29

This post is in reference to the following blog post: Report this Ad [beta] provides more control over your advertising experience

To start off, a quick introduction of myself; my name is Jascha Drel, Product Manager for Advertising at Stack Overflow. In this role, I am part of Teresa Dietrich’s organization.

Our team is continuously making improvements to the ads experience on our sites. One of the changes we’re excited about sharing is the new Report this Ad feature now available on some of our sites. We want to continue to give our communities more native control over their ad experience. We needed a quantifiable way to receive feedback and this feature will allow that.

We enforce strict guidelines on the types of advertisers we work with but cannot fully control what is shown due to how advertising delivery works. However, being able to hear from our users about problematic ads will be a huge benefit as we can address those problems faster and with more accuracy. The Report this Ad feature will allow us to quickly see and investigate ads reported by you which matters to us because we want to hear your thoughts.

Furthermore, this new feature ensures that the members of our communities have a built-in way to report ads they consider problematic when we test any new ad formats/solutions.

We are launching this feature on 5 sites initially (Server Fault, Database Administrators, Arqade, Game Development, and Unix & Linux) to get a good understanding of what we can expect in terms of numbers of reports whenever launching this to the entire network. Prior to launching it network-wide, we’ll share general results we’ve seen here.

What happens once an ad is reported?

All reports will be consolidated into a dashboard that the team will look at several times per week to spot irregularities in terms of the total number of reports, report categories, and on an advertising campaign basis. We will not be replying to individual reports, but we will reach out to our advertisers, to share feedback or notify them if the ad being reported does not fit our guidelines. We may also evaluate our policies in response to large numbers of reports not covered by them. Since we don’t have a baseline for the number of ads that are going to be reported using this new feature, we will consider reevaluating the process if the current process doesn’t scale.

  • 4
    Thank you, Jascha. Do you have an idea how long this five-site beta is going to run, or will it be terminated based on other conditions? (6-8 weeks, I assume?) – nitsua60 Mar 24 at 21:08
  • 1
    Just curious, why don't you use your name and avatar here on Meta? We won't bite. :) – Shadow Wizard is Ear For You Mar 24 at 21:44
  • 7
    Welcome to Meta SE :-) And thanks for this! A lot of people have been complaining about inappropriate ads; it's good to know they've been heard. – Rand al'Thor Mar 24 at 21:46
  • 4
    Could you comment on the state of the ads experiments on the wider network, most of the seriously problematic ads were part of that. My impression is that this was put on hold a while ago, but I haven't checked ads in a while. What are the plans for expanding ads across the network, which does mean the inclusion of potentially more problematic ad providers? – Mad Scientist Mar 24 at 22:09
  • 4
    @MadScientist Yes, that experiment is over - currently we don't have a set, planned, way to run ads across the network, we're still working on that. The idea is exactly to test this feature and see how it can help us with problematic ads before we go there. – Cesar M Mar 24 at 22:21
  • 3
    @nitsua60 The plan is to run this for a week or two on these 5 sites if no blockers appear from those weeks of test we’ll roll it out to the entire network @ ShadowWizardisEarForYou Hahaha, I was expecting anybody to bite on Meta SE ;). I just prefer to use a nickname, and I did notice today that I should get an updated avatar for myself 😅 @ Randal'Thor Happy to hear you like the feature :) – JD-Stack Mar 24 at 22:21
  • 3
    @CesarM the reporting feature does leave a lot to chance, ad selection is extremely dependent on location and other stuff. What would be really nice would be a way to easily get an overview of ads that have been shown to users, even if you're not targeted yourself by those ads. I assume this is impossible due to the way ad networks work, I just mention it in case I'm wrong about that. – Mad Scientist Mar 24 at 22:33
  • 1
    Can we report blog posts that are ads being mislabeled as announcements? – curiousdannii Mar 24 at 23:38
  • 3
    Uhm. Probably not. 🤣. That's probably a long, nuanced conversation for meta – Journeyman Geek Mar 24 at 23:59
  • 2
    @MadScientist Talked to the team, and yeah, an overview of the ads is either very difficult or outright impossible, one due to how ads work, and secondly, due to the volume we serve, it's one of the reasons we built this feature, it can't simply be checked all at once :( – Cesar M Mar 25 at 19:26
9

Thanks for this feature, hopefully that will put an end to the out there.

Purely out of curiosity, how were the test sites chosen? I don't see any obvious correlation between them (though maybe that's the point), and I'm a bit surprised SO is not in it; I'd expect way more feedback from SO, given the traffic there (though Server Fault is quite busy as well).

In particular I can't help but notice that the sites where the spammy ads apparently pop the most aren't included in that list. Going by the questions in the tag above and this answer, they would be Travel, Workplace, Music.SE and SFF. Is them not being part of the experiment a deliberate choice?

|improve this answer|||||
  • 5
    "to get a good understanding of what we can expect in terms of numbers of reports whenever launching this to the entire network" - it seems like they don't want massive amounts of traffic/feedback at first. They're starting with a small, controlled experiment before expanding to the whole network - that's the point. (Plus, don't complain about non-SO sites having something special happen to them ;-) SO gets enough special attention as it is!) – Rand al'Thor Mar 24 at 22:11
  • 4
    @Jenayah, I think Rand al'Thor covered our thoughts well :). The sites were chosen based on the combination of traffic numbers and the amount of non-house ads we’re currently running there. We want to figure out how the feature works and how often it is used. These sites should give us a good indication. – JD-Stack Mar 24 at 22:16

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .