I'm going to offer the contrary advice that hiring by employee referral is not a good approach, for these reasons:
1. Your genius employee does not necessarily know lots of other geniuses. In fact, he might not know anybody. Yes, there are brilliant programmers and engineers who are well-connected, but there are ten times as many brilliant programmers and engineers who are solitary, independent types. If you press them for a referral, you'll get a referral to his brother-in-law who took some computer courses or the guy that was nice to him in university but hasn't found a job yet.
2. Friends hiring friends leads to cliques. For a startup, two or three friends working together is ideal. But for a larger enterprise, you want to avoid that. Of course, cliques will form by themselves, but why create them from day 1.
3. It's very cynical to say this, but chances are that your employee is not going to recommend someone smarter or harder working than he is. Why make himself look bad? (One of Paul Graham's articles says that this is true of mediocre programmers, but not of the top ones. Perhaps when you're starting out, you want to work in a room full of geniuses, but what if you are the top dog in the room? Are you going to bring in someone that can obviously show you up?)
Many hiring methods lead to negative quality direction. Example: Making all your hires from Monster.com would be a negative trend. But a lot of people seem to believe that hiring based on employee referral is a positive. I'm sure it's not the worst way--and it's probably much better than relying on Monster.com for instance--but I think that employee referrals lead to quality loss over time.
We should acknowledge the second edge of this sword. When a company invests in its employees' networks, it not only expands its ability to recruit more people directly, it also expands its employees' ability to find new jobs.
This may still be a net win for everyone, but it's complicated. Imagine you run a company and have identified a few bad employees. Surely you should not reward them by sending them to conferences, right? But wait, maybe you should, because then not only might they spend time learning, recruiting, and doing other things that might be more productive for your company than whatever they're currently screwing up, but they might also find a new job which is both better for them and solves your problem of how to get rid of bad employees without undue liability.
So send your best, most loyal, most enthusiastic employees out into the world to meet other people and spread the word about your company. And send your worst ones too!
If you can identify legitimately bad employees, fire them; they are not just a waste of money but an anchor around the neck of the employees who aren't bad.
Of course, this assumes a lot of things such as that you have USA/California style "at will" employment, you are capable of determining that the employee(s) in question are legitimately bad and not just hamstrung by stupid process or demoralized due to reasons you could reasonably fix, etc.
Here's a wild ass idea: have you engineers do all the resume reviews, screening calls, and interviews. Leave HR and most of management entirely out of it.
Then just hire people who can intelligently speak to a relevant resume, and who get along well with your group. If they don't work out, fire them.
People spend entirely too much time obsessing over how to hire people, usually mumbling some nonsense about how they'll destroy their team or be sued out of existence by a bad hire. Personally, I think it's a combination of desperation to cover one's own ass at all costs, and providing an illusion of control. If they'd focus on improving the workplace and their own communication rather than fretting about voodoo hiring theory, they'd be much better off.
Thats what we do. We give HR a spec, they handle talking to recruiters etc.
We get a stack of CVs. We break them into a pile that sound OK and we give each a telephone interview.
If the telephone interview sounds good we send them a scenario that should take a couple of hours, we have a few tailored for different specialities. If we get a good sample back from them we move onto face to face.
Face to face tends to be about an hour with a couple of senior developers and a more management type.
Seems like a poor prescription to ONLY go off of referrals. The bigger the company gets, the more likely people will not want to bring in talent that they might view as competition. I know that's cynical, and perhaps placated by referral bonuses, but it could happen. Also, what about people straight out of college that find you on LinkedIn and apply to a job posting (which I did in NYC since I knew no one), and end up getting trained to be a senior employee? There are just too many scenarios where you're limiting your pool of prospective employees, why not try everything and double down on what works without discounting new opportunities to bring talent in.
1. Your genius employee does not necessarily know lots of other geniuses. In fact, he might not know anybody. Yes, there are brilliant programmers and engineers who are well-connected, but there are ten times as many brilliant programmers and engineers who are solitary, independent types. If you press them for a referral, you'll get a referral to his brother-in-law who took some computer courses or the guy that was nice to him in university but hasn't found a job yet.
2. Friends hiring friends leads to cliques. For a startup, two or three friends working together is ideal. But for a larger enterprise, you want to avoid that. Of course, cliques will form by themselves, but why create them from day 1.
3. It's very cynical to say this, but chances are that your employee is not going to recommend someone smarter or harder working than he is. Why make himself look bad? (One of Paul Graham's articles says that this is true of mediocre programmers, but not of the top ones. Perhaps when you're starting out, you want to work in a room full of geniuses, but what if you are the top dog in the room? Are you going to bring in someone that can obviously show you up?)
Many hiring methods lead to negative quality direction. Example: Making all your hires from Monster.com would be a negative trend. But a lot of people seem to believe that hiring based on employee referral is a positive. I'm sure it's not the worst way--and it's probably much better than relying on Monster.com for instance--but I think that employee referrals lead to quality loss over time.