Signal Tries to Run the Most Honest Facebook Ad Campaign Ever, Gets Banned | Hacker News Reader

Signal Tries to Run the Most Honest Facebook Ad Campaign Ever, Gets Banned

gizmodo.com

351 points1cvmask92 comments

mostly a dupe: https://news.ycombinator.com/item?id=27040470

Facebook is pushing back on these claims: https://twitter.com/Kantrowitz/status/1389735736644091906

"These ads were never rejected because they were never set by Signal to run. The ad account has been available since early March"

I'm glad that both FB and Signal have both made clear claims which may be proven or disproven. That makes everything much easier for onlookers. As both parties have made strong plays, time will adjudicate this matter.

Karl Popper rests peacefully.

Wow, that sounds like a response made without input from their PR department. They basically obligated themselves to show these ads now if Signal tries to run them (even taking at face value their claim that they were never set to run).

So, they never ran the ads as a publicity campaign, caused some Facebook exec without enough PR experience to promise they will be run, and now they run... a publicity campaign... seems like Signal is winning twice here to me.

It's not obvious to me.

If FB's claims turn out to be true, then it will look very bad for Signal — that they staged evidence to create a fake story about FB, and went on record to double-down on their dishonesty.

Yes but do you honestly believe Facebook over Signal? Facebook literally lied to Congress... Signal is about transparency. I seriously doubt Signal lied. They have no incentive to lie here, if the ad would have run they would have won, it didn't they're winning. It's a win-win for them.

It is possible that neither of them are lying and that Signal just misinterpreted what the Facebook ad system was telling them and jumped the gun a bit.

This honestly seems most likely to me.

It does mean that Facebook’s messaging is confusing enough to cause that though.

Or they "misinterpreted". Plausible deniability enables this win-win scenarion.

Problem is that the images Signal have provided don't really provide much concrete evidence that the ads were rejected after infact being run.

So far even the screenshots Signal posted contradict and disapprove Signal's story, and are in line with what facebook is saying.

Signal is pushing back on these claims: https://twitter.com/signalapp/status/1389767227570798595

And then Facebook pushed back again: https://twitter.com/Kantrowitz/status/1389773302160297985

The screenshots Signal posted showed they were from early March and show no ads rejected. I never thought I would say this, but so far I'm leaning towards believing Facebook over Signal.

Facebook would not push back on this unless they were very sure of their position, because the spotlight is going to be on them harder.

Interesting. After reading the back and forth between Signal and Facebook, I think Facebook’s argument is stronger.

It sounds plausible that Signal’s account was briefly suspended due to a payment issue, and someone in Signal’s marketing is simply mistaken, confused, and/or inexperienced at using Facebook’s ad campaign tools.

These tools get more and more complicated by the day, so that could very well be

Looking at the screenshot provided by Signal it's pretty clear that the account is disabled due to some sort of payment dispute and/or account balance problem.

They have plausible deniability because FB bans are well known to be totally opaque. They don't like to tell you what you did wrong.

The image they posted clearly has a statement on it next to the account disable banner which relates to an account balance problem. It also has 0 rejected ads against the campaign.

Doesn't seem to be totally opaque to me.

Disclosure: I worked at FB, but my only exposure to ads was in an advisory role on an ads product that as far as i know hasn't launched and doesn't seem to be launching soon and my advice wasn't taken anyway.

Looking at the big modal style box in the middle of the screenshot it says:

"Ad Account Disabled

This ad account, its ads and some of its advertising assets are disabled. You can't use it to run ads." And there's a see details button.

Maybe the see details indicates a fixable problem, and some of the text behind seems to maybe indicate a fixable payment problem, but it's not clear at all from the text of the big box that the problem is fixable. It seems perfectly reasonable to me that when Facebook says you can't use this account to run ads, it means you can't use that account to run ads, end of story. There's not a call to action to contact customer support or fix the payment or anything, just disabled, you can't run ads.

I'm looking at this image https://pbs.twimg.com/media/E0lx59fWEAI9HcX?format=jpg&name=...

From looking at the screenshot behind the dialogue box there is a link to "verify the account", which looks to me like an initial fix.

My point being that being told that I cannot run ads, doesn't mean that Facebook doesn't like my ad campaign. There are many reasons I would think why an ad account is disabled, to simply assume this was because FB didn't like your campaign despite having no rejected ads doesn't seem very smart and is a big call. If Signal had followed through to try resolve the problem and had actual evidence that it was due to the nature of the campaign, then they should be showing that, not conjecture.

It's super unclear. Verify the account should fix the 'Ad account status warning', since it's in that section. But if that would fix the ad account disabled, you'd expect to see it mentioned in the modal's text.

And why is it an ad account status warning if you need to fix it to proceed.

I mean maybe Signal overreacted if the text in the big box didn't mean what it said. But I don't expect them to be experts at running advertising campaigns; and I kind of expect advertising campaigns to be easier; at least if it's supposed to be self-serve. If you're expected to work with an ad rep, sure it can be clunky and hard; train the reps, sell to Fortune 500, cash the big checks, whatever. Maybe the story here should be bad FB UI drives away advertiser who thought they were banned.

That's my point, if it's not clear exactly why the account is disabled, it is irresponsible of Signal to create a blog post blaming it on Facebook simply not liking their ad campaign. This isn't a couple of Tweets sent out harmlessly by Signal during troubleshooting, they wrote an official blog post about it.

Signal should have verified the problem first, then ran the ads campaign and posted a legitimate response from Facebook regarding why. The situation makes Signal look nefarious.

When Facebook says you can't use your account to run ads, how do you verify the problem? Does that channel actually work?

How would one know they could use that channel? Since there's no contact link or phone number on the thing that says you can't run ads.

I guess the moral of the story is FB doesn't care about user experience for users or advertisers.

> if it's not clear exactly why the account is disabled, it is irresponsible of Signal to create a blog post blaming it on Facebook

This is called the benefit of doubt, and it’s something you lose when your company and culture gain a reputation of chronically lying to the public, regulators and legislature.

I would be cautious of writing blog posts and sharing tweets based on the benefit of doubt, when the information you provide to substantiate your argument shows signs of going against your claim.

Facebook would likely reject thousands of ad campaigns as they would go through an automated clearance platform. Therefore why are there no rejected ads at all in the screenshots provided by Signal. Without that information I would have expected Signal to have dug deeper to resolve the issue to find the reason why before launching a public claim.

As I said in my earlier post, it's not the benefit of doubt but conjecture.

> would be cautious of writing blog posts and sharing tweets based on the benefit of doubt

Facebook has lost the benefit of doubt. That means when one is in doubt, Facebook isn't given the benefit of assumed good faith.

> when the information you provide to substantiate your argument shows signs of going against your claim

Reading this as someone who doesn't use Facebook Ads on a day to day basis, I don't see anything in their evidence that goes against their claim. It's only the fact that multiple credible accounts on HN have corroborated that claim of Facebook's that leads me to trust it. At the very least, the messaging therein sows doubt.

You've repeated similar comments across this thread. It might make sense to consider where the other side is coming from. This may come down to nothing more than horrible UI at a predictably sensitive time combined with Facebook's arrogance in believing they can skate by with minimal support costs and Signal's trigger finger on its Twitter account.

I disagree, Signal claimed that their ads were rejected. Yet the images provided show that for the campaign in question there were no rejected ads. They need to provide something a little more accurate to argue that.

My argument is based on the fact that Signal have come out with an official blog post about it, because they believe that Facebook have rejected their ads and have disabled their account because of the campaign. They have not said that Facebooks UI is confusing and they’re unable to figure out how to unblock their account, they have made the assumption that the campaign was rejected because and I quote “Facebook is more than willing to sell visibility into people’s lives, unless it’s to tell people about how their data is being used. Being transparent about how ads use people’s data is apparently enough to get banned; in Facebook’s world, the only acceptable usage is to hide what you’re doing from your audience.”. This is a massive call to make and is a big overstep in judgement without much to substantiate your claim (based on the images). Even if it was due to Facebook’s confusing UI, I think it is irresponsible for Signal to simply assume this is what Facebook have done and why.

If it has come down to what you say it has, it does not look good for Signal.

I am in IT, I got a sting at social marketing for 5 years and I am slowly backing away from facebook and GA. You can't do both and keep up-to-date on the tools and their usage at the same time.

It looks like:

1. Signal planned this as a PR stunt, anticipating that Facebook would reject the ads.

2. After setting up their ad account, the account or payment method was flagged for additional verification, triggering the dialog they showed in the screenshot in their blog post.

3. That dialog has a big red padlock that says "Ad Account Disabled". That's perfect for Signal's needs, so they stop here and don't even bother getting the actual ads reviewed and rejected.

Signal's post is careful not to say that the ads were rejected, only that their account was banned.

In response to the pushback, on Twitter, they posted another screenshot showing that the account was restricted, but still no evidence that they actually attempted to run the ads.

> Signal planned this as a PR stun

I would say both Signal + Facebook. I only use iMessage.

In Apple I trust =)

You only use iMessage up until the point you need to SMS a device that isn't an Apple device. Then its not in Apple's ecosystem anymore.

How do you know that they aren’t using iMessage to only send iMessages?

I don't but my point still stands, Signal and Messenger are software agnostic (to a degree). There is every possibility that when I send a person an SMS, it will fall outside of the iMessage ecosystem.

If I send someone a message on Signal or Messenger, it will only come through to them via that platform. With iMessage you cannot guarantee that.

I've been using Signal as an SMS app replacement since it was called TextSecure, and it still works that way - if you send to a non-Signal-registered number, it sends over ordinary SMS (with a visual indicator to let you know), very similar to the way iMessage works.

Difference being that Apple doesn't have a visual indicator before you send the message to let you know you're sending it to a non-iMessage device.

Edit: I was completely wrong, iMessage does do that, my apologies.

I haven't paid too much attention, but that hasn't been my experience. As recently as today when I type in a new number the text in the window defaults to green (a text message), after a brief lag it turns to blue when it validates it as an iMessage account. This all happens before you enter any text in the body.

Sure it does. Apple shows the phone number/contact name and the send arrow in green if it is being sent to a non-iMessage recipient.

I mean it's similar with Signal: you can send Signal messages until someone in your contacts (which are most) doesn't use Signal so you have to send an SMS.

The problem is that Signal prefers to also set itself up as the default SMS app and doesn't clearly show which messages in the feed are what kind. As someone who has to explain this difference to his father about twice a month I don't like this "feature", it's utterly confusing to people who don't know or don't care about the distinction between web-based messaging, SMS and MMS(now defunct here) when they want to send a picture.

I don't understand this, so I'm guessing this is a "feature" of Android... but Signal _mixes_ SMS messages and Signal messages inside the Signal app, and doesn't visually differentiate?

One of the fundamental differences between iOS and Android is that on Android you can replace most preinstalled apps with a third-party app for the same purpose. You can use third-party camera apps, keyboards(iOS too for some time), messaging(SMS) apps, web browser engines, ...

Signal requests the status as default messaging app, which imho is a problem when combined with inexperienced users who tend to just click "OK" to get rid of popups and error messages, and now don't understand why they can't send a picture to someone in their contact list. And yes, Signal just lists everything in the same interface. For non-technical users this can lead to a lot of unexpected behaviour.

Thanks for the explainer. I think I now see why so many Android users are so totally wedded to WhatsApp.

However, you don't have to make it the default. I have separate SMS and signal apps, it's not hard to set up.

This is incorrect, there are small minor identifiers but you need to have an eye out.

It really does differentiate. I just checked, and the input box you have to tap on to type says "Unsecured SMS". Now, some folks might not know what that means, but I think it's a stretch to say is not clearly marked.

Perhaps Signal should adopt blue and green coloring so people know they’re getting protected chats, or maybe a padlock to communicate security.

It's already visible on the "send" button IIRC. I think it's blue when it's secured, grey otherwise

The Signal target group is people who value privacy. You appear to not be in that group.

I've never understood corporate tribalism.

>our advertising policies prohibit ads that assert that you have a specific medical condition or sexual orientation

Are there any grammy-like awards for corporate speak performances? This is gold.

It's funny that if you look at the background of the image Signal posted you see pretty clearly their account was disabled due to some unusual account activity not ad violation.

https://pbs.twimg.com/media/E0lx59fWEAI9HcX?format=jpg&name=...

https://pbs.twimg.com/media/E0lx59bXIAMn5hr?format=jpg&name=...

The screenshots literally showed their were no rejected ads.

smh signal

And you can see that the screenshots are from March 4th.

> if it's not clear exactly why the account is disabled, it is irresponsible of Signal to create a blog post blaming it on Facebook

I trust HN’s opinion on this. But speaking as a layman, nothing in those images screams “easy solution here” to me.

Facebook isn’t Google when it comes to service. But I can totally see someone well meaning at Signal interpreting that messaging, and the lack of any call to action other than “see details” in the error box, as a permanent and irrevocable (without a Twitter or front-page-on-HN tantrum) ban. Hell, the second image even calls out "not complying with [their] Advertising Policy or other standards."

Right, but when your ad campaign shows no rejected ads and you don't actually know if it was due to the ad campaign.

Then I would be reluctant to spread Twitter posts and Blog posts regarding that being the reason. If signal investigated and came back with a legitimate Facebook response, then absolutely.

It's not the benefit of the doubt, its conjecture. Based on the information I can see I have no reason to believe that the campaign was rejected by Facebook, the screenshots show that the campaign did not even run, nor were any of the ads rejected.

I'm not trying to defend anyone however the profession I work in, I cannot raise observations unless they're validated and I don't think Signal have validated enough the reason why the account was disabled. I also don't think the photos they have provided directly point to it either as it is highly plausible that the deactivation is related to something completely legitimate.

You are assuming that people would know that your ad account on FB will only get banned when they have some "rejected ads". I, for once, learned that from your comments here.

You seem to be familiar with FB's terminology and UI, which most of the people are not. That's not on the people, but instead on FB to fix.

Sure, Signal is not doing themselves any favour by jumping the gun, but there is no reason to defend FB's UI when it can be higely improved too.

I’m not assuming that at all, that is my point and I’m not familiar with FB’s terminology nor have I ever used their ad platform. However Signal have said on Twitter that their ads were rejected and the screenshots show that they weren’t.

My point is that Signal have jumped the gun and made some serious claims against Facebook when their account may have simply been deactivated for any number of reasons. There is no evidence based on the images provided to show that FB deactivated the account because they didn’t agree with the campaign. Now that doesn’t mean that isn’t the case, but it’s a big overstep by Signal to then go and assume that.

I could setup an ad account tomorrow and and run a campaign and find that the account has been deactivated. Do I then assume that it is deactivated because FB didn’t like my platform? Do I create a blog post and share it on Twitter also assuming it was because FB didn’t like my campaign? No, I investigate and if it comes out that I wasn’t in compliance with their policy that may or may not be easily interpreted, then I write the posts. Until then all you know and all you’ve shown the public is that your account was deactivated for some reason.

I don't have a lot of sympathy for facebook here. If you disable people's account you need to be clear about why or you are going to suffer blowback.

Is anyone else getting concerned about Signal's trustworthiness lately?

Why? Can you please elaborate more. I recently started using Signal and like it very much. It seems that more and more of my contacts are also starting to use Signal.

They've been doing some weird stuff recently. Like this stunt and the reverse-hack thing, and they also run a cryptocurrency now. That last point alone is a huge red flag for me.

Frankly stunts like this are comforting. If Moxie had been bought, I'd expect a much more subtle hand.

He's running a cryptocurrency, doesn't need to be bought.

The "reverse-hack-thing" is a big trustworthyness bonus.

Even though Signal does seem like a step in the right direction, the fact that Signal is still a centralized platform makes it less trustworthy than say, Element. I must say this nonsense with the ad campaign isn't helping either.

Trust doesn't get easier when you are decentralized.

I rather like the idea of this ad campaign. If it shakes up a handful of people to be more aware of how much data is gathered and used against them, I consider that a win.

Once they started adding shitcoins you know they’re going the way of keybase

No.

It's irresponsible unless they were waiting for the ban and purposely orchestrated it for publicity

I am surprised to hear that commenters consider a person with a closed Google/Facebook/Amazon account of being able to clearly know why it was closed and how to remedy that situation.

I have read many HN articles and pleas for help describing the exact opposite - a cancelled/blocked account and no proper explanation.

Of course in this gamed scenario, both parties claim the above! Ludicrous.

But this is not a case of the account staying locked and Signal being left with no recourse but to assume the worst (a reasonable approach to megacorps).

At least, according to Facebook, it was only locked for a few days and at no time after clearing up the payment issue did Signal actually setup and attempt to schedule the ads to run.

Well, according to Facebook. Whose logging could be wrong for all we know.

In this scenario you can see on the screenshots posted by Signal that it looks to be related to a payment dispute or account balance problem. Its also silly given that it says there are no rejected ads, yet Signal said the ads were rejected.

It's one thing to not know why your account is suddenly disabled, but it's another to attribute that to Facebook simply not liking the campaign you wanted to run. Perhaps if Signal followed up and were provided with concrete evidence that it was due to the campaign being taken down, then it would bolster their story.

The problem is that plenty of news articles have already reported on Facebook banning the account due to the campaign. So it seems like Signal have got what they wanted, however from my point of view this is a scummy move by Signal and makes them seem more nefarious rather than open and transparent.

Ok, so if Signal was banned for different reasons and facebook claims the ad campaign wasn't the problem... it means that Signal can create another account and run the same campaign again, Facebook shouldn't block them for that in the future. Let's see what happens.

In that case they will get banned, because Facebook does not allow advertisers to let the targeted group know why they were targeted.

https://twitter.com/dmarti/status/1389737052091027460/photo/...

Not the first time someone has tried this technique... Remember:

Want to buy black hole on eBay? Cheap black hole available to buy now!

Those ads just stuffed every dictionary word permutation into the ad API. Can you imagine how much load that put on ad systems? I can totally imagine Facebook deactivating an advertiser who tries to upload and activate 1 trillion+ ad creatives and sets of targeting criteria, especially if their total spend was low.

These systems are highly scalable and ad buyers commonly use programmatic ad buying to test many combinations

Programmatic ad buying still requires each creative be approved on most ad networks. This technique would require billions of creatives, most of which would never even be used.

I can see how ad networks may not want to take on that...

They aren't that scalable though. For example, every ad system I've worked on requires every ad creative in RAM. Having an ad creative with each combination of a life event, job title, interest, and city might be 100 bytes * 10 life events * 1000 job titles * 1000 interests * 1000000 towns/cities = 1000 TB of RAM.

That needs duplicating in every ad serving datacenter. (say 20 for facebook)

Will facebook be dedicating 20000TB of RAM to signals ad campaign? Thats $86 Million for the hardware alone. I think not...

Where did anyone say Signal intended to target every possible combination of criteria?

This is lesson in media, PR, liars, spin, misrepresentation, attempts to explain, yada yada. There are hardly any good actors here: Signal for misrepresenting, the press for turning a blind eye, Facebook for collecting too information etc..

This is a really bad look for tech.

Cui bono: who benefits? This seems an asymmetric affair in favour of increasing Signal's visibility (infamy is still fame) and in reducing people's trust in big tech. A win for privacy team, even if you can debate whether it's a win for Signal in particular.

Every single word in his headline comes as no surprise to me. What about you?

Well, it looks like Facebook did not ban the ads. But that Signal have confused their ads being rejected for a payment dispute with their account.

Is there any way that you can sign up without a phone number?

They are working on it.

They said that like 3 years ago though…

In asia phone number is really the only viable default registration option if you want to go mass market.

I’m afraid that doesn’t make any sense at all. Can you try again?

Seems pretty clear, what don't you get?

Registration via email isn't nearly as popular here as it is in the west. If you are going to choose a registration option to initially support, phone number is the correct choice.

Dubious story. While Facebook is one of the worst examples of surveillance capitalism the evidence is very questionable in this case. Screenshots saying 0 ads run(?)/rejected(?)

Reply on news.ycombinator.com