Clicks + views
Credit for the ads people watched.
One day under Clicks + Deterministic Views
Orders Clicks only gave to unpaid touches
8 orders credited to direct, organic, email or other unpaid touches.
View-through purchases reported
- Meta ad A3 orders
- Meta ad B1 order
- Google Ads campaign1 order
f = min(1, 5 / 8) = 5/8. Each of the 8 orders gives 5/8 of its credit to the ads with views and keeps 3/8 on its unpaid touches. The day moves 5 orders to ads, never more than the platforms reported and never more than the 8 orders in the pool.
Worked example: 8 unpaid orders, 5 reported view-through purchases.
Many buyers see a video, never click, and come back later through search or by typing your address. Click-based reports give that sale to whatever came last. The views models give it to the ad that started it, without counting any sale twice.
Shopify
Meta
Google Ads
TikTok
YouTube
Instagram
Snapchat
Pinterest
Microsoft Advertising
Amazon Ads
AppLovin
Reddit
LinkedIn
X
The Trade Desk
Vibe
MNTN
OpenAI Ads
Klaviyo
Where views come from
From the platforms you already connect.
View data comes in with spend, over the same read-only connections, so there is nothing new to set up.
Meta1-day view-through purchases for each Facebook and Instagram ad, by day.
Google AdsView-through conversions for each ad and day, including YouTube, and for Performance Max by campaign.
TikTokView-through purchases for each ad and day.- A measurement partnerViews a clean room or a platform's matched export ties to your orders, for any platform.
Order-matched views
Views matched to the exact order.
Views matched to orders by a partner
Order A
- Meta ad click1/2
- Meta view1/4
- Meta view1/4
Order B
- Direct visit0
- Meta view1/2
- TikTok view1/2
Each paid click is one share and each platform with views is one share; a platform's share is split across its views. Once a view counts, unpaid touches get nothing.
Rules: views up to 24 hours before the order; a view within 30 seconds of a paid click on the same platform is that click.
When a measurement partner can say which orders followed a view, each of those orders is credited from its own views. A view up to 24 hours before the order counts, and a view within 30 seconds of a click on the same platform counts as that click.
MTA has no partner feed of its own yet. Until a partner sends views for your store, every platform uses the day-level method above.
Clicks + Modeled Views
Views estimated from your own history.
Clicks + Modeled Views, fitted per brand
unpaid orders(day) =
weekday + trend
+ Σ coefficient(platform) × impressions(day and day before)
coefficients ≥ 0 · video views as a second input where reported
- 1Weekday and trend take out the regular rhythm of your unpaid orders.
- 2What is left is shared out by each platform's impressions, and no platform can take credit away.
- 3A day's view credit stops at the number of unpaid orders that day.
Refitted nightly over up to 120 days. Needs 28 days of orders with pixel data.
MTA learns from your own store how the daily sales no ad click claimed move with each platform's impressions, over up to 120 days, and credits views from that. Until your pixel has 28 days of orders, it uses each platform's own reported view rate instead.
It shows how sales move with impressions. It cannot prove that a view caused a sale, and it says so.
Honest totals
Blank where the data is thin, whole where it counts.
One day, Clicks + Deterministic Views
| Row | Impressions | Orders | |
|---|---|---|---|
| Campaign: Prospecting | 48,210 | 31.4 | Shown |
| Ad: Spring video | 46,900 | 30.8 | Shown |
| Ad: Test cut B | 640 | — | Blank under 1,000 |
Blank means held back, not zero. The ad set and campaign above a thin ad still show their credit, so ad rows can add up to less than their campaign.
Illustrative rows. Spend, impressions and visits stay on blank rows.
An ad with fewer than 1,000 impressions on a day shows its spend with its view credit left blank, so a handful of claimed views can't give a tiny ad a huge ROAS.
View credit only goes to sales no paid click already claimed, so every order still adds up to one order and your totals still match Shopify.
Ask it
Answers to the questions you ask every week.
Each card opens the report that answers it.
Are our video ads behind sales that show up as direct traffic?
Switch on a views model and see which ads earn credit from people who watched and bought later.
Clicks + viewsOpen itWould this campaign still look good if we judged it differently?
See the same channels and dates under every attribution model, side by side.
Model comparisonOpen itWhich ads claim more sales in the platform than they actually drove?
Each ad shows the platform's own reported revenue beside the revenue your orders credit to it.
Creative analyticsOpen itIs our Meta spend lifting branded search?
Put Meta spend beside Google brand-search revenue and see how closely the two move together.
Metrics explorerOpen itWhich campaigns brought in new customers last week, and what did each one cost?
New-customer revenue, orders and cost per new customer for every platform, campaign, ad set and ad.
Sales attributionOpen it
Explore
More ways to find what sells
Multi-touch attribution
Every sale counted once and split across the ads that earned it.See multi-touch attributionCreative analytics
Each ad's image or video beside the buyers it brought in.See creative analyticsSales attribution
Spend and real revenue for every platform, campaign and ad.See sales attribution
Numbers that add up.
Three rules every report keeps, checked by automated tests on every change.
1×
Each sale counted once
However many platforms claim an order, its revenue is split across the touches that led to it and never counted twice.
= Shopify
Totals match your store
Orders with no tracked ad get their own Unattributed row, so changing the model moves credit between rows and never changes the total.
1 currency
Spend and revenue in yours
Spend billed in another currency is converted once, at that day's European Central Bank rate, so ROAS never mixes two currencies.
Find out which ads really sell.
Book a 45-minute call with the Growth Dept team. We look at how you measure sales today and show you what MTA would add.
