[zztop44]

It does have findings, it’s just a long article. Maybe longer than it needs to be. Looks like their SERP ads CPC increased by 30% and conversions decreased, display ad conversions plummeted (partially offset by cheaper CPMs) and YouTube ads switched to optimize for watch time instead of conversions seemed to perform better and lead to cheaper conversions.

[c0l0]

The one number that should be of interest to anyone who's not a professional "growth marketer" is missing, though, don't you agree?

I skimmed the whole article, read your posting, and even though I know some of these words, I still have no idea if the decision was hurtful to the business, or if it did not move the needle at all, or if it even was a net positive, all things considered.

[the_mitsuhiko]

> I still have no idea if the decision was hurtful to the business, or if it did not move the needle at all, or if it even was a net positive, all things considered

It's incredibly hard to tell. For us the goal of eliminating Cookies was important given the stance we have on privacy so everything went from there. The folks working in marketing for sure were not happy with the directive as it makes their job much harder.

But the world is still spinning, even without cookies. That's enough to call this a success.

[justinclift]

> It's incredibly hard to tell.

Heh, that's very much like the old "... and nothing of value was lost" then.

[Root_Denied]

Yeah, the fact that they don't have a clear indication of impact means one of two things:

- Cookies weren't doing a whole lot for them to begin with, and removing their use had negligible impact.

- The other work their marketing department did to try and compensate for discontinuation of cookie usage interfered with the test, making the results useless for evaluating the value of cookies.

I'm leaning towards the second one based on the response above you.

[brookst]

Agree, but your second bullet is kind of the same as the first, right? It’s saying that on the whole, inclusive of changes to marketing processes, removing cookies didn’t impact bottom line enough to be measure able as positive or negative.

[Root_Denied]

There's definitely a difference between "Cookies provide X value" and "We were able to mitigate the loss of cookies by doing A, B, and C - but we don't know the value of doing A/B/C individually."

If the point is to quantify the value of cookies then you need to measure that as an independent variable. Attempting to compensate with other actions, each of which you also don't know the independent value of, means you will be unable to quantify the true value of cookies or the value of your other actions.

What if none of the tools they're using, including cookies, provide any measurable value? How would you know? What if the value of cookies (X) and item A is positive in terms of impact, but B and C are both negative, but not enough to offset A so it comes out in a wash?

[yencabulator]

My personal favorite possibility: Nobody is actually able to connect advertising spend to incoming revenue in any rigorous way.

[hnbad]

This seems to be very much the case for behavioral ads as far as I can tell. At some point Google et al convinced everyone that tracking the user's every move across the Internet made for better conversions than just providing ads that are relevant to the content they appear alongside and everyone just nodded along. Given that we now know that Meta (née Facebook) blatantly lied about the effectiveness of video content (which single-handedly killed various news sites who made hard pivots based on these claims) I'm not convinced behavioral ads really do perform that much better. The biggest value add of Google ads was that it provided some safeguards against abuse and a reasonable expectation of safety and moderation.

[bawolff]

Well that's fair, i do feel a little let down by a lack of answer to "here's what happened" part of the headline, which largely went unanswered.

[the_mitsuhiko]

Marketing budget went into marketing things. Results were in line with expectations :P

[zilti]

The same stance on privacy that somehow makes it okay for you to scan all your users' code for AI training without consent?

[the_mitsuhiko]

I'm assuming you are referring to our terms of service change. The short answer is: we are not. We posted an update to the blog post here that gives more context: https://blog.sentry.io/ai-privacy-and-terms-of-service-updat...

[dxdm]

Could you elaborate on what they are doing?

[zilti]

Sentry scans all data sent to it to train their problem solving assistant, no matter if you use it or not, and there is no opt-out.

[the_mitsuhiko]

That is incorrect. We are not doing that.

[hnbad]

It sounds like you're still using tracking, just without cookies. I understand the push to eliminate cookies was motivated by Google announcing it would force the decision eventually but it sounds like you're still using alternative means of tracking and identifying users.

Laws like ePrivacy in the EU do indeed have specific provisions regarding cookies but e.g. the GDPR is much broader than that and would still apply. How truthful is it that eliminating cookies is motivated by a strong stance on privacy rather than just getting a head start in marketing instead of having to scramble when Google pulls the plug? It doesn't sound like you reduced the tracking and behavioral analysis beyond what was technically unavoidable?

[ToucanLoucan]

> I skimmed the whole article, read your posting, and even though I know some of these words, I still have no idea if the decision was hurtful to the business, or if it did not move the needle at all, or if it even was a net positive, all things considered.

I mean, based on my interactions with marketing people, they often don't really know if much of what they've done has helped the business at all, and the majority of their work (apart from actually creating marketing copy and interacting with customers) seems to revolve around figuring out how to attribute booms in business to their previous campaigns while building plausible deniability for inevitable busts. Don't get me wrong, not ALL of it is completely incomprehensible: email and referrer links are pretty straightforwardly calculable in terms of their impact; but things like "brand awareness" campaigns are nigh impossible to actually gauge the impact of.

[ameister14]

I never understood why businesses accepted this. At least with digital marketing, it is possible to track pretty much everything and connect marketing campaigns with concrete results. It's actually not impossible to gauge the impact of a brand awareness campaign if you run the campaign properly using good principles of experiment design, especially if you associate it with keywords and track those or are using search ads.

[njitram]

This is probably whats going to happen everywhere, Google will be even more dominant in adtech with 3rd party cookie removal. Everyone moving to 1st party advertising (SERP, Youtube ads), display ads are race to the bottom without targeting..