Here's a scenario we see weekly:
A customer clicks your Meta ad on Monday. On Wednesday, they click a Google ad. On Friday, they buy - a $100 order.
You open Triple Whale's dashboard. You're using Triple Attribution because it sounds like the most comprehensive option.
Your Meta revenue shows $100. Your Google revenue shows $100.
Total attributed revenue: $200.
Actual revenue: $100.
You just doubled your numbers without knowing it.
This isn't a bug. It's how Triple Attribution works.
And it's exactly why most e-commerce brands are making budget decisions based on inflated data.
In the 60+ e-commerce accounts we've audited over the past 18 months, ROAS is commonly overstated by 25-40% because of this single mistake.
Brands scale channels that look profitable on paper but are actually bleeding money.
Here's the thing: the fix takes 30 seconds. You just need to know which model to use when.
Three models. Three purposes. Zero confusion.
ALLOCATE
(Linear Paid)
Budget Decisions
/\
/ \
/ \
/ \
/ \
/ \
ANALYZE ----/------------\---- AUDIT
(Linear All) (Triple Attribution)
Full-Funnel Platform Comparison
Think of Triple Whale's attribution models as three different lenses for looking at the same data. Each lens answers a different question.
Use the wrong lens, and you'll get the wrong answer.
Purpose: Budget decisions across paid channels
How it works: Divides conversion credit EQUALLY among paid touchpoints only.
Example:
When to use it:
The key insight: This is the ONLY model that gives you true budget allocation data. Everything else over or under-counts. (Yes, really.)
EcoBio Boutique used Linear (Paid) to discover Facebook was their primary first-touchpoint for converting customers.
Based on this, they adjusted Pinterest to lower-funnel campaigns only - resulting in a 33% decrease in cost per acquisition.
Purpose: Full-funnel marketing analysis
How it works: Divides conversion credit EQUALLY among ALL touchpoints - paid, organic, email, direct.
Example:
When to use it:
The key insight: This model shows you how non-paid channels support your paid efforts. If you're only looking at paid attribution, you're blind to the channels doing 30-40% of the conversion work.
Many brands discover their email list is doing heavy lifting they weren't measuring.
A customer might click a Meta ad, then convert through email - but if you're only using Linear (Paid), you'd give Meta all the credit.
Purpose: Platform comparison ONLY
How it works: Gives EACH platform 100% credit for conversions they touched (last click per platform).
Example:
When to use it:
When NOT to use it:
The key insight: Triple Attribution exists for one reason - to compare apples-to-apples with individual platform dashboards.
Both Meta Ads Manager and Google Ads give themselves 100% credit for conversions. Triple Attribution does the same, which is why the numbers will match more closely.
But the moment you sum these up across channels, you're counting the same orders multiple times.
One of the most common questions: "Why doesn't Triple Whale match what Meta shows?"
There are four reasons your numbers will always be different - and understanding them is the difference between confusion and clarity.
| Platform | Default Window | Extended Option |
|---|---|---|
| Triple Whale | 28 days | Up to 365 days |
| Meta Ads | 7-day click | 1-day view, 7-day click max |
| Google Ads | 30 days | Up to 90 days |
Triple Whale will capture conversions that Meta "forgets" after 7 days. If a customer clicks your Meta ad, waits 14 days, then buys - Meta won't count it. Triple Whale will.
Fix: When comparing to Meta, set Triple Whale to a 7-day window.
Meta counts "saw ad → purchased within 1 day" as a conversion. Triple Whale's pixel is click-only - it can't track views.
This means Triple Whale will typically show LOWER Meta revenue than Meta reports. That's not Triple Whale under-counting - it's Meta including view-through conversions you can't independently verify.
What to do: Understand the gap is view-through attribution. Based on our account audits, view-through typically accounts for 15-25% of Meta's reported conversions.
Some of that is real brand lift. Some is people who would have converted anyway. The honest answer: you can't know the exact split without running holdout tests.
User clicks on phone, buys on desktop. Meta and Google have better cross-device tracking through logged-in users. Triple Whale may miss these conversions.
What to do: Accept some tracking loss as the cost of privacy-first attribution.
75% of iPhone users opted out of tracking. Meta uses "modeled" (estimated) data for these users. Triple Whale uses actual clicks only.
Post-iOS 14.5, Facebook Pixel capture rates fell from 80-95% to 60-70% of backend sales. That's a 15-25% signal loss on a platform that was already imperfect.
What to do: Don't expect perfect matches. Focus on week-over-week trends, not absolute numbers. If your ROAS dropped 40% this week vs last week, that's real - even if the exact dollar figures are off.
When you need to compare Triple Whale with Meta or Google, follow these steps:
Step 1: Match Attribution Windows
Step 2: Use the Right Model
Step 3: Understand the Discrepancy
Step 4: Switch Back for Budget Decisions
Use this when you're unsure which model to pull:
"I'm deciding how to allocate budget across channels" → Linear (Paid)
"I want to understand how organic/email supports paid" → Linear (All)
"My Meta numbers don't match Triple Whale - why?" → Triple Attribution (for comparison only)
"I need total revenue for my dashboard" → Linear (Paid) OR Linear (All) → NEVER Triple Attribution
"I'm scaling a channel and need accurate ROAS" → Linear (Paid)
"I'm presenting to stakeholders and need clean numbers" → Linear (Paid) for paid analysis → Linear (All) for full marketing picture
| Model | Credit Method | Use For | Duplication Risk |
|---|---|---|---|
| ALLOCATE (Linear Paid) | Equal across paid | Budget decisions | None |
| ANALYZE (Linear All) | Equal across all | Full-funnel analysis | None |
| AUDIT (Triple Attribution) | 100% to each | Platform comparison | HIGH |
For operating your business: Linear (Paid) or Linear (All)
For auditing platform discrepancies: Triple Attribution
Never: Sum up Triple Attribution across channels for "total" revenue
(Yes, we've made some of these ourselves. That's how we learned.)
Mistake 1: Using Triple Attribution on the "All Channels" dashboard
Mistake 2: Expecting Triple Whale to match Meta exactly
Mistake 3: Ignoring Linear (All) entirely
Mistake 4: Making budget decisions on default settings
Use this before your next budget meeting:
We learned attribution the expensive way. (Most people do.)
Mistake: Trusting Triple Attribution as our default for 4 months. We scaled a Meta campaign from $5K to $15K/month based on Triple Attribution ROAS. The numbers looked great.
Profit didn't follow. When we switched to Linear (Paid), the real ROAS was 38% lower than what we'd been reporting. We'd been scaling based on double-counted revenue the entire time.
Mistake: Ignoring Linear (All) completely. For our first year using Triple Whale, we never checked Linear (All).
Turned out email was assisting 30%+ of our paid conversions. We almost cut the email program because it "wasn't driving revenue" on its own.
Linear (All) showed the real picture. Email was the silent closer we almost killed.
Mistake: Obsessing over exact number matches between platforms. We spent weeks trying to make Triple Whale match Meta exactly.
It won't. It can't. Different data access, different attribution methods, different tracking limitations.
The fix was accepting directional accuracy over false precision. Week-over-week trends matter more than matching to the dollar.
Attribution isn't about finding the "right" number. It's about using the right lens for the right question.
Linear (Paid) tells you where to spend money. Linear (All) tells you how your full funnel works. Triple Attribution tells you why platform dashboards show different numbers.
Get these three straight, and you'll stop making budget decisions on inflated data.
That's it. No fancy software upgrade needed. Just the right model for the right question.
This playbook is part of the Tegra Growth toolkit. Questions? tegra.co