> Well, okay, but turn it round: what do you have to lose by listing the jobs?
Years ago, Triplebyte had a blog post on the front page of this very site where we mentioned, offhandedly, that we'd gotten jobs for "hundreds of engineers". That was true - the number was roughly 800 that we could track, probably into the low thousands counting ones we didn't - but there were extended threads analyzing that number to prove that we never actually got anyone jobs and our whole brand identity was a lie.
Had you done the numbers, you'd have found that 800 was totally consistent with TB's stage as a company. At 30k a placement, 800 placements would amount to 26m in revenue; at 10m or so of annual revenue and rapid growth that's a totally reasonable lifetime number. The complaint was, I think, totally unfounded. But that didn't stop it from being the top thread on that post (IIRC) for some hours.
That's the kind of thing that makes it hard to be transparent. As much as we want honesty from companies, we (or at least a lot of us) also want to go with the winners, and we often have unrealistic expectations of what "winning" looks like (because our intuitions are tuned on inflated numbers that are either lying or selected). In fact, just down this thread, you'll find someone arguing we can't really be serious about recruiting because our website was not made by a graphic designer.
> Maybe some that would have given you their CV would be put off if they see only listings for jobs they don't match - but they don't match them, so they're not going to help your business any time soon anyway. So: the ones you gain by not listing don't help you, and the ones you lose are the ones that would.
The problem is that, in the limit, this results in no one ever signing up for anything. Because candidates want us to already have the job they're a match for, and companies want us to already have the candidates they're looking for. And even if clients were willing to go for it even without candidates in the pool, it's not like candidates regularly check to see if new ones are posted on a small site. (If you did do that on your recruiters' site, I can tell you - with the backing of considerable data - that you are in a rare minority.)
One person we have at an onsite literally right now signed up a month ago when we had no roles for him. We had a role that was a fit come in a few weeks later, matched him with it and then with another, and now he's got a 50-50 or so shot to get a job in the next few days. Had he just bounced day one, that wouldn't have happened.
There's an analogy here in terms of chemistry. Trying to get both candidates and companies to both be around at the right moment is second-order kinetics, proportional both to the rate at which candidates check your site and to the rate at which companies have open roles they want you to fill. Trying to get one side to sign up provisionally - and it makes sense for that to be candidates, since job searches tend to last longer than open reccs and candidates aren't having to pay for it - is first-order kinetics, proportional only to the rate on the other side. Since at small scale n^2 is << n, it's really important for us to encourage the first-order state of affairs rather than the second-order one.
Of course, none of that gets at whether not posting all our open roles is consistent with honesty. But you're making a strategic argument here, and I think it's incorrect.