High Signal Podcasts Evidence ledger
Method
Browse
← Back to evidence

Evidence receipt / evaluation

Published · transcript-backed

Yuriy Timen: evaluation

1 Sept 2022 Lenny's Podcast How to grow a subscription business | Yuriy Timen (Grammarly, Canva, Airtable)

“It was only good for correlative insights. And the only way to determine causality is through real controlled experiments, randomized control experiments through incrementality testing, which is typically really hard to do cleanly and also companies have always been often wary about doing it because you have to turn off a channel potentially in a key demo and you're like, "Yeah.”

— Yuriy Timen

Source trail

Everything needed to verify it.

Speaker
Yuriy Timen
Attribution
Verified speaker
Claim type
evaluation
Recorded
1 Sept 2022
Publisher
Lenny's Podcast

Transcript context

…You're just doing- Yeah, I think it's useful if you're spending a considerable amount, what's considerable, I'd say worth of a hundred thousand a month update media. And if you have some level of channel complexity, so you're not just like a Google Go or a Facebook, but maybe you're on three plus the channels. Then I think it still makes sense. The other ones in the incremental space, they have very different methodologists actually, because at end of the day, this might be obvious to folks, but maybe some will find value. Click based attribution or the digital attribution were all fawning over cookie based and click based, a real parameter based attribution. It never demonstrated a causal relationship between our media spend and business results. It was only good for correlative insights. And the only way to determine causality is through real controlled experiments, randomized control experiments through incrementality testing, which is typically really hard to do cleanly and also companies have always been often wary about doing it because you have to turn off a channel potentially in a key demo and you're like, "Yeah. " The benefit is to learning of whether it's actually incremental, but the cost or the sales that I will lose today. But the only way to really know how effective your paid media is through ongoing incrementality testing. So there are two companies that are addressing that. Two that I'm excited about. One is measured, can be found, measured.call, amazing domain name. Amazing domain name, go with that.…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.

Search evidence