The guide every DTC brand owner needs to read before making budget decisions
This is the number one reaction when brands first connect an attribution tool. Your Meta Ads Manager shows 4.0x ROAS. ConversionPath shows 2.5x. Your stomach drops.
Nothing is wrong. This is exactly what accurate attribution looks like.
Here's what's happening: Meta takes full credit for every sale where the customer clicked any Meta ad in the last 7 days. But what if that customer also clicked a Google ad, opened an email, and then bought? Meta still says "I drove that sale." Google says the same. Klaviyo says the same.
ConversionPath sees the full journey and splits the credit fairly. That's why Meta's number goes down - it was never that high in reality.
Typical example
| Channel | Platform Says | Reality |
|---|---|---|
| Meta Ads | 4.0x ROAS | 2.5x ROAS |
| Google Ads | 3.2x ROAS | 2.8x ROAS |
| Klaviyo | $85K revenue | $52K revenue |
Platform numbers add up to more than your actual Shopify revenue. ConversionPath numbers match exactly.
This is the mistake that kills growth. A brand sees Meta drop from 4x to 2.5x in their attribution tool, panics, and cuts spend. Then Google performance drops two weeks later. Then overall revenue drops.
Meta creates demand. Google captures demand. Cut Meta and you cut the supply of new customers that Google was converting.
Think of it this way:
The Restaurant Analogy
Meta is the billboard that brings people to your suburb. Google is the sign on your door. Email is the waiter who brings them back. If you take down the billboard because "people are coming through the door anyway," eventually the foot traffic dries up - and the door sign stops working too.
Rule #1: Use ConversionPath for decisions. Use platform data for optimisation.
ConversionPath
The Judge
Platform Data
The Driver
You don't let the judge drive the car. And you don't let the driver decide the verdict.
When Meta looks bad in ConversionPath, ask these three questions before making any changes:
1. Is my blended ROAS healthy?
IF YES
Yes → keep scaling. The system works.
IF NO
No → diagnose further (see below).
2. Is total revenue growing with spend?
IF YES
Yes → Meta is doing its job even if its ROAS looks low.
IF NO
No → check creative fatigue, offer, or landing page.
3. Are new customers increasing?
IF YES
Yes → Meta is acquiring. Don't cut it.
IF NO
No → Meta might need new audiences or creative refresh.
If all three are "yes" - ignore the low Meta ROAS. Your marketing system is working.
Stop obsessing over individual channel ROAS. These are the metrics that tell you if your business is healthy:
Blended ROAS
Total revenue / total ad spend across ALL channels. The single most important metric for scaling decisions.
Target
3.0x+ for most DTC brands
MER (Marketing Efficiency Ratio)
Ad spend / revenue. Shows what percentage of every dollar earned goes to ads.
Target
Under 25% is healthy
NC-ROAS (New Customer ROAS)
Revenue from first-time buyers / ad spend. Shows if you're profitably acquiring new customers.
Target
2.0x+ means you're growing sustainably
CAC (Customer Acquisition Cost)
How much you spend in ads to get one new customer.
Target
Depends on your AOV and margins
Blended ROAS is above target AND revenue is growing AND new customer acquisition is healthy. Increase spend 15-20% at a time.
Blended ROAS is at target but not growing. Maintain spend, focus on creative refresh and audience testing. Don't scale into efficiency issues.
Blended ROAS is below target but new customers are still coming in. Fix creatives, test new angles, review landing pages. Don't cut spend yet.
Blended ROAS is declining AND revenue is flat or dropping AND no new customer growth. Reduce spend, pause underperformers, restructure campaigns.
No. And that's the point.
If ConversionPath showed the same numbers as Facebook, you wouldn't need it. The whole value is that it tells you something Facebook can't - the truth about how all your channels work together.
What you should expect:
Use platform data to drive. Use ConversionPath to navigate.
Optimise your ads inside Meta and Google - that's where you test creatives, tweak audiences, and manage bids. Then come to ConversionPath to understand the big picture: is the whole system profitable? Should you spend more or less? Which channels actually matter?
The brands that win aren't the ones with the highest Meta ROAS. They're the ones who understand how all their channels work together - and make decisions based on reality, not inflated platform numbers.
You've probably seen ads claiming "60% of your ad revenue is invisible" and pitching "server-side tracking not reliant on cookies." That's marketing spin — but it's pointing at a real problem, and we solve it.
Your browser pixel alone loses 10–25% of conversions to iOS ATT, Safari ITP, ad blockers, and incognito. Server-side tracking is how that data is recovered.
What "server-side" actually means. Instead of relying on a script in the customer's browser to tell Meta and Google about a purchase, your server sends the event directly to the ad platform's API using hashed email, phone, and click IDs. No browser involvement. No ad blocker can touch it. iOS can't block what it never sees.
Browser pixel only
Server-side + pixel (us)
What we do server-side, today:
Meta Conversions API (CAPI)
Every Shopify order fires a server-to-server Purchase event to Meta with hashed email, phone, first name, last name, city, and fbclid. Survives iOS 14.5+ ATT, Safari ITP, and ad blockers — Meta matches the conversion back to the ad click using the hashed PII, not the browser.
Google Enhanced Conversions
Same server-to-server flow to Google Ads. Hashed PII + gclid lets Google re-attribute conversions the pixel missed — declared as FIRST_PARTY data so Google treats it as high-trust signal for Smart Bidding.
Shopify orders/create webhook
HMAC-verified webhook is our source of truth for orders — not the browser. If the pixel fails entirely, every order is still captured and attributed from Shopify's server-side payload.
Klaviyo server-side click ingestion
We pull email clicks directly from Klaviyo's Events API. iOS Mail Privacy Protection and Apple Mail pre-fetching can't hide these — Klaviyo logs the click server-side before any browser script runs.
IP + email cross-device matching
When a customer clicks an ad on mobile and buys on desktop from the same home WiFi, we match by IP + hashed email server-side. No cookie needed — the browser pixel can't do this because the browsers are different devices.
Shopify Web Pixel (sandboxed)
Our browser pixel runs inside Shopify's Customer Events sandbox — no URL for ad blockers to match against. Sees checkout.shopify.com where every ScriptTag-based competitor is blind.
The honest part no one else will tell you. "No cookies at all" is marketing fiction. Every serious attribution tool — Triple Whale, Northbeam, Rockerbox, wetracked.io, and us — uses a first-party cookie to link pageviews to a session before the customer enters their email. What's actually being killed by browsers is third-party cookies (the tracking kind), and none of us use those. Our _cpid cookie is first-party, 365-day, and set on your own storefront — not affected by the third-party crackdown.
One gap, honestly disclosed. Capturing the first-touch UTM parameters on a customer's landing page still requires the browser pixel to fire once. If they block it entirely, we know which ad drove the final conversion (via CAPI/Enhanced Conversions) but not the full pre-click journey. No attribution tool solves this — it's physics.
Same tracking technology as the enterprise tools. More features. A fraction of the price.
14-day free trial. No credit card required.
ConversionPath - Stop guessing. Start knowing.