The Problem
Most marketers are flying blind. They're spending thousands on ads but have no idea which channels actually drive revenue. They rely on platform dashboards that lie, give credit to the wrong touchpoints, and make bad campaigns look good while killing winners.
The culprit? Attribution models. And most people are using the wrong one—or worse, they don't even know which one they're using. If you're making budget decisions based on last-click attribution (the default in most tools), you're systematically under-investing in top-of-funnel channels and over-investing in bottom-funnel ones.
This isn't theoretical. We've seen companies cut profitable Facebook campaigns because "they weren't working," only to watch their overall revenue drop 30% two months later. The ads were working. The attribution model was lying.
What Is Attribution?
Attribution is how you assign credit for a conversion across multiple touchpoints. A customer rarely sees one ad and buys. They see a YouTube ad, visit your site, leave, see a Facebook retargeting ad, Google your brand, click a paid search ad, and then convert. Which ad gets credit? That depends on your attribution model.
Different models split credit differently. And the model you choose changes everything—which campaigns look profitable, where you allocate budget, what you optimize for, and ultimately, whether you scale or stall.
The Main Attribution Models
Last-Click Attribution
Last-click gives 100% of the credit to the final touchpoint before conversion. If a customer sees five ads but converts after clicking a Google search ad, Google gets all the credit. Facebook, YouTube, email—nothing. Zero credit.
This is the default in Google Ads, Facebook Ads, and most analytics tools. And it's wildly misleading. Last-click systematically undervalues awareness and consideration channels (like Facebook, YouTube, display) and over-credits intent channels (like branded search and retargeting).
The result? You kill top-of-funnel campaigns that are actually creating demand, and you over-invest in bottom-funnel campaigns that only capture demand someone else created. It's a recipe for short-term wins and long-term stagnation.
First-Click Attribution
First-click does the opposite—it gives 100% of the credit to the first touchpoint. If someone discovers you through a Facebook ad, that ad gets all the credit, even if they later clicked five other ads before converting.
First-click overvalues top-of-funnel awareness channels and ignores everything that happened after. It's useful if you're trying to understand how people discover you, but it's terrible for budget allocation because it ignores the nurturing and retargeting that actually closed the deal.
Linear Attribution
Linear splits credit equally across all touchpoints. If there are five touchpoints, each gets 20% credit. It's more balanced than last-click or first-click, but it's still naive—it assumes every touchpoint is equally important, which is rarely true.
A customer seeing your brand for the first time is not the same as a customer clicking a retargeting ad after visiting your pricing page. Linear attribution treats them the same, which distorts reality.
Time-Decay Attribution
Time-decay gives more credit to touchpoints closer to the conversion. The first touchpoint might get 5% credit, the second 10%, the third 20%, and the last 65%. It acknowledges that later touchpoints often matter more, but it's still arbitrary—the decay rate is just a guess.
Time-decay is better than last-click for understanding the full funnel, but it still overvalues recency and undervalues early awareness.
Position-Based (U-Shaped) Attribution
Position-based gives 40% credit to the first touchpoint, 40% to the last, and splits the remaining 20% across everything in between. The logic: the first touchpoint creates awareness, the last closes the deal, and the middle just nurtures.
This is more realistic than linear or time-decay, and it's a decent compromise if you can't use data-driven attribution. But it's still a guess. It assumes first and last are always the most important, which isn't always true.
Data-Driven Attribution
Data-driven attribution uses machine learning to analyze your actual conversion data and assign credit based on what actually drives conversions. It looks at thousands of conversion paths, identifies patterns, and gives credit to the touchpoints that statistically increase conversion likelihood.
This is the gold standard. It's not perfect—it requires enough data to train the model (typically 400+ conversions per month)—but it's the only model that adapts to your business instead of forcing you into arbitrary rules.
Google and Facebook both offer data-driven models, but they only look at their own data. If you want true cross-platform attribution, you need a third-party tool like Triple Whale, Hyros, or Northbeam.
Which Model Should You Use?
It depends on your business and your goals, but here's the hierarchy:
If you have enough data (400+ conversions/month), use data-driven attribution with a third-party tool that tracks across platforms. This gives you the most accurate view of what's working.
If you don't have enough data, use position-based attribution as a starting point. It's not perfect, but it's miles better than last-click and gives credit to both awareness and conversion touchpoints.
Never use last-click alone for budget decisions. It will systematically kill your top-of-funnel and make you over-reliant on branded search and retargeting—which means you stop creating new demand and start cannibalizing existing demand.
Use multiple models in parallel. Don't rely on one view. Look at last-click (to see what closes deals), first-click (to see what creates awareness), and data-driven (to see the full picture). Make decisions based on all three.
Implementation
Start by auditing your current attribution setup. Log into Google Ads, Facebook Ads, and your analytics tool. Check what attribution model you're using. If it's last-click, you're under-crediting awareness campaigns.
Next, switch to position-based or data-driven if you have the data. In Google Ads, go to Tools > Conversions > Attribution Models. In Facebook, go to Events Manager > Attribution Settings. Choose your model.
Then, re-evaluate your campaigns. Look at performance under the new model. You'll likely find that some "losing" campaigns are actually winners, and some "winning" campaigns are just stealing credit from earlier touchpoints.
Finally, invest in cross-platform attribution if you're serious about scaling. Tools like Northbeam, Triple Whale, or Hyros cost money, but they pay for themselves instantly by preventing you from killing profitable campaigns.
Results
One client switched from last-click to data-driven attribution and discovered their YouTube campaigns—which looked like they had a 5x ROAS under last-click—actually had a 12x ROAS when you counted the full funnel. They tripled their YouTube budget and revenue jumped 40% in two months.
Another client found the opposite: their "high-performing" retargeting campaigns were getting credit for conversions that were going to happen anyway. They reallocated budget to cold prospecting, and while retargeting ROAS dropped, overall revenue grew 25% because they were creating new demand instead of just harvesting it.
Attribution isn't sexy. But it's the difference between guessing and knowing. Fix your attribution, and you'll stop killing winners and start scaling with confidence.