Why not use a headhunter? Find the most aggressive, well reputed agency in the area and they’ll do the job for you. We’re pretty much in the same boat and I’ve never gone more than three months before placement even without a job over the past 12 years
But it’s not just inefficiency for me. I have a skillset that is a common enough commodity there are thousands of openings, but rare enough that there are openings, the positions aren’t all filled at the present. (embedded CS, Linux device drivers, BSP, that sorta thing).
I would like to efficiently find what the best offer possible out of all these thousands of opening is. Also I would like to even be seen by a particular company if I wanted to work there.
I can do neither. There is no practical way for me to apply to every company offering a job with this kind of commodity skillset, and even if I set out to do this, each application system wastes my time one way or another. Moreover, even though I have strong experience that means I can work on a massive range of embedded systems, if I look at specific listings, it is not uncommon to see demands for something I haven’t touched - and almost no one else in the world has. Rarely used CAN bus analyzers, some specific software package for PLC controllers, that sort of shit.
On the company hiring side, that position sits open because they can’t get a qualified candidate very quickly unless they overpay or take a large risk.
Instead, well, I know some people. And they know some people. And instead of having access to the thousands of jobs that I am qualified to do, this system you are defending means I can only access maybe 10? 100? total job openings with a personal recommendation. I mean they have to be in a company a friend that knows me well works in, and they have to have this specialized skillset required and the position is open…see what I mean?
It’s like trying to sell a pink iphone. People buy them all day on ebay and it’s a common commodity but maybe all your direct friends either already have a phone or don’t want a pink one.
Yes, the job matching mechanism has become ridiculous. I believe this is for several reasons:
-
The bifurcation of the economy into “high level” (investment bankers, lawyers, full stack engineers, etc) and “low level” (food service, retail, etc) service jobs.
-
Overreliance on HR systems (ATS) to screen resumes and make hiring decisions.
-
Treating workers and disposable cogs to be hired and fired as needed.
-
Perceptions that you need to hire superstar “unicorn” 10x employees to transform your company.
-
The relative ease for people to spam resumes and overwhelm recruiters.
-
The rapid pace of technology change, leading skills to become obsolete much faster.
Except that’s not really true. And it’s that myth of hiring “superstars” and setting that as the standard that I think causes job searching to take much longer and generally just makes work miserable for everyone.
Yes, for certain roles, there are superstars. Like the full-stack engineer from Google who built one of their major platforms. Or some well-connected industry thought leader who goes to work for McKinsey. But the vast majority of the job market are normal people looking for normal jobs.
You have friends and family? College alumni? Do you go to Meetups and industry groups? Do you interact with customers, vendors or alliance partners from other companies?
Other than it has become a source of spam from every headhunter, financial advisor, sales rep and tenuous connection you’ve ever had with something to sell?
Yeah, it is inefficient. I am not making 10K, per month.
How are we measuring efficiency?
Efficient from the employer point of view? Just like anything else, firms may have an advantage in efficient hiring practices. I’m not in HR so perhaps someone can weigh in on how they evaluate their own hiring practices. Maybe they measure the time and money spent in acquisition in relation to the value provided by the employee. I imagine many companies do not have an adequate means for measuring employee value in many occupations, but some like sales seem like they would be easy to measure. In any case, they should be making improvements in hiring practices that are guided by data. This is no different from the evaluation of other business processes. There seems to be a mechanism that leads towards more efficient hiring practices.
Maybe you are talking about efficiency from the applicant’s point of view? This is much simpler. Every individual has the means to measure the resources he has allocated to job acquisition in relation to forecast benefits. This depends largely on personal preference. Some people see jobs that they want and develop a plan to attain them. They attend college, get internships, study, develop intangible good-employee habits. Now they are qualified. They get a job. This was quite a bit of effort, but they now have the job they struggled for. How is this any different from later career job searches? You expend resources (time and money) in order to get the job you want. If the result is accurately forecast, and the effort was “worth it”, I don’t think this is inefficient.
Maybe you expend effort and don’t get the job you struggled for. Is this “inefficient”, probably. But it is a result of your flawed forecast.
If you are saying that job search processes are too resource-intensive, perhaps. Sounds like an entrepreneurial opportunity. There are a multitude of approaches to take for a job search. Which approach you take is job-dependent. Perhaps you have identified that for your occupation, there is a lot of work involved in getting a job. There’s also a lot of work in getting a job for a surgeon. Many years of learning. Maybe it is because these tech companies are not mature, and the best job-search technique has not been identified. Maybe it is because individuals do not make data-driven decisions when approaching a job search, which is understandable as developing such a process may take more time than it is worth.
You’re not wrong, but I think this is partly where things fall down for employers. There are a few “data points” that they can’t or just don’t want to measure. For example, a lot of highly skilled candidates refuse to take personality tests or unpaid multi-hour projects to prior to being granted an interview. (This seems to be the new trend among hiring companies even for white collar professionals.) It’s not that the candidates don’t want to demonstrate that they can do the job. It’s that the companies expect these things from candidates before they even do more basic screening like discussing the job requirements and salary expectations. Putting the cart before the horse, if you will. You can’t measure how many really good candidates you could have hired but lost due to these practices because when they bail out at the entry point of the process you can’t distinguish them from flakers and crappy candidates (because you haven’t evaluated them yet).
It’s also hard to measure candidates who are lost because the interviewers themselves are bad at interviewing. I have an anecdote (that I’ve heard echoed by many others on the jobs subreddit) where the hiring manager asked me to meet some of the people I would be working with. One of them was very argumentative with me, and despite putting on my best “customer service” smile and attitude I couldn’t figure out what answer she was looking for. I left that interview thinking that I wouldn’t want to work with a bully so I lost interest in the job, and since they ghosted me I assume they felt that I was the one with the attitude problem. It’s natural for the manager to side with his staff, but sometimes it actually is his staff who are the behavior problems.
Which brings me to another issue: the significant increase in ghosting candidates. You can’t measure data when you … well, don’t measure it. When you ghost a candidate instead of asking them how they felt the interview went, that’s a lost data point.
Also not giving the “correct” answer.
Once on an interview I was akised do I prefer working by myself or with a group. I said both. Not good enough. They wanted me to choose. I went back and forth that I was fine with both but finally said maybe by myself.
But they took that as the wrong answer since they wanted people who said “group”.
The advantage firms have is that they are the ones with the jobs. And there are a finite number of them.
But I think what we are talking about in terms of “efficiency” is job seekers matching up with paying jobs.
I’ve been a part of the hiring process at a number of places I’ve worked. It is absurdly inefficient. Usually consisting of rounds and rounds of phone screens and interviews by people who often have never been trained in interviewing. At the end, you are given a job (or not) largely based on “feel”.
Later career searches are very different from entry level career searches. For entry level, companies typically actively recruit new college grads into specific roles. The paths are generally (but not always) pretty clear. Even if you don’t have a perfect degree for the role, a lot of companies will hire candidates because they demonstrate achievement and aptitude through their academic and extracurricular achievements.
Later job searches become more challenging. Careers do not follow a uniform path, even within the same industry. Not everyone has the same aptitude or success. A person might struggle at one company but excel at another. Some people have missteps in their career they have to recover from. As you get 10, 20, 30 years or more into your career, you develop more “baggage” that you have to “explain” to a recruiter or hiring manager:
-Why did you change jobs so frequently (or why haven’t you changed jobs more frequently)?
-Why have you been at your current level so long?
-How familiar are you with the latest tech that maybe your last company didn’t work with at all?
-Why do you have a gap in your resume?
-Is this person’s skill set too deep and narrow or too broad and thin?
-Are there certain skills missing? (Or are they actually there and he just assumed he didn’t need to put “proficient with Word”?)
Of course, you will likely never get to explain any of this in the six seconds a recruiter will look at your resume before deciding which pile to toss it into.
You can’t “analyze” your way into a job.
The entire problem is that the “formal” approach for applying to and obtaining nearly any corporate job does not work. That is to say, going to their career page and applying for a job that matches your qualifications. 99.9% of the time, your resume will go into a black hole and never be seen again.
So companies and applicants have created all these “back doors” - networking, recruiters, LinkedIn posts, so on and so forth - in an attempt to circumvent the process. Which ultimately makes the process more chaotic.
And assuming you actually get the attention of the company, the process ends up being weeks or months of phone screens, Skype calls, batteries of in-person interviews, all designed to see if you would be the perfect fit. After which, you might not get the job, as presumably you are not the only candidate.
I think this one is the primary driver, at least from the vantage point of my career.
Companies seem to be approaching hiring as a much more transactional event than in the past. From what I understand about the past, and what I recall about my first jobs 20 years ago, hiring someone was very much about the long term- you were making sure this person had the basic skills for the job, but you were mostly trying to make sure you were finding someone who would be a net asset for the company- someone whose ability to learn, grow and innovate would help the company over the long term, even if they didn’t fit the exact requirements for the job position you had. The expectation was that you could figure out the parts you didn’t know.
Today employees are viewed as much more modular- we have a “slot” that requires certain skills and experience, and we want to fill that slot in the short term with the expectation that the person who was hired will walk in and start working at a high level.
So you start ending up with situations where someone would be a terrific fit for a job, but doesn’t get a look because he has experience with a different ERP system in the same industry, or a very similar, but still different programming language, or they don’t have audit experience, but are accomplished accountants otherwise.
If you don’t like dealing with all that, go work in a small industry, or a niche within an industry. I don’t have to do any of that stuff because I work in a field where I am no more than one or two degrees of separation from everyone else in the field.
I’m not sure how what I’m saying couldn’t be true. I agree that “superstars” are rare and that some companies have an unreasonable belief that they can hire them at below-market wages. But that doesn’t invalidate my general point. What I’m saying is that there’s a continuum of skill/desirability, and at the high end, people do not stay on the job market for any time at all (the superstars), and on the low end, they submit hundreds of resumes and are never hired (or are soon looking for another job).
Sure, there are plenty of people in the middle. People who probably consider a dozen or two jobs and maybe get to the interview stage with five and get two offers. But the low end is overrepresented in any sample of resumes or applicants. Because they keep applying to jobs and not getting them. Any process that makes it easier for the “normal” people to spread their job search far and wide will be inundated with unqualified applicants.
And as the higher skilled people realize that the slush pile is an inefficient way to find work, they find ways to get out of it. All the ways that people have mentioned.
Are there really that many “slush pile” people though? Eventually I’d think they have to get to a point where they’ll drop their expected pay, or otherwise apply for jobs they can do adequately.
Sure, but there are always new job-seekers appearing as the old ones come out of the applicant pool, and most of them can’t or won’t limit their search to jobs they are qualified for.
Note that there don’t actually have to be that many of them to make the slush pile problem pretty bad.
Like, imagine (throwing out numbers for demonstration purposes, not claiming these are an accurate measurement of the population) that the job market consists of 95% who apply to jobs in a reasonable manner and 5% people who apply indiscriminately to many jobs that they are unqualified for.
The reasonable ones know what their qualifications are, they spend some time to determine which jobs would be a good fit, and they focus on applying to a handful of jobs that they have a reasonable chance of getting. Those people maybe apply to 10 jobs a month. And they get a job after a month or two.
The unreasonable ones can easily apply to 100 or 200 jobs a month. All they have to do is figure out what keywords to jam into their resumes.
So, even though the field of job seekers is heavily weighted toward reasonable applicants, the population of people applying for any given job has a strong bias toward the unqualified.
Mechanisms that make it easier to apply for more jobs make this problem worse, not better. Reasonable people aren’t going to apply for 1000 jobs that they’re only remotely qualified for, but unreasonable people will.
Another thing that impacts the current job market is the fact that so many Boomers can’t afford to retire, so they’re trying to hold onto jobs later in life, which blocks young people from entering the market (in aggregate). Laws that push out legal retirement age are also causing this.
Either you haven’t been in the job market recently or you’re being too flippant. What you describe really depends on your geographic location. For example I’m in a major metro area, a hot job market and have in-demand skills. While I only applied to around 5-20 jobs each month, it took me 1.5 years to land a new gig. That search included two actual job offers that fell through (one was yanked due to incompetence of the company, the other one I lost due to my own fault). The two failed offers were 9 months apart and I landed my current gig 5 months after the last failed offer. It’s a numbers game for all but the 1% best candidates these days.
I haven’t been in the job market recently, but I wasn’t intending to be flippant, just illustrative.
Whatever the numbers are, can we agree that the continuum I suggested exists? Higher skilled people apply to fewer jobs and find a job faster on average and lower skilled people apply to more jobs on average and are out there looking longer/more often?
Yeah that’s terrible. What if your employer had gone under or laid you off during that period? Not having a job cuts your odds down even further.
And it doesn’t have to be your fault. My current employer did some layoffs, then figured out they accidentally eliminated a key position and are now in the process of refilling it…
I am from a field like that and I have always worked and dealt mostly with very small businesses employing less than 25 people. And virtually all hiring was done via word of mouth and personal references. And while there are some defined job descriptions, more commonly hiring was done because a certain person had become available or someone’s kid needed a job. So these companies were constantly moving around and shifting job descriptions and responsibilities to fit the skills of the employees of the moment.
This was at its worst inside the sales agencies, in my business they were selling products to builders that had been specified by architects and designers. And doing personal favors for architects and designers was a key business tactic. And friends and kids need jobs. I’m surprised I survived in that world and it was a fluke that I got in in the first place. Because I found my job through a help wanted ad. And I was with that company for 10 years and never saw them advertise another job opening.
You’re flirting with the Lump of Labor Fallacy.
Were you actually employed during this 1.5 years and looking for a job, or were you literally unemployed?
If there’s one thing that hurts you in trying to find a job, it’s not having one to begin with. It’s like prospective employers assume you’re in iamthewalrus(:3='s “slush pile” and are out of work for an actual job performance/ability related reason.
It’s like job hunting leprosy or something. And it’s completely stupid.