Automattic trial data and employee retention
Matt Mullenweg does a lot of interviews, and much of this one at the Lean Startup Conference is similar to many others, highlighting how WordPress got started, the culture of open source and how Automattic works. But at the end he mentioned a few numbers about Automattic I hadn’t heard before.
He said that Automattic ends up hiring around 40% of applicants that get accepted to their trial program they use for hiring. I’m not sure how it works for less technical positions, but I know they utilize a code test for some of their more technical positions as a means for pre-screening applicants before a trial. They also do interviews before the code tests.
He also noted how pay works for trials, as well as how they structure them to accommodate people that are currently in other jobs:
We do trials. Everyone that joins the company, we hire them on a contract basis first. Nights and weekends. We’re not asking them to leave them a job or anything like that. So it’s… and it’s a mutual trial. So it’s a standard rate: $25 per hour – same for whether you’re coming in as a CFO or as a Happiness Engineer.
I was also impressed that Mullenweg noted he still reviews every incoming resume, even as they have reached 225 employees.
And finally, he notes that in eight and a half years, only about 10 people have left the company and they’ve “let go” another 25-30.
The trial pay and specific retention numbers are both new to me.
> We’re not asking them to leave them a job or anything like that.
I need to call false on that. As I understand it (ahem, from personal experience) the trial required full-time work, thus requiring one to leave their current job.
Imagine asking your current boss for somewhere between 3-8 weeks off, and telling them that your goal was to not come back at all!
The trial does not in any way require quitting your present job. You can do the trial on weekends or evenings – most of us have done it that way.
Kathryn is correct. I was able to successfully complete my trial while working another full-time job.
I wish I had this comment to refer to about 18 months ago! Part-time wasn’t an option for the t(h)eam I did a trial for. I don’t know if that was an official position or not, but it was presented to me as the only option.
I used to be far more skeptical of the trial program for the reasons people would expect. That said whatever they’re doing, clearly its working. From a possible employee’s standpoint, I would regard the benefit of the trial being that its not a one way trial. Automattic culture is probably not for everyone and if you’re worth your salt, you should be interviewing them, too.
That said, the $25 is insulting. For a company so proud of its benefits and work culture, the rate should be more respectful of the candidates time and energy.
I wouldn’t be surprised if $25/hr is their entry level salary as well, so it might not be that much of an insult to some. It is odd that it’s the same for everyone though.
Where I come from, I know people who would give up a body part for $25/hour. I would not be insulted at all by that wage. Of course, a lot of that depends on where you live (New York City vs. smalltown Alabama, for example).
As far as i know they are hiring remote workers around the globe, in my country experienced PHP programmer in mayor city, can make around $12/hour before taxes (sic!), so $25/hour on a trial is hardly insulting.
> That said, the $25 is insulting
It helps insure that the reason the candidate is doing the trial isn’t for monetary reasons – the trial projects are small, or should be for strong candidates. It’d be nice to see data on the typical # of hours of work paid for trial projects.
Having people do the trial while staying in the current jobs also filtered on those candidates who were very good at managing their own time and energy, two things very important for working well in Automattic’s remote work culture.
While I was there (2010-2012) they were a highly desirable company to work for, and still are, which means they can have their pick from the best (in their evaluation) applicants, and these elements helped winnow the field on factors they felt were beneficial to them, just as any major organization would.
I was a trial employee for a period during the summer of 2011, and it was really rough on me to try and maintain the trial period alongside my existing work and other commitments with my free plugins.
We had also just had our first daughter a few months prior and I was trying to get back into a normal schedule again. That, combined with my wife in night classes for college, made it pretty hectic. But I just couldn’t give it my full attention like it needed, not when to pay the bills and keep our daughter fed I needed to maintain my existing income. Which meant, I couldn’t give up my existing work to take on the $25 hour contract work that may or may not turn into a full time position. I couldn’t risk not paying the bills and letting go of ongoing existing client projects in the pipeline. My trial project lead was also on a weird schedule for a bit which made it hard to sync up for a little while.
The environment was very lax, I found myself constantly wondering if I was doing enough, doing a good job, and if I was good enough of a developer to be on a team of people with way more experience than me (at the time).
It would probably be easier for other people, but it was a mix of bad timing and not having the financial cushion to gamble on something like the trial period with no guarantees.
It’s not that Automattic is a bad company to work for / with, just that I found it particularly hard under the circumstances I was in.
Also Mullenweg wrote this article about trials for HBR:
Comments are closed.