- Inceptly
- Posts
- There are two conversion numbers on Demand Gen - and they're not measuring the same thing.
There are two conversion numbers on Demand Gen - and they're not measuring the same thing.
If you've ever opened a YouTube or Demand Gen report and seen Conversions (Platform Comparable) sitting higher than Conversions, you've probably wondered which one to trust. Both are correct. They're just measuring different things. |  Author: |
Want to brainstorm with us on new ways to scale your business with YouTube Ads (and other performance video platforms)?
Join us for a free YouTube ad brainstorming session here:
Not all conversions look the same
On Demand Gen, a conversion can happen in three different ways:
Click-through (CTC) — Someone clicked your ad, then converted. The classic path everyone optimizes for.
Engaged-view (EVC) — Someone watched your video meaningfully, didn't click, but converted later.
View-through (VTC) — Someone saw your ad, didn't click or engage, but converted anyway.
This is what makes video different from Search. The path to conversion isn't always a click. On YouTube and connected TV, influence happens before the click — or instead of it entirely.
So why are there two columns?
Conversions uses Data-Driven Attribution across all your Google campaigns. Credit is distributed across every Google touchpoint in the path — Search, Shopping, Performance Max, Demand Gen. When a Search campaign closes the conversion, it gets the credit. Demand Gen's contribution upstream gets diluted. This is your optimization column — it drives Smart Bidding, CPA, and ROAS calculations.
Conversions (Platform Comparable) work differently. It does two things the standard column doesn't: it isolates Demand Gen from the rest of the Google ecosystem, giving full credit to the last Demand Gen touchpoint in the path — and it includes view-through conversions, which the standard column usually excludes by default. The result is a measurement model that aligns with how Meta, TikTok, and other social platforms report performance.
That's why Platform Comparable tends to run higher: it's capturing both the view-driven conversions and the Demand Gen influence that DDA redistributes to other campaign types.
How to use both
Think of them as answering different questions:
Conversions → How is this campaign performing within the broader Google ecosystem?
Conversions (Platform Comparable) → What's Demand Gen's standalone contribution, including users who converted after seeing — but not clicking — the ad?
If you're benchmarking Demand Gen against paid social, Platform Comparable is the right column — it's built exactly for that comparison. If you're evaluating overall Google account efficiency, Conversions is the one to use.
And practically speaking, it's also the number Google's own teams will reference when they review your campaign performance. Good to know before that call.
The gap between the two columns isn't a discrepancy to fix. It's a signal worth understanding.
Want to brainstorm with us on new ways to scale your business with YouTube Ads (and other performance video platforms)?
Join us for a free YouTube ad brainstorming session here:
 | Bobo Slijepcevic, Director of Media Buying & Analytics From black holes to ad clicks, Bobo took a cosmic leap from astrophysics to analytics. After years of teaching physics and explaining why Schrödinger’s cat is both alive and dead (but definitely not a good pet), he joined Inceptly in 2022. Now, he spends his days decoding YouTube metrics and buying media like a physicist shops for particles — with precision, curiosity, and the occasional caffeine boost. |
🎯 Inceptly’s top picks:
Essential reading you can't afford to skip
AI-generated ads are everywhere right now… but most of them feel cheap.
Not these.
Tai Chi Walking is running conversational YouTube ads powered by AI visuals that are both cost-effective and surprisingly persuasive.
One ad starts with a simple line:
“Daughter’s wedding is in a month…”
From there, it turns into a perfect direct-response structure: emotional hook → objection handling → bold efficiency claim → clear CTA.
We break down exactly why this works.
This “fake” expert helped scale an ad to $3.1M in spend.
The presenter?
An AI-generated avatar in a suit speaking at a conference.
But the real story isn’t the avatar.
It’s the direct-response structure behind the ad:
problem agitation → pattern interrupt → unexpected solution → authority → scarcity.
In the breakdown, we analyze why this ad works and what marketers can steal from it.
Get in touch with us by responding to this email or tagging us on LinkedIn or Instagram and sharing your thoughts. Your feedback helps us keep our newsletter relevant and interesting.
