
- Single-touch models are simple but one-sided: first-touch favors awareness, last-touch favors closing channels.
- Multi-touch models share credit by fixed rules, while data-driven attribution learns from your own conversion paths.
- Google Ads and GA4 now offer only data-driven and last-click, so comparing other models means running them on your own data.
- Choose by sales cycle, channel mix, and data volume, and fix your tracking before you trust any model.
Want to Know Which Campaigns Really Drive Revenue?
Colibri’s Analytics & AI-Ops team can audit your tracking, connect your platforms to your CRM, and set up reporting that compares attribution models on data you own. Schedule a complimentary call with our team to get started.






