10% lab days
I’ve had the opportunity to work at a company which implemented “10% lab days” and over the course of 3 years seen how it evolved and worked in practice.
I don’t mention any names because it’s not a sponsored post and I don’t represent the company. Nevertheless, the company is one of the largest media houses in Scandinavia behind some of the top 10 visited sites in Sweden and Norway.
First, let’s see where it came from. Google famously practiced the “20% time”. The idea is pretty simple but the impact is huge:
“We encourage our employees, in addition to their regular projects, to spend 20 percent of their time working on what they think will most benefit Google. […] This empowers them to be more creative and innovative. Many of our significant advances have happened in this manner.” — Larry Page 2004
Notice the “they think will most benefit Google” part. We’ll dissect this in a bit.
Now before we jump into cargo culting (blindly copying someone else’s best practice hoping for the same results), let’s step back and see what environment is necessary for this to work.
The underlying idea is to spread the R&D responsibility to the entire organization instead of limiting it to a dedicated department. The hypothesis is:
- Great ideas may come from anywhere
- The most practical ideas come from the people who do the regular work (as opposed to an isolated R&D department)
Not every company has the culture to take new ideas seriously and evaluate them. Even fewer companies have the tools to capitalize the great ideas.
There are 3 important aspects that are required for this to work:
If the company does not have a clear and visible vision, it will be hard to give direction to new ideas even if the employees get 20% extra time. It’s a bit like shooting from the hip. Some ideas stick, some are a total waste of time. It is enough to have the vision, but it should be visible and part of the evaluation framework for ideas so that the employees know what is expected of them.
“Leadership is the capacity to translate vision into reality “ — Warreng G. Bennis
If the employees are seen as code monkeys trapped between the sprint cycles and overloaded by concrete tasks to squeeze the maximum amount of immediate results, they’ll not be seen as a source of ideas. This is especially true for top-down management style where people have titles like “head of …”. If one person is the “head”, the rest of their org is the “body” which is supposed to execute their ideas.
“It doesn’t make sense to hire smart people and then tell them what to to , We hire smart people so they can tell us what to do. “ — Steve Jobs
For lab days to deliver on their promise, the employees need to be given the mandate to choose their own agenda. If there’s not a full trust that they will spend the time efficiently, you can provide them with some loose framework and guideline about what this time is supposed to be used for. See What kind of work? section for examples.
Not everyone is a self-learner. Some may choose to do regular work and that’s OK. It’s important to put the employees on the driving seat with a map for the company destination. Let them drive.
If the company leadership is primarily focused on short term results (tactical) than long term objectives (strategy), they will not see the benefit of “waisting” 20% of the payroll for no immediate result. This is of course a very narrow vision because investing in R&D can yield huge long term results.
Sometimes to move fast you need to go slow.
If you work in an industry which cannot benefit from investing on R&D or finding new solutions to old problems, there’s no need to invest in innovation. For anything else, it’s hard to justify not to.
Before jumping into the implementation, let’s highlight some of the common counter-arguments and examples for how not to implement it.
Putting employees in control of 20% of their paid time without any structure will not fly by management, nor does it guarantee to yield the expected results which are worth literally 20% extra pay.
In fact, it may hurt the performance of the other 80% if not executed properly. For example:
- Someone may start a project that, unlike their regular duties they are passionate about and it can easily swallow more than 20% of their time and/or their most productive hours.
- Company trade secrets and intellectual property may unwantedly leak to an open source project that comes out of an initiative
- An employee may quit the company to start their own business based on an idea they started on the time that was paid by the company.
These are not hypothetical examples. I’ve seen all three in action but as I said, our 10% strategy evolved over time to provide a framework for innovation.
Alright let’s get to how this idea was implemented at one of the largest media companies in Scandinavia.
At the basic level, 10% of the employees’ time is spent on doing things outside the regular duties. Why 10%? We thought Google’s 20% is rather too aggressive and our revenue and market didn’t justify a higher percentage. However, the ambition was to spend this time smartly and productively using a framework. After all, we did not have Google’s revenue and in our local market hardly any other company made this level of investment. So it was a good first step and there hasn’t been any serious complaint to increase it. 10% provides a good balance between regular work and innovation for us but other companies may want higher or lower. One thing is for sure: start small and evolve it over time.
What kind of work?
People use this time for different things:
- Some took online courses
- Some fixed tech debt and improved code quality (may sound boring, but at least this was a regular change to improve the code)
- Some including myself did open source projects (in a later post I’ll discuss how to set an open source strategy but if you’re in a hurry, here’s some great material)
- Some created a new solution to an old problem and demonstrated their proof of concept.
- Some managers read a book that would improve their leadership skills or understanding of the market
- Some tried learning new solution technologies to see how it solves the problems at hand
- Some spent time on improving thedocumentation
- Some built automated tools to reduce toil
- Some used this time to add more automated tests to the code base to cover edge cases
The key here is that this time is not controlled by product managers and the employees decide how to spend this time. Some even chose to do regular work and that’s fine. It is opt-in. Not everyone feels like innovating all the time. In fact after the initial excitement curve flattened, many found themselves doing regular work but a bit more relaxed.
“This is a great idea, but maybe something for 10% lab days?” was a common theme in conversations where the nature of a solution was ambiguous and needed some work.
When to do it?
It would be stressful to pull out 10% of your time for R&D when the rest of the org is doing regular work. You may be someone else’s dependency and either starve their work or get pulled into their regular duties. To reduce this kind of pressure, it was decided that we all do “10% lab days” simultaneously on every other Friday. Why Friday? Because:
- If your idea needs more work, you can continue working on it over the weekend if you want. A big warning here: this was not an expectation but rather a flexibility to allow for continuity.
- Many of us who had on-call duty were free Friday the week after. Although we would miss one lab day, at least we knew that we’re not someone else’s dependency for regular work.
- Fridays are usually shorter and more cheerful, so why not add to the fun and let people do something they desire to do?
Friday every other week was marked in a calendar invite that was sent to everyone with a description about what’s this about (new employees appreciated that) with a link to read more on an evolving document with further information.
Who did it?
The 10% lab days applies to the entire org which this initiative came from. The company had several satellite companies which gradually adopted this initiative as they heard about the results. Good ideas may come from anywhere, therefore there is no point in limiting it only to a subset of the skills at the company’s disposal.
The company vision is clearly defined at the company level and present at many meetings as well as on top of the landing page. Note that I didn’t mention the mission.
While vision focuses on tomorrow, the mission statement focuses on today.
When defining a broad innovation framework, it is good to have some direction but not be too strict. You want to leave some space for experimentation but no gap for misinterpretation.
Moreover, the company had a slack channel (
#lab-day) where the ideas were discussed and sometimes people teamed up together to work on a new PoC (proof of concept). It was strongly recommended that people discuss their ideas to give them visibility. Some teams skipped the standup on that day but my team wanted to hear about everyone’s idea for the lab day for an opportunity to pair up.
Based on our experience running lab days for a few years, there was some areas that could improve:
Some of the work that came out of the lab days were good enough to expose publicly. We have a public Github repository where some of our best work is shared. Some engineers may choose to publish their work under their own name for publicity. We had a clear open source strategy to define which work goes where.
Another aspect is that sometimes an employee (or a few) got too attached to an idea that they wanted to productionalize it. The company was well prepared for that. There were people at the innovation hub who would help evaluate the idea and if needed dedicate resources to grow it to its own spin off. However, it has happened that an employee quit their job to build a business on the idea they started at lab days. In those cases, as long as they didn’t breach the contract (taking company intellectual property or trade secrets or other employees), it was fine. After all, it’s hard to know if they wouldn’t do the same if they were hustling in their own time. It’s good to part ways on good terms.
Some missed it
We also mentioned that the on-call compensation day fell on a Friday and therefore one would miss the lab day. Obviously, employees who didn’t work on lab days missed it and it was more common than one may thing (for example some employees worked 80% to use the generous parental leave available in the nordic countries).
Balance: micromanage vs anarchy
Finding a balance between setting expectations and leaving enough room for innovation is eaiser said than done. Too loose and you’ll create slack, too tight and you’ll create confusion about how lab day is different than regular work days. In the absence of a clear strategy and policy, the initiative can be interpreted differently by different people.
What could help our setup was a clear policy that anchors the expectations and a clear strategy that anchors the motivation.
The strategy is a plan of action while the policy is a set of common rules and regulations, which forms as a base to take the day to day decisions.
The net cost is less than 10%
Since the lab days are opt-in, it’s very normal for the company to spend less than 10% of its personnel doing actual lab activities. It’s all too common for people to see the lab day as an extra day to finish off their regular work. What’s important is that the manager don’t pressure the employees to loose their lab days doing regular work.
Moreover if the lab day is on Monday or Friday, there’s a good chance that the employees may spend some of their free time on their idea as well. To be clear it should not be an expectation, but you don’t want to totally block the possibility either. Many top talens have hobby projects as a learning tool and they work on it in their free time. If they happen to want to work on it during a lab day, they shouldn’t be blocked.
It’s a perk
Full disclosure I left that company back in May 2021 amidst the “great resignation”. Nevertheless, that is the company I stayed there the longest. Lab days were one of the main reasons for that. Admittedly my salary raised more than 10% when I left.
We discussed the motivation pros and cons of lab days as well as its implementation at one company. For lab days to deliver its promise, there are 2 important requirements:
- Employees who are self-starters and take initiative
- Clear policies that helps them spend their time on what is impactful
Hackathon is another implementation of the same idea within a more limited time. It is often a good first step to test the potential of the organization to innovate.
If you want to promote 10% days at your company, feel free to share this article internally, or publicly.