Media buyer: Facebook updates have ‘wrecked the ad platform on the backend’
digiday.com
digiday.com
Yes, Facebook is a complete train wreck, which survives in spite of its abysmal technology and support for its real customers rather than because of them. Even then, speaking only from the point of view of a much smaller business that has been a regular Facebook advertiser in the past, it probably left 33-50% of the money we would have spent with it on the table simply because we couldn't spend that money on the Facebook platform due to all the bugs. We redirected those funds to other channels instead. And yes, we too only got around Facebook's systems demanding nonsensical or literally impossible things that would have totally blocked our access at least once because we knew a senior dev who worked there and could escalate internally.
Unfortunately, this model seems to have become SOP for big tech service providers in recent years. Google are similarly non-communicative and have also provided such a bad experience to us as an advertiser that we stopped all our ads on their platforms for a long time. It's not just the big advertising platforms, either. Every payment service we use (including some that are darlings of HN) has messed us around horribly over the years, too. Unless you're spending literally millions of dollars per months with these tech giants, it seems they've decided it's not commercially sensible to offer any meaningful support, and presumably to accept that in some cases they'll lose customers entirely and damage their reputation as a result.
It's like living in a run down tenement, no matter how many rats and roaches you have to deal with due to a lack of maintenance, you have no choice if the alternative is the street. Yet the owner makes money no matter how horrific the building.
The advertisers are effectively the customer, with the social media platform acting as a vehicle to reach the willing end-user.
With this model, a polished & efficient dashboard/UI for advertiser engagement is as important (if not more important) than the front-end experience that provides end-user engagement.
It might be the case that there is simply not enough subscription revenue in the world to fund such an operation without advertising. People forget just how much objectively good and useful stuff we have access to because of ad-supported $0 products. Hell, you could argue that Tensorflow and PyTorch themselves are ad-supported products.
The sticking point is that your old ad-supported classifieds paper and network news channels weren't surveilling you across the Web.
And as much as I think we need strict data privacy laws, it's hard to ignore the fact that Facebook ads used to be for penis pills and karate scams, but now for the most part are things that I (or at least my demographic) might want to buy.
I guess a vast majority of people simply don't want to pay if there is a somewhat acceptable free alternative.
The reason I do not pay for YouTube is that despite paying, I'm still the product - Google is still collecting every bit of data about me, to spam me on their other channels. Instead, I use YouTube (and every other Google product) logged-out, deleting cookies often[0]. I can't do that if I pay for YouTube.
[0] I used to reject cookies, but Google has become unusable that way -- everything goes back to defaults which are exceptionally bad for me.
For me, this is similar to why I won't pay for the New York Times subscription, either. I mainly only do stuff on both sites when I find a link to their site. I don't go to their site to do stuff on my own. Why would I subscribe to a site that I don't go to on my own? I do subscribe to ArsTechnica, for example, because I go there frequently.
Furthermore, I might not mind ads, but the ads on YouTube are very obnoxious. There's no intelligence, so you get weird situations like watching a 2 minute ad before being able to watch a 30 second video. Or ads cutting off the speaker mid word. Who thinks that's a good experience? If they can't get that stuff right, why would I pay them?
And on top of that, I find YouTube's "Hey did you know you can pay us money now?!?!?!" pop-ups all over the page to be obnoxious enough that I have decided I will never pay them any money.
Were I a regular visitor who wasn't so off-put by their ham-fisted advertising, I might have considered it, but now they've made going to their site so unpleasant I just avoid it if I can, or hold my nose for a few minutes when watching something with ads. Then I get away from their site as quickly as possible.
Things are going well so inefficiency is fine (and it really may be fine), instead of fixing problems the best people are looking for new sources of growth.
"Good luck using a better ad platform with no users to target your ads to".
The real alternative is simply not playing their game which is essentially an arms race with other advertisers anyway.
That might be true for huge advertisers trying to reach mass-market audiences.
For smaller businesses operating in niche markets, it is plausible that you can get better returns through other promotional channels that appeal directly to those markets, without needing all the nasty tracking or dubiously effective targeting that come with the likes of Facebook and Google. The idea that you "have to" advertise on the big, generic networks because that's "where all your potential customers are" is well out of date, at least in our experience.
They want to "get things done quickly" and inevitably break code elsewhere on the platform. The guy in the article is right - there's no other platform around where these sort of issues appear so often.
Which, in the case of the article, is exactly what happened.
But this is the game. I've spent millions of dollars on G/FB/etc and that's just how it goes.
This is insane.
> They also restrict your account so you can't do anything, but they leave the ads running and keep billing you. This bankrupts people who can't pause their unprofitable campaigns.
This is insane too.
The vague reporting of likely spend/speed of showing ads has always scared me of putting a large budget on a single campaign upfront.
There are a ton to criticize about google's adplatform (from user data handling to transparency to semi-monopoly to ...) but at the end of day if I make a proper well targeted campaign I get matching results.
Not all of them and there are some obvious false in there (ergo the need for transparency), but for every company who I have personnaly done the adwords campaign, as long as I disable the display network and make sure the campaign is well made, I get my bang for the buck.
On the other hand I'm still not sure why anyone who isn't scam or retargeting advertises on facebook, its "leads" are absolutely terrible in quality. I'm the kind who is willing to spend lots of time and a/b testing on my campaign (since at the end of the day it's the source of my money), but no matter what I do it costs me between 3 and 20 times more on facebook than on google to get the same number of end game sales.
It's a terrible ad network. When I dabbled in adult toys sales a while back I had to deal with adult networks ads and even THEY were better. Honestly it's impressive to me how bad it is.
(ads are purely textual and non retargeted, real products or services with great reviews)
I'm saying if I have a buck to put, it's worth several times more on adwords than on fb.
If you are traffic starved then sure fb could be usefull but I suspect the vast majority of small/medium business runs out of ad budget before they run out of adwords growth capabilities, and they value qualified leads much more than raw traffic.
Again OF COURSE it varies by business and I'm sure other people have the opposite experience, but mine has been clear cut and I shared it.
> gave yourself a lot of credit as a marketer
I in no way intended to do that. If I did, I apologize. I may be bad, or good, or decent, or terrible at it, I don't know. The point I was trying to make on that part of my original comment was simply: I do spend time on my campaign and A/B test difference, not as-in "I'm great", but as-in "I didn' just make my first campaign there in 5 minutes without trying to improve it or learn from my mistakes" because I think anecdotal evidences given from that point of view have no value at all.
So again, sorry if that is indeed what can be understood in what I wrote because that would be egoistical and wrong, but I genuinely don't believe that's what I wrote.
I have never sold below 100$ product, I have for services but I was taking a loss for the first 6 months per customer to acquire a base.
My best performing campaign in an entirely different market (education) is for a 4000$ product/1 year service combo, cpa is at 150$ but basically one out of every 10 leads is a sale (cyclical depending on time of year so this is a "make bank for three month" thing).
It's a $70bn ad network. All that revenue comes from highly liquid ad budgets. It's naive to conclude that fb ads "don't work" from anecdote... Not working most of the time is the norm for advertising, but the scale comes from the cases where it does work.
Anyway... this is just not the case here. The advertisers that pour budgets into fb aren't suckers. They're highly roi-sensitive and data cognizant.
All advertising fails much of the time, but when it doesn't it scales.
How do you know? Facebook obviously carries ads from huge brands, but also has a very long tail of smaller businesses and other advertisers. While it's reasonable to assume that whoever is spending a few million here and there on behalf of some major retail enterprise is watching the numbers pretty closely, I see no reason to assume that a thousand small businesses who are spending a few thousand here and there are doing the same. It's also worth noting that several things Facebook has done in the past have prompted reports of Major Brand X pulling $Y million in ad spend, so apparently the people who do have the sophisticated ROI monitoring don't always like what they see.
I will be honest and say that the prefix was assumed to be understood without having to be written, but I understand how without it my statement could be seen as something else (and rather obviously wrong).
I've heard (no direct experience) that FB leads tend to be high-quality when one is advertising, e.g., mobile games monetized through in-game purchasing.
For references my own ads have been looking for leads (not traffic, not subscription/registration), and the areas have been education, IT services, home help services and niche market retail sales (aka low sales but of several thousands per unit products).
I recently had to open a support ticket with Paypal, they are exactly the same, unfortunately
I understand the incentives that led to this, but it's still crazy to think about it, since it's obviously a worse use of time for everyone (both me and their agents) just to get some metrics to go up
This smells like a lawsuit.
Do whatever Coca-Cola, McDonald's, and the other big and hugely successful advertisers do. They have entire staffs to make sure their ads end up in the right places at the right prices.
See what your competitor is doing, and then do slightly better. Assuming you can afford it.
I used to run FB ads for my brick and mortar ca. 2016-18 to the tune of about $1k/day at it's height (I know that's not a lot necessarily just providing context). I never had any issues at all, the experience was actually pretty great. But, I've heard similar stories to yours from a lot of people so I know it happened I'm just curious if it was a change in the last couple years or if I just got insanely lucky.
Locking away your backpack in a supermarket, bag inspections in the cinema, online tracking, and captchas. And good luck if you ever get annoyed at the barrage of unnecessary captchas and just click random stuff... You'll get shadowbanned.
Because nothing says "robot" more clearly than getting offended by useless interruptions.
Why would you not just call your bank up, and get it charged back/refunded at that point? Thus Facebook would probably "ban" you (like most merchants when you do a chargeback), but since you're already restricted, who cares. At least you aren't losing money from them.
I think I had my account banned the last 3 times I created a new ad. None were against their policies. My account go reinstated each time but still.
Sadly, competition is difficult to flourish in the current environment because of Facebook's market share and tactics.
It's not as bad in the US compared to other countries especially developing countries where Facebook is pretty much the only online channel. Sometimes, people only have access to Facebook and not the rest of the internet.
It's about pride. Not everyone has pride in their work, in their person, or in their actions.
I once worked with one of these "good enough" people. He bought his wife the cheapest things because they were "good enough." All he was focused on was the ticket price. Not the quality. Not the workmanship. Not going with a good company instead of a bad company.
It's become more common these days in everything from consumer goods to personal behavior. There was an article in the newspaper yesterday saying that sales of Folger's Crystals is up 14%. It's "good enough" coffee. There's a congressman up for re-election right now who admits to having an affair with an intern. His campaign doesn't hide this fact that would have made him un-electable ten years ago. His behavior is "good enough."
It's like all the Facebook and Google employees who are working on things that make the world a worse place. They're making money, so that's "good enough" for them. Someone else can be moral.
And similarly most of the positions that Amazon, Facebook and Google recruiters have approached me for are also in that field (even though I don't have any professional experience in that area, nor any keywords in my LinkedIn profile). I understand it's where the money is, and it's probably a good idea to work in a profit-center for once instead of being viewed as a cost-center by the MBAs, but I still have a soul.
I'm at the age where I'm starting to look back at the various phases of my life and think about whether they were wasted or not.
A lot of my similarly-aged friends are doing the same. Some who worked for the wrong tech companies really regret it now. At the time they thought it was cool because they were living the tech dream. Now they wonder what it was all for.
EDIT: for a while this reached 10 points, now it is at 2... makes me wonder, what I wrote so horribly wrong that deserves all these downvotes?
[1]: https://addons.mozilla.org/en-US/firefox/addon/hide-fixed-el...
1. This should probably be on by default with some indication it's "working"
2. It makes me so sad that I can't use extensions like this on my Android browser
Turning it on by default would break a lot of sites because they pop open one of those illegal cookie popups and disable scrolling until you've gone through it. If you can't see the window, all you get is a broken page with no explanation what went wrong.
The addon is open source[2] so if you'd like the feature nonetheless, you should be able to do it with only a few lines of code.
[1]: https://blog.mozilla.org/addons/2020/09/29/expanded-extensio... [2]: https://github.com/andfarm/HideFixedElements
1) Try uMatrix. It blocks third-party JavaScript by default, which is what's used to power a lot of these elements. I don't see the big "subscribe me" bar at the bottom, so I assume uMatrix is preventing the JavaScript that creates it from loading. This will also help with performance on your "ancient workstation".
2) Reader mode seems to eliminate all of this cruft and give you just the meat of the article to read.
There's a newish rule on HN that we're not supposed to complain about the look of the linked web sites.
I don't mind when people point out that a site is unusable, because then I don't bother clicking through to it. But it's their site, so their rules.
For comparison, AdEspresso is not nearly as complex but hugely more productive for doing split testing in Facebook campaigns. You build your ads on their interface and they submit via API.
I have no problem getting good customer support from Google's paid products (YouTube Premium, YouTube TV, Google One, etc).
In a slightly more serious tone, the interviewee pulls out the old “won’t someone think of the small businesses?” line at the end. Is there any evidence that this is true? Is is primarily small businesses that are hurt with this kind of instability?
Marketing / advertising is a sub-sector of exploitation.
Modern advertising is technology deployed for the purpose of exploiting psychology to effect persuasion.
Marketing is not the same thing as advertising, but it can frequently be described identically.
This has been a universal problem that I've observed in most tech companies i've worked with. Hyper-focus on the optics and less on the content / results.
I dont know if its a by product of giving front end design equal / greater weight than back end design / analytics or something more systematic like the apple design influence. Maybe both. Eitherway seems like websites are getting easier to look at, more clever, and less useful all across the board.
Maybe customers receiving bad support spend just as much, or possibly even more, than those receiving good support.
It also shows how developers rank the priorities of bug fixes.
I'm an Xcode user.
That means that I experience lots of bugs; some, pretty damn nasty. I tend to only report the ones I consider serious, but I find it a bit depressing that so many of them seem to remain untouched.
That's because Apple assumes that I have a much higher tolerance for bugs than the vast majority of their users.
They're right. I keep using Xcode, despite the bugs. I'm not a fan of other IDEs, and my experience with them, is that they are every bit as buggy as Xcode, except that they also run slower.
But user interface (and security) bugs in the general operating systems get high priority.
I suspect the same goes for Facebook. They know that they are the only game in town, and that marketers will tolerate a much buggier experience than users.
I would like to see software quality rise all over. Speaking only for myself, I spend a great deal of time refining the interface for administrative functionality, and have a standard policy of "Don't move on to the next item until you believe this one is 100% bug-free, and completely usable.", even if it means delaying, or even reversing, progress of the project.
That's not a popular stance, but it does tend to result in pretty high-quality software. I'm writing a social media app, right now, and this philosophy has paid off in a big way. It does mean that the team needs to wait longer for some functionality, but the app is constantly in a state, where they can simply hand it to potential investors, and not worry about them encountering highly embarrassing bugs. I'm hoping that the end result will be a real joy to use.
What other alternatives are there?
Even doing a simple Instagram promotion in the App has to jump through the FB app for authentication which can hang and show me a blank screen.
They have so many profit and revenue levers, this is another example of one.
So many companies would pay for ironclad API access and relationship.
I believe that the lack of transparency around their APIs is by design for political reasons not a lack of engineering know how issues or ignorance issues.
They make it sound like a bad thing but pretty sure it's not true anyway. If Facebook saw that their ads performed the best with Comic Sans they would go with it and any design changes they make are surely done to increase ad revenue.
If you're spend $100-1000 a day you get tests and early UI/UX. Spend more get out of the baby support tiers. Someone has to the best test bed.. might as well be the smallest clients.
That aside, IDK I read it in the group I was in and felt it would be liked here.
Can you name any other business that could function like this? Health care is the only ones that come to mind...
It's just like playing along to Google's game (including vague "exact" matches for adwords) because there's no alternative, and no recourse.
It stinks.
And to top it off, almost all digital ads are money losers for the buyer. Its the few that work that get backed into their ROI calculations. Which are essentially a tax added to the purchase price of anything marketed and sold via these channels.
The business is a fraud.
The irony is that one of the ways these big ad tech companies promoted themselves as being better than traditional advertising was because they were supposed to be so transparent compared to the old model where you had to go through an ad salesman and then measure your results in sales.
At least when you have an ad salesman, you have someone to complain to. And their boss, if need be.