There's a certain group that started out as a marketing job placement service then turned into its own marketing agency that takes the best jobs for itself, and now draws on its membership to offer supposedly helpful presentations that promote a certain business that is likely paying for this form of content marketing to a live, albeit virtual, audience. One of these is scheduled for today. This is the banner for it with the hosting organization cropped out.
If this organization that is not able to survive on recruiting fees any more wants to branch out into charging businesses for representation on their platform, that's fine. But they should be transparent about sponsored content and not try to hoodwink people with bogus statistics that don't even fit the argument made for using AI in today's job market because those statistics are all rooted in a job market that predates this reality by years -- if the stat ever had any validity at all!
I put in the highlights to make it clear why this irks me so much. If this were a student paper, I'd give it an F despite its mechanical correctness and superficial polish.
1. There is no citation for the source of the stat with what is actually considered a fairly large range of 70-85% that undermines credibility. If you did a real survey, you'd have a specific number and not have to hedge your bets this way. And if you can't tell me where you found this out and what the context of it what, I don't believe you at all. You could be making it up altogether or be sharing something that was based on a sample size of just 3000 jobs 10 years ago (as it turns out).* There's no way of knowing if this is current and truly relevant if you don't share that source.
2. The second stat, "up to 80% of jobs are secured before they’re ever publicly posted"**makes no sense at all. Yes, I'm sure that many jobs are destined to go to the founder's son, his neighbor's daughter, or his roommate from college because nepotism and cronyism still dominate the world. But once the job truly is secured, there is no reason to post it. It's like making a claim that most houses are sold before their open house. If an agent has the contract, she is not going to run the open house (despite the fact that open houses are about gaining potential buyer lists rather than selling that particular property) because admitting to people that the house is not for sale undermines her credibility.
3. The structure of this is meant to lead you on to point 1 and point 2 add up to 3. But it doesn't at all. It doesn't follow from any of this that AI gives you a leg up in job search. What does give you a leg up is being related to or friends with the people who are making the hiring decisions. No matter how amazing you are at AI, you'll be passed over for the manager's college pal or cousin.
The more you know, the less you trust
I asked the woman who posted this the source of her statistic. She answered a few days later:
The Openarc source is dated 2025. So you'd think the stats are current and should take into consideration how things changed during the pandemic and the incursion of generative AI. But you'd be wrong because Openarc is referencing a 2021 source https://www.thomasnet.com/insights/85-of-jobs-get-filled-through-networking-here-s-how-to-grow-your-professional-network/, and that 2021 source is citing a February 2016 LinkedIn Pulse survey inkedin.com/pulse/new-survey-reveals-85-all-jobs-filled-via-networking-lou-adler/ that relies primarily on 2015 data drawn from a survey of just 3000 people. Here's the graph it features: *
In other words, this woman is presenting herself -- or the man who is going to give the presentation to market his company -- as an expert by throwing out bogus stats. What was true 11 years ago could still be true now, but we don't know that. Certainly, the job market has changed drastically in that time.
Now let's look at herd source for the claim that 80% of jobs are secured before they're posted. It doesn't substantiate that at all. On the contrary, what the Fast Company article https://www.fastcompany.com/
The more you look into, the more she loses credibility. She's either deliberately misrepresenting 11 year-old stats as current and never established stats as valid or she doesn't even know enough to realize how absurd it is to rely on them in 2026.
This is ridiculously common among lazy, incompetent, and not altogether honest writers who provide outdated and even clearly unsubstantiated information. When I want to cite a stat, I always trace it to the original source to get the full context, as I wrote in Why you always need the original source.
Posts
