Thoughts on Google’s 20% time

Everybody loves to think one little trick can make their organization transform into a super creative powerhouse. With the rise of Google, no single tactic comes up more in innovation circles than their concept of 20% time. Simply put, employees get 1/5th of their time to work on projects of their own choosing.

For the myths of innovation book i spent time studying lots of concepts, models and approaches similar to 20% time, and even talked to a few Google employees about how they see the idea. What follows below hits on most of the erroneous assumptions I’ve heard people make about the concept.

Here’s a short report:

  • Google’s 20% time is more of an attitude and culture than a rule. First, hourly time isn’t tracked there, so there’s no way to enforce or even know what percentage of time people are spending on side projects. But more importantly, the entire idea seems to function more as an attitude – that new projects should be spawned by whoever has the best ideas, not who is in what place in the hierarchy, and the culture is based on this fundamental belief. There seems to be way more support for the pursuit of ideas generally than in most cultures, and simply creating a 20% rule doesn’t give you that culture. G-mail, Adsense & Google News are three examples of major offerings initiated by a self-motivated engineer. See Google employee Joe Beda’s blog post for one of better first persons accounts you’ll find online.
  • It’s worth noting that people at Google work very hard on their 80% time. It’s not as if every Friday is 20% day and work shuts down on all existing projects so people can do their 20% things. Google culture, much like Microsoft in the early 90s, has a very strong, competitive work ethic, and peer pressure and pride drive many people to work hard. Like many tech companies, the vibe is that, yes, if you have an idea you should follow it, but not to the determent of other responsibilities. Time for 20% projects is protected, but more by individuals than by managers. Managers spend little time tracking engineers (span of control is wide, with managers typically having 10 or more reports, influencing people and code more than “managing their direct reports”). I’ve heard different things from employees in different groups at Google about how this has changed as the company has grown (10k employees and counting) and perhaps the variances in their culture will continue to grow. (Read Steve Yegge’s excellent post on software development process at Google).
  • The 20% time concept isn’t new. 3M developed a 15% time rule in the 1950s with the same exact intentions and basic philosophy. Masking tape and Post-it notes are two notable products that were concieved and developed by individual engineers working without formal budgets, plans or management support. I’m sure other companies and organizations in the past have had similiar attitudes about creativity (Edison’s Menlo park lab likely qualifies). For more on 3m’s approach read this short Wired article. Also, the Google founders mention at their talk at TED that Montessori school philosophy influenced their ideas on 20% time (Jump to 8:50).
  • Google’s culture has a resistance, or even distrust, of hierarchy – they often use voting, peer review, and debate to make decisions or decide which new projects and features to add. With that structure the 20% time idea makes sense as they want self-motivated creatives putting ideas in the hoper for others to review, evaluate, or contribute to, rather than waiting for executives to spend weeks making big vision documents and marketing plans, dividing things up into smaller and smaller pieces, before allowing creatives to make (creatively constrained) contributions. 20% time complements, or perhaps even depends on, what is a unique culture for a large, 10,000 person company. It’s the lack of dependence on hierarchy that empowers individuals, and this is the thing people at more conventional companies have the hardest time comprehending. 3M also had a strong maverick, anti-structure vibe that made their 15% successful. Giving people time is one thing, but it’s the culture of the org they get that time inside that determines how useful that time will be to the company.

20% time experiment: Atlassian, a software development shop, just announced a serious 20% time experiment, adopting the idea in their culture and blogging about it as they go.

Disclosure: Don’t take my word for it alone – While this is based on some research, and although I have visited Google several times, I have never been a Google employee and if you start with the links above you’ll hear from more authoritative sources on Google management and culture than myself. If you know of others I should read, please leave ’em in the comments.

84 Responses to “Thoughts on Google’s 20% time”

  1. Joserra

    I was wondering in my blog (sorry, in spanish) how to move that 20% Atlassian experiment to another kind of software enterprise: To a service oriented enterprise, where you work for projects, not for products.
    I suppose the idea is to leave employees decide how to perform better, for example, but I think this is a different one and not so interesting for most developers.

  2. olie

    Joserra, To apply the time in an environment that doesn’t have projects, the developers work on what needs to be done. This might be tools to make their own jobs easier, automated test tools, tracking tools, automating your own business, etc. Basically, anything that YOU can use, becomes your “product”, even though you might not sell it to customers.

  3. Ben

    Thanks for the comments on having to create a creative culture and not just give employees some dedicated time! I even wrote about it in my own blog!

  4. Peter McNicholl

    What is the ownership rights of the employee that comes up with a successful idea, which gets developed and launched within the 80/20 system at Google.

    Is the ownership 50/50 between them and Google? 80/20? 100% Google?

    1. Carl Gutekunst

      Peter, you have the same fundamental misunderstanding that most people do regarding Google’s “20% time.” This is not about an engineer setting aside time to work on whatever he/she wants; it’s a formal project, recorded in your project planning database. It needs management approval, both for the time and for the project topic, and it’s expected that you including updates in your weekly status report and in your quarterly objectives report.

      What makes it unusual is you can work on something completely outside your normal responsibilities, so (for example) a UI designer might work on an operating systems project. Or an Email developer might take on a security project. Many departments also solicit Googlers from other departments to take on 20% projects for them.

      Whatever comes out of the 20% project is obviously 100% Google’s.



  1. […] Thoughts on Google’s 20% time – great article from Scott Berkun on Google’s famous 20% time. Note: Google do not have timesheets! The 20% is not actually recorded – but still seems to work Bookmark It Hide Sites News […]

  2. […] » Thoughts on Google’s 20% time “[…] the entire idea seems to function more as an attitude – that new projects should be spawned by whoever has the best ideas, not who is in what place in the hierarchy, and [Google’s] culture is based on this fundamental belief […]“ (tags: Google innovation creativity culture) […]

  3. […] Have personal projects Many would have written about it. If your employer supports your personal project then nothing like it. If not make sure to have some personal project. You can start from writing simple tools/scripting. Once you feel better try to contribute to a open source project. Later you can have your own big project. There is nothing like self learning, your personal projects will give you a chance to learn what you want. It also gives a chance to escape from **ordinary regular work**. Updated: You should read Scott Berkun’s post on 20% time. […]

Leave a Reply

* Required