Ad Fraud on LinkedIn

I observed the same with Google's "Smart" display ads. I ran a campaign for a couple of weeks and watched all of the traffic with HotJar. The first fishy thing was 99.9% of the traffic came from foreign countries despite the target being US/Canada. When I saw the targeting wasn't effective, I even set up specific geographic exclusions for every country != US/Canada and still the traffic was from mostly underdeveloped countries. Then there was the behavior of the "users" when they visited the site. A large percentage of them went straight to the search box (which literally no humans do on the site) and typed in something topical but completely unrelated to the site's topic, e.g. "covid". Another large percentage just scrolled up and down randomly. It was interesting to see the size and relative sophistication of the fraud going on but not exactly $500 well spent.

Hard to know how much is ad fraud and how much is people just not being interested in what is being advertised. Don't see why I should care about yet another webinar especially if I'm already a successful CEO and get five thousand webinar requests a day. Might click for more info but then after 10 seconds on the page (and seeing it's more BS) I'll probably bail.

Same as this quote:

>“You’ll see the call to action is ‘Request Demo,'” Gellis said. “So, arguably, a user who sees this ad and clicks legitimately on it will be looking to get a demo of the analytics tool we were marketing.

No, if the only option you give me is "Request Demo" but all I want is more info then I'll click "Request Demo." Then when I don't get more info or info that I like I'll bail.

edit: Also at only 11 clicks, the metrics are pure noise. Sure they each cost $45 but it's still only 11 clicks. Can't even blame the CPC estimation for being off at such a low number.

I lost a couple of thousand dollars using their services (literally 5x my ad budget) because they "refill" your ad account based on the ceiling you intend to spend - without an email notification or other notification in your main profile. It cycled through my ~$300 ad budget and charged me that amount multiple times before I caught it on my bill. Never received the refund they said they'd send either.

People say that half of all advertising dollars are wasted, you just don't know which half, but this company has perfected it to 5-10x what you mean to spend on advertising is wasted, and you don't know until they've taken it from you and then shown you the fine print.

In short, avoid.

> “You’ll see the call to action is ‘Request Demo,'” Gellis said. “So, arguably, a user who sees this ad and clicks legitimately on it will be looking to get a demo of the analytics tool we were marketing.

This guy is deluded. People click random things to get more info, and they might not have liked the more info they got so didn’t go ahead with the demo.

I advertised on LI about a year ago. I had reasonable success (in terms of new leads/signups). Here is what I remember from the experience.

I found that the interface for managing ads was terribly buggy and shouldn't even be considered production ready. One bug that I remember is that it wasn't possible to remove ads, and the button to suspend an add didn't work most of the time.

In typical LI fashion the interface was full of dark patterns, it was really tricky to limit the expenses. In the blink of an eye you could be spending hundreds of USD per day.

Analytics were somewhat available, but basically useless. For no reason the CPC could jump to dollars per click, and no way to discover why.

Advertising, especially for a tech SaaS, is hard. The tools available are usually terrible to use, expensive and often shady. In my experience, doing advertisements is one of the worst parts of starting a new service.

I don't know anything about LinkedIn ads, but I watched some of the video that this post is based on: https://www.youtube.com/watch?v=jKuyxgWuiRM

It sounds like they are moaning that their client side rendered React-based landing page had a lot of bounces which they don't believe because client side rendered react pages are known to be "very very fast" and "some of the very fastest available technology". (This is around the 15:30 mark)

That's baloney, client side rendered JavaScript landing pages are much slower than server-side rendered.

LinkedIn is one of those things you have and probably only use when you need a job or are desperate for clients.

No one is there to get decent information or to genuinely connect with people. It's the digital equivalent of the super fake networking seminars people go to. Everyone is a seller and on one is a buyer.

Makes sense that their ad network is sketchy and buggy.

Sooo, I've spent hundreds of thousands on LinkedIn Ads on behalf of clients. Here is my take:

Ad platforms work in varying degrees for different products and companies. You should not dismiss an ad platform based on poor performance alone without understanding why it performed poorly.

It's sort of like with airlines: If you boycott an airline for every minor offense, pretty soon you'll have no option but to take the train.

In this case, it's disingenuous to paint an entire ad platform--that does hundreds of millions in revenue[1]--as "awash in mistaken clicks and bot traffic" based on a campaign that ran for just 1.5 days and resulted in 11 clicks. (They say they saw similar results from larger campaigns, but they don't show them.)

But Ok, let's say we take out the generalizations. I get that it's disconcerting for any company to find they were charged for traffic that bounced within a second. In that case you need to look for the root cause, and see if it can be resolved or worked around.

In this case the root cause should be painfully obvious to experienced marketers like the authors: Most of the 11 clicks seems to be from mobile traffic (as seen in their Fullstory report), where it is remarkably easy for users to accidentally click on an ad in their news feed. They do mention this possibility in both the video and the title, but they seem to underestimate the likelihood of this being the root cause, and talk too much about bots instead. I've seen this exact misclick behavior in Twitter and for that reason I stopped advertising there.

Unlike Twitter, however, LinkedIn lets you work around this. They offer a "lead gen" campaign type, which requires two clicks (one on the ad, and one more on the "submit form" button) and sharing of contact information before the advertiser is charged. I've still seen funky results from this--such as users saying they don't recall submitting any form--but in much fewer cases than with simple image ads.

Today, when I run LinkedIn campaigns, it is almost always with lead gen ads. The results are far better than this anecdotal post suggests. Some issues still come up, and I've found that attempting to troubleshoot them leads to a better outcome for the business than dismissing the entire ad platform.

[0] https://www.gkogan.co/blog/how-ad-campaigns-fail/ [1] https://www.businessofapps.com/data/linkedin-statistics/#3

LinkedIn is the weirdest product ever. Everybody is there, nobody is using it for anything. Except shady recruiters.

B2B SaaS founder here. I had exactly the same experience. I built my own code for tracking signups, and discovered that the actual conversions were a big fat zero. That's when I stopped burning money on LinkedIn ads.

Mind you, this experience wasn't much different for other ad networks. Quora was slightly better in that at least there seemed to be some interest shown by people landing (or perhaps this was the only ad site that sent actual people my way?).

The use of the term "study" is quite misleading. This is hardly a "study".

I think the title should be "Someone has a bad experience marketing on LinkedIn, gets frustrated, and assumes LinkedIn is to blame".

> LinkedIn’s ad platform reported 11 clicks. RMG’s logs showed ten clicks.

This is well within the IAB's 10% discrepancy. Also that sample size is bullshit. I could show you the sample size of my $7,000 campaign from a couple months ago that recorded a slightly larger discrepancy, but the fact of the matter remains that LinkedIn was the best performer of all ad networks I tested.

Again: That's not to say there's no fuckery afoot but that is a hell of small sample size to refer to as a "study".

Amazing how much wealth in Silicon Valley has been built on top of profits from rampant ad fraud.

I deleted my LinkedIn account because it attracted an endless number of bad business partners. I got sick and tired of all the "Joe Blow"(s) on LinkedIn who claim to be "Joe Blow Inc." (I thought the whole point of a corporation is that it is greater than one person.)

I've certainly found work on LinkedIn, but my experience is that for other kinds of sales and networking it is a terrible time sink.

Based on the intense level of BS I can't imagine that advertising would have really worked there.

> LinkedIn Ads network is likely awash in mistaken clicks and bot traffic

Just like every single other adnet? The very fact of massive adfraud groups allegedly specialising on one or another net means that they must be making money somehow.

By the definition, those guy must be one step ahead of the adnet, or they don't make money, and fizzle in a few months.

My impression is that there are groups which have consistently defeated all, and every bot detection method the entirety of ad industry threw on them.

Why I deleted my LinkedIn account? Literally the baddest UX ever. Lots of self-serving, self-promoting, click-baiting people. There is a thing called web. In this thing you can have a blog, in this thing called blog you can actually reach your ideal audience. LinkedIn is a tool created from the start to serve a corporate idea, which in my view is useful if executed with a boundaries. Where are LinkedIn boundaries? They don't exist.

> Error establishing a database connection

Based on the title I'm guessing this could have been delivered as a static web page delivered by a CDN.

I've never understood why some people want to set up their own web servers to deliver static content.

A sample size of of 2,000 impressions is absolutely useless. Even the larger example is completely useless.

I don’t even considering analyzing a campaign with less than 50k impressions, and I won’t look at a variable if it’s under 10k. Realistically I want a few hundred thousand impressions before trying to make big claims about fraud.

I'm confused, why would a fraud bot click on ads that makes LinkedIn money? Normally ad-fraud is when a fraudulent publisher is set up, made to look legit, hosts ads, which a bot then clicks the hell out of. Where's the incentive here for a bad actor on LinkedIn?

So the same as every other ad network then? Only half joking.

Our niche SaaS sees 90% of Google Ad spend going to India, Russia and South America unless we create a short-list of countries to specifically target. And Google is full of dark-patterns around your targeted countries, by default it selects "people in, or interested in, your targeted locations" unless you drill down and find an obscure setting to change this. Then you only need to deal with the domestic and/or more sophisticated ad fraudsters, which will still probably end up as the majority of the spend.

I am not convinced by their single ad, one day of data (11 clicks total) that they have proven anything.. This was a waste of time unfortunately...

Some apps, and many webpages on Android have this dark pattern of miniscule transparent x's to close ads and popups, that somehow don't register very easily. I'm on a Pixel 3 and I've replaced the device for unrelated reasons, it's not my screen. And my fingers are normally-sized.

What that means is that people get this popup video playing, get a little X, try to hit the X but hit the thing behind it, or click on something else taking you elsewhere, or starting the video if by miracle autoplay ISN'T overridden...

And so forth. I could go on and talk about how I have to zoom in with two hands, crane my neck and open the little x so it's a bigger X, so I can close it...

But I'd rather focus on the question - how is this sustainable? That hostility makes users hostile, and it makes your entire ad infrastructure feel like a dirty enterprise, not the "it runs the Internet" philosophy that I'm guess adtech says to itself. Here's the thing - if I'm the company trying to advertise, do I even want to pay for clicks that are unwanted?

Sounds like this person is bad at advertising and marketing, and looking to blame anyone but themselves.

My guess is the content of the ad was poorly paired with the landing page. For what it's worth, I ran a LinkedIn campaign, had a $10 CPC with about a 40% conversion rate from a click to lead.

Also that sure is a lot of "conclusions" to make very little data.

I may have a meagre business and live in a meagre world but would never hire a single person who subscribes to linkedin nor participates in their blatant games. It's the most ridiculous HR wetdream and yet some seem to have no problem playing along.

Lay your own beds with that stuff. It's only a matter of time before you'll live in shame for what comes out and probably write woeful blogposts about how fooled you were and now so enlightened never again to repeat such clear mistakes.

Having self-worth really isn't that hard. Selling yourself and all your friends for pennies is incredibly easy though, which is what is happening here.

Please keep digging those graves. The internet never forgets, nor forgives. The internet is forever, make no mistake.

Facebook gives you better returns as well as Google Ads, Linkedin was not a good convertion rate at all and they don't really care about it so yes you're correct about the statement of "Money Pit" it's because we're used to Facebook and Google conversions that we expect the same from Linkedin and actually it's different platforms with different goals and objectives, not sure why you should expect the same "money's worth" on every platform out there each one has a different objetive and different users that's how it works they're different.

This reminds me of our Google AdWords experience for our enterprise analytics product. We started with an experienced marketing consultant and later a Google-provided account manager joined. We were tuning our acct setup to get signups at an industry avg, yet some reason the leads rarely qualified. Checking in a month or two into their experiment, I realized half our budget was going to middle school students using google as a calculator for their algebra homework!

That was a bit of an 'aha' for me for their position in the market.

I know nothing about advertising, but does it really matter if you are getting bot traffic if you are still getting conversions from the ads? If you know that for every $10 you spend on LinkedIn ads, you get 1 person to sign up for your service, then it doesn't matter how many mistaken clicks or bots clicked on the ad since click through isn't the main metric you are using. Isn't that why websites have so much tracking? So they can attribute an ad to a conversion rather than just looking at the number of clicks?

A pay-for-performance model (e.g. pay per sale) eliminates all the risk for advertisers.

Does this exist anywhere, aside from affiliate deals?

I'd be happy to pay 50% for each digital product sale, since those are 96% profit.

Those are very small sample sizes to make any assumptions from. 11 clicks for their test, and only 256 clicks for the larger referenced test. The larger test then makes a huge assumption in user behavior:

"So, arguably, a user who sees this ad and clicks legitimately on it will be looking to get a demo of the analytics tool we were marketing"

Uh, no. Just because the ad brought them to your site, doesn't mean the user wants to interact with your video. You got them there, you have to work to keep them.

I don't quite understand the economics of the ad fraud on the closed (on site) LinkedIn ads.

How does a third party (not LinkedIn, and not the advertiser) make money from this?

It’s very clear this company doesn’t have an experienced media buyer running their campaign. This blog post is /exactly/ like a nontechnical person spinning up a bunch of aws stuff and complaining about billing fraud.

Media buying is a niche skill, and you can’t just throw money at LinkedIn and expect it to work. (Especially for true for a /branding/ agency and not a media buying agency)

The biggest mystery for me is why so many people believe so strongly in Facebook Ads. I have never gotten ANYTHING other than bot traffic from FB, or anything above 0% conversion, after repeatedly (mis-)spending thousands on FB ads. I must really be going all backwards about it, or something.

(P.S. this is for content that otherwise gets a very healthy conversion rate from organic traffic)

Here is a recent video from a meetup hosted by LinkedIn where an engineer explains exactly how LinkedIn advertising works. https://www.youtube.com/watch?v=p2YLoyFXvIY&list=PLihIrF0tCX...

I’ve experienced the same with Snapchat and their geofencing ads. I geofence a perimeter around a venue where we run an event and run a branded filter. We have a team of 5 in the venue during the hours the ads run. None of them was able to see the filter come up. Snapchat on the other end shows hundreds of impressions and tens of uses. Surely a fraud.

Even the job posting on LinkedIn seems very shady. This was the result of $32 daily budget LinkedIn suggested us for running our job ad: https://i.imgur.com/7fYv6jy.png

Out of two people who applied, one seems like a bot/empty profile.

"FullStory reported mouse movements that were nothing like what an actual human would do."

That web sites can tell where I move my mouse is simply unacceptable. What can we do to block JS from reporting mouse movements? (Yes, I know I can turn JS off altogether, but most sites aren't fully functional without JS.)

I am so surprised to be redirected to rate the linkedin app while using their mobile web interface, I usually give them 1 or 2 stars for wasting my time. Maybe I actually misclicked on one of those pop ups that takes me to their app.

makes me wonder what I may have agreed to on various sites by pure misclicks.

We experienced a similar issue with LinkedIn ads, we had a small campaign for two weeks to a page which was only designated for this specific LinkedIn campaign. Apparently we got a few clicks (around 30) but there was not one single request for that page in our webserver logs.

I’ve never used LinkedIn, do they have any real competitors or is this just a monopoly being a monopoly?

My profile is on LinkedIn because I feel like I’ll miss opportunities if I’m not on there.

I’d love to remove it and go somewhere that doesn’t do this.

Are there any non-terrible, more purely motivated sites for an individual to post their background and what they’re looking for so folks can view them?

Simple linkedin lifehack:

1. put utf8 emoji at start of your name with space following it and then your first and last name. whatever emoji do: smile, country flag, beer. you name it.

2. discard any connect request or inmail or message that starts with 'Hello <emoji>' - it is automated

3. profit :)

it's pretty easy to click on competitors ads with bots to run out competitors credit and then run your ads for the rest of the day for cheap prices. But the same happens with Google adwords. I guess they don't have any way to prevent it

I logged into LinkedIn yesterday and it took me to a special page where it excoriated me for using “automated tools” to make fake activity or something. It demanded I click an “acknowledgement that I will henceforth comply” or I was subject to account termination.

Frankly I felt like refusing and telling LinkedIn to fuck off. I have no idea what it was accusing me about. I almost never use the site. I have a 25 year career in an industry where I am known and the industry knows me, so LinkedIn is zero value to me. I would never find a job there nor look for a candidate there.

So whatever their systems are doing, they’re incompetent and generate false positives.

Shamefully, however, I clicked “I will comply”. Bah bah bah bleets this Sheeple.

I've long contended that linkedin is essentially like a seedy dating site where we're all the women and the recruiters are the men hounding us.

All LI Marketing I've done has resulted in net zero leads. My own content create if free, apart from my time cost, and is much more fruitful.

This is a reminder to myself to someday explain how these ad-click botnets work, how they make money directly and indirectly by clicking on your ads, and why the big ad tech companies need them to survive.

The open question is whether there will still be perceived value when these things are known. So far, I don’t see ad spending affected by these exposes.

I can't think of anyone who uses LinkedIn for anything other than finding jobs.

I have never clicked on a Ad from linkedin purposely. The thing is, everytime I go to check my profile (which is on the left top part of the screen - I'm dextrous) I end up clicking on a Ad with the palm of my hand (smartphone).

likely? It's completely filled with bot traffic. At some point someone needs to create verified accounts on LinkedIn.

The issue isnt the lack of clicks perse, its the fact that LI and others digital ad providers charge so much for something that is neither market driven nor validated by independent auditors. $45 per click? People need to know that LI just made that figure up based on cursory examination of competitive fees. They price their traffic based on guesses and and then add 10-30% every few days to see how much people will spend. If they spend more, its worth more... But its the furthest thing from market based pricing as there is literally no open market or visual competition for the traffic for 99% of all ads served on these platforms.

Imagine owning a gas station on an island and charging as much as you want. $50 per gallon, $100 per gallon and then, not even delivering a full gallon! That's the same thing here. Without competition or regulation the prices are set by greed and greed alone.

I dont agree with bowdlerization happening on Hacker News, wrt legitimate titles. This is effectively censorship @dang (I'm addressing it to you because you speak most often here, and set the tone, not because you personally did it. ) https://en.wikipedia.org/wiki/Expurgation

I do understand that having less clickbaity _content_ is noble goal (and reason we read HN), but let US do our job in giving upvotes to good topic or flaging bad ones, while having ORIGINAL titles author of the post intended, not the one the censor considers proper.

I can't read the source article (site is down) but judging from the header I'm kinda glad LinkedIn is not selling many ads. The last years it's become an even more fake stream of glossy PR marketing crap filled with stock photos of smiling faces, and people commenting how much they adore the company line. I get enough of all that on the company's own sites, thank you.

I get that people don't want to be critical in public, but you don't have to repost every piece of marketing BS that comes your way :) Especially when I know some of these people tell a completely different story in person.

And the membership fees for being invisible are ridiculously high now. There's constantly people bothering me with their 'services' which are totally not relevant to my role if they'd even bothered to glance at my profile. I wish I could report them somehow.

The only reason I still have it is because it's almost impossible to get a job without it :) Because the business world loves it. But it's more fake than FaceBook now. Everyone is one successful talented employee who is completely in line with the corporate hive mind.