What If You Could "Date" Companies?
fastcompany.com
fastcompany.com
The freelancer who works a gig at a company for a couple months before finally accepting a full time job from them is one of the oldest stories in consulting.
Done correctly, freelancing tends (strongly) to be more lucrative than full-time work, most especially in markets like this one, where sellers of talent have leverage over buyers. Why would anyone in a market like this concede anything to a company like GroupTalent? Serious question; I'm sure there's some reason.
That's also why I balk at companies that give take home tests to screen applicants.
Since you think about hiring a lot, I'm curious: do you think the above is accurate? If no, how would you phrase it differently? If yes, how would you recommend testing that hypothesis instead?
In addition to the argument that the economics of temp- to- perm don't work out in this market for tech talent, there are also fundamental problems with the approach.
Most importantly: lots of people can perform well for short periods of time, especially if they don't have to deal with interpersonal conflicts (which take time to develop) or technical design disagreements (which only happen at inflection points in product/service design; ie, it might take 8 months to discover whether someone can "disagree and commit" to work on a team where they disapprove of some technical decision). Also, lots of people can work well on a project for a couple weeks before they get bored, and some of those people lack the ability to kickstart their engagement back up again once they feel "burned out" by a problem domain.
I think those are some of the most likely failure modes of new employees, and they also happen to be different from the failure modes that classic interviewing and work-sample testing are likely to suss out; ie, you can cover the non-psychological non-work-ethic concerns about new candidates with a relatively straightforward hiring process, so temp- to- perm might not even buy you much even if you don't assume that the approach is costing you the best candidates.
The problem is that companies looking to fill full-time slots are looking to fill full-time slots, not to engage a succession of contractors.
By requiring candidates to perform in a contractor role prior to signing on full-time, companies are effectively disqualifying most candidates that are currently employed. That's because very few people sniffing around for a better job than the one they have now can afford to drop out of the market for months or even weeks delivering work as a contractor. That's not how most people switch jobs; instead, most people search for their new job from the comfort and security of their current job, knowing that if (say) interviews don't work out, they can take their time looking for the next lead, knowing that they can continue to execute on their current job.
This is a problem for employers because virtually everybody you'd want to hire in this market is already employed.
Typically they offer to take a slightly-low full-time weekly salary+benefits number, divide it by 40, and pay that as an hourly rate. Oh, goody.
No company wants to pay a contractor the going rate for contractors, and then face the unenviable task of talking that contractor's rate back down again when making them a full-time offer. Nor do they want the bother of haggling over the details of a contract deal that will almost always prove, one way or the other, to have been a wasted opportunity. (If the worker proves to be competent, the contract will be replaced by a full-time offer as quickly as possible. If the worker is incompetent, the contract will be terminated at the earliest practical date. In no case does anyone care much about the contract. It is a consolation prize for everyone involved.)
My personal philosophy is to treat contracting and employment like oil and water. Both can be tasty. But they are different things and one does not merge seamlessly into the other.
Anyway, I learned. Thanks.
The problem here is obvious: the best candidates don't ever have to put up with stuff like this. They're actively courted by prospective employers.
Sure, not every candidate can be "one of the best candidates". But why would a rational employer select a screening process that overtly rules those candidates out?
The only time it is ok to ask for a take home assignment is when the person is at most a couple of years out of school, and they haven't proven themselves yet.
I personally refuse doing to do take home assignment as I have enough projects in my belt that I can show. It is usually a sign that the other party is not mature enough, and they don't know what they are looking for, and once you join that company they will treat like a junior engineer, when you might be past that.
Maybe yours is different, but I think the first problem is they typically have a lack of clearly defined goals. I know that probably is the idea behind these - to see what someone comes up with - but it can be very intimidating. Especially given the fact that the interviewee is already nervous enough and now they are given something ambiguous. I think that's more nerve-wracking than the interview itself.
Second, I think that code-style tends to get graded on these more than the content. Unless it's completed in some really novel way (which isn't always a positive) there's nothing to do but nit-pick how the code looks.
Third, its disrespectful of the interviewees time, especially relatively large coding assignments. If a candidate has a job, is it fair to expect them to go home and spend a couple hours working on something that is essentially worthless?
Finally, typically the candidate gets no feedback. I had one that was a 'pass' and I never found out why. You are essentially grading me on an assignment and then don't have the decency to tell me what was wrong? This is related to the previous item as a time-waster.
So, those are my reasons. Just my two cents.
(1) Yes, ambiguity is death. It's bad for the reason you say it is (it psyches you out, and discovering how people perform when they're at their most psyched-out is a pointless exercise). But it's also bad because open-ended tests are hard to grade and compare. Work-sample tests need to be standardized and they need to be defined rigorously enough that you can draw comparisons between candidates, and then draw comparisons between the performance of hires and the tests.
(2) Yes, grading superficial issues is also bad, for the same reason: when you give someone "points for style" (or demerits), you lose the ability to make apples-apples comparisons with candidates over time.
(3) I flat out disagree that take-home work is disrespectful of candidates. Most interview processes demand that employees submit to several (3+) hour+ long phone screens, then show up at an office for 1-2 days of interviewing. Allowing candidates to do the most important technical portion of the interview on their own time, on their own schedule, in their own location is simply better for the candidate than forcing them into a room with a whiteboard.
(4) Yes, not giving feedback to candidates is one of the great sins of tech hiring. Some of this is folkloric; for instance, there's a belief that providing feedback to candidates you opt not to hire sets you up for frivolous complaints down the road. But there are ways around that problem.
As for #3, if you aren't a fairly exclusive 'brand' (like Google, Amazon, maybe Facebook) and you are known for having a long interview process, I think a fair number of applicants won't even bother. I've been asked by Amazon to interview a few times and their process just seems like a big pain for not much gain. Unless I was right out of college I just wouldn't find it worth the effort.
So if a recruiter tells me an interview process is extensive and way too involved for the market, I'll just pass. When jobs are plentiful a candidate can afford to be picky.
EDIT: sentence clarity.
Simply put, GT gives developers deal flow.
I'm not suggesting that you need to be Patrick McKenzie to execute this strategy.
This doesn't scale, but I'm glad for your startup it works. You like it because you don't have to pay unemployment/benefits for someone that you end up firing, probably, so of course you are going to support something like this (being the one hiring).
I don't care about not having to pay unemployment/benefits. The reason I like it (or rather, need it) is because our organization is little more than the employees we have, and I want to make sure that they are awesome.
I like contract-to-hire overall, but one problem is that you get lifetime contractors who have no intention of ever converting to perm roles, and they just stay as long as they can. Companies are more inclined to invest in and train a contract-to-hire employee over a pure contract employee. The brain drain can be a problem.
Another issue is contract-to-hire employees usually are paid a premium, and get accustomed to the money, making them less apt to convert. This is particularly troublesome for those that get their benefits through their spouse/partner, where the extra money seems substantial because they aren't paying for benefits or COBRA out of it.
Probationary periods are rather useless in any at will employment state (in the US, not sure about international law). Why put "We can fire you in 90 days if you don't perform" in an employment contract when you can simply fire someone on day 2 if you really wanted to? It seems like a decent motivational tool perhaps, but in markets like software it is just another reason for candidates to potentially turn down work, where the alternative (no probationary period and no expectation of even 90 days of work) is better for both parties.
"At-will" means that companies have the right to execute strategic layoffs, and also to set performance standards whereever they wish, as long as they're uniformly enforced across that job description. (Both of these, I'd argue, are reasonable.) It doesn't allow companies to vary their performance standards for different individuals, or to fire "for any reason". For example, if you work for a 5,000-person company and you're fired for performance after failing on one project and being denied transfer, and you can prove that someone else was allowed transfer under the same circumstances, you can possibly win a discrimination or harassment case (especially because any negative communication about you by management that impedes your chance at transfer can be construed as harassment; this is why the common practice of making performance reviews Enron-style-- that is, part of the transfer packet-- exposes a company to severe legal risk).
If you apply to a 5-person company and get fired in Month 3 because your boss (probably the CEO) isn't feeling it, you probably don't have legal recourse, because as owner and manager of the business, he has the right to select personnel. If it's a 5,000-person company, and you weren't give a few chances to prove yourself on multiple projects, you can raise accusations of personal bias. (That's why companies implement those horrible 18-month policies against internal transfer-- it's to create the impression of a uniform performance standard-- however, those only have legal weight if enforced uniformly, which they never are.) Even if you don't win, you can bring a lot of HR records into fresh air (which a company doesn't want) as they try to prove that you failed according to a set of rules that is enforced uniformly. Companies would rather pay severance than have the courts getting into their HR records.
The probation period is generally more "at-will" than regular employment. For most companies, probation means a couple of things. First, it means that a no-show constitutes voluntary resignation and therefore can't require the company to pay unemployment. (This isn't an issue at our level, but most people are not as ambitious and diligent as we are, and no-shows happen.) Second, it blocks the "employee should have been allowed transfer before termination" argument because the going assumption is that no one transfers in a probation period. Third, it often means no vacation is accrued, which is relevant when people are terminated early on (because they usually haven't used it). Fourth, and perhaps most relevant, it means that one should not expect severance if terminated in that time.
This is somewhat misleading. It definitaly allows them to vary their perfomance standards for different individuals (just not to discriminate on a set of specific prohibited bases) and to fire for any reason except for set of specifically prohibited illegal reasons. Having consistent standards (both in theory and in documentable enforcement practice) for everyone isn't a requirement, it just helps you prove that the firing was not for a prohibited reason, and that the standards you were applying to that particular person were not discriminatory on one of the specifically-impermissible bases.
Many of the things that people perceive as "legal requirements" aren't actually legal requirements, they are things that are common practices designed to demonstrate that the actual legal requirements are being met.
For one example that can be played (by employee) to a win, the most common case of a good person being fired seems to be one in which he's seeking transfer to another team, and the manager decides to punish the "disloyal" report, often with negative reviews and "performance improvement plans" that block transfer outright. In that particular case, it's arguably retaliation (over a legitimate use of internal mobility procedures) and harassment (negative communication interfering with job performance) for the manager to be doing so.
Technically, if a manager interferes with a report's ability to transfer, that is harassment, and bad performance reviews qualify in companies that use Enron-style reviews (i.e. reviews that are part of the transfer packet). That's why Google has secret calibration scores: it's a CYA move. Managers can give positive verbal feedback and annotate it with a smear that the employee never sees. This makes it less likely that a report's lack of mobility will result in a lawsuit.
That's why if you ever see a PIP, you should never sign it, and in fact leave the room immediately. "I have to go to the bathroom". Then you start your internal mobility search immediately. If you do it properly (this includes that you were never there to see the PIP) then the pursuit of transfer happened before the PIP, which hints at a retaliatory context.
This move doesn't work under a probationary period because anyone who seeks transfer when on probation is (reasonable man standard) deemed unwilling to follow workplace policies-- or, more informally but to the point, a flake.
No, its because having those procedures in place is how they prove in court that they did nothing wrong. If you can prove that you did it for a specific legitimate reason, that proves that you did not do it for a specific illegitimate reason.
Now, its true that that also means that a case is less likely to end up in court, but that's because the evidence exists for use in court, its not some feature of the process independent of its evidentiary value in court.
For example, most of the time, a person will pursue a case if he or she perceives moral superiority over the other party. So the employer's goal is to equalize that difference, either by bringing itself up or by bringing the employee down.
Severance payments are there to make the employee feel good about the company. It didn't work out, but the company was fair and everyone can move on to a better arrangement. This is a decent way of doing because, after all, most people don't want to sue ex-employers. They want to move on with their careers.
A PIP's purpose is to make the employee feel bad about himself. PIPs actually make it easier for employees to sue, but are justified on the basis that they will be less likely to do so if they think they have less recourse. I have no idea how this works out empirically.
That's not in dispute. The point you are missing is that the value of the practices I was addressing (which weren't, you'll notice, PIPs) is still the evidentiary value those practices provide in court, because that evidentiary value is itself the mechanism by which they prevent disputes from ending up in court in the first place.
I think you are oversimplifying what probationary periods mean. I've worked with (not for, but represented) more than one employer that had a so-called probationary period, and both looked quite different. One did not provide benefits during that period but vacation did accrue. A probationary period can vary by employer.
If they try to claim that it's "for cause", then they have to say what the cause is. If you're the only person fired that day, it's not really a layoff.
Many companies will give you a decent severance and not get into specifics, though. In that case, you should take the deal and move on.
This stuff's really complicated, with a lot of variation by state, company, and circumstance. However, the "can fire for any reason" claim is blatantly incorrect. Companies don't pay severance or run "performance improvement plans" because they're nice; they do it because at-will employment is very complicated and firms are risk-averse.
In addition, I'm not sure how contract-to-hire for 4 months is different from hiring someone for 'full time' and then firing after the few first months if they don't perform well.
In this market, if you are a good developer, the company will most likely want to hire you. In my experience, it is the company that needs to court the developer.
Oh wait, the article's just a PR piece for a startup using a deceptive title instead of something more honest like "GT takes the dating model to the recruiting process."
[1] This might not actually be the case but their info pages only talk about apps.
Essentially, it's a system that allows people to collect some stability income upfront by selling call options on future consulting services. This would also allow intermediaries (like GT) to arbitrage. For example, I'm damn sure there's someone out there who'd pay $750 per hour for my time (or, at least, could justify that much value) but I can't find that person-- and he probably only needs a few hours of my time, here and there. I'd be happy to sell a call option (strike $100/h) for $150/hour and let the option holder take most of the profit ($500) in order to have him doing my marketing and finding those $750/h high-value clients (which I can't).
The one way my consulting call options would differ is that the option holder only gets 95% of the profit; 5% goes back to the consultant. That's to make sure the consultant knows the actual price he or she is billed at.
Employment is already a call option, really. Take software engineers. The median of us is worth less than our salary, but the good ones are worth 5x and the great ones are 20-100x. That's hard to predict, because it's not only a factor of the individual, but also project/person fit and being assigned to high-yield work. Many software engineers at large companies could be rendering over $1 million per year in value, but are being held on evaluative Fourth Quadrant Work ( http://michaelochurch.wordpress.com/2013/01/01/fourth-quadra... ) while management decides whether or not to trust them with a real project. So there's already a call-option dynamic in place; it's just that right now, it works entirely out of employee favor. I want to fix that.