Commuting back from a day of making and testing samples in the lab on the tram, I often find myself browsing longingly through the ever sleeker and satisfyingly minimal desk-setups of software engineers on my feed: An ergonomic Herman Miller chair, a standing desk of course, an arms-width wide monitor backlit with LEDs colour graded for optimal productivity and cosiness. It is a long way from the incessant banging of a peristaltic pump, smell of wool, and messy mixing pots that decorate our lab-bench.
Welcome to software envy.
It is a feeling that runs rife in the advanced materials start-up ecosystem. Yet another calendar app raises millions of dollars in funding, an AI start-up claims a market size that's multiple times the size of the British economy, the facility of scaling a new product to thousands of users with minimal capex on your friendly neighbourhood cloud-provider. Working in hardware can feel particularly hard. Why make my own product, when I could sell SAAS to optimise everybody else’s product development.
However, I do like to channel this blatant jealousy into a force for good. I follow a collection of random software engineers, keep up with tech-news, and read the biographies and teachings of silicon valley veterans because the software sector grows and iterates so fast and effectively, there is nothing else like it. Advanced materials might lay the foundation, but software builds atop. There is so much to learn from the last few decades in software development, that applying this to materials R&D has been the guiding compass in my career so far.
At Vector, we are not only looking to innovate in the products we develop, but also how we carry out that process. This experimental approach permeates all aspects of our company, but in this post I’ll focus on our most recent implementation inspired by Silicon Valley; task management!
Start-ups are typically time rich, cash poor. So the key to scaling successfully is maximising the impact of time spent. This leads to the workaholic founder trope, wherein any successful entrepreneur must spend every waking minute smashing out tasks. This rhetoric pays the bills of loud alpha male productivity gurus, but in practice it is unsustainable and scales terribly without near-infinite streams of money and labour. In the real-world, scaling needs to happen sustainably and thoughtfully. The hard part is not writing a long to-do list, it’s picking the few right tasks to assign your energy to, in a conservative cadence which keeps that energy available week after week. Rome was not built in a day.
Our practical implementation of this philosophy draws from the sprint practice, popularised in ‘Agile’ software-development. I won’t delve into the project management theory that accompanies this practice, but rather detail our approach and why it works for us.
Sprints define our cadence of work, they last for 2 weeks exactly.
1. At the start of a sprint we consult our backlog of tasks and the calendar for upcoming deadlines/deliverables/milestones, the team then populates the sprint board with relevant tasks. Administrative tasks are included as well as technical work. Granular detail is not required but we aim to be thorough.
2. We estimate how much resource each task requires. We use hours and reach a rough number collaboratively. Estimates are averaged across all team members with no weighting for seniority. We will self-report actual resource taken to assess the accuracy of these estimates at the sprint end to get better at the practice. It is generally better to have overestimated than vice versa.
3. The next step is key; prioritising. We assign one of three variables to every task “Must Complete” when it is a business critical task. “Should Complete” when it is an important task to do but not time constrained (these often become Must Completes in the following sprint). “Nice to Have” when the tasks would bring joy, but only if we can spare the time.
At this point, you’re thinking, ok great but what if you have more “Must Complete” tasks than resources in the team? This is where it hurts, but you have to downgrade, shift or remove tasks until you can afford the effort. This might feel frustrating, like you’re moving slower, but it is a healthier rhythm: A steady cadence compounds gains through improved work-life balance, employee satisfaction and ultimately higher performance at work. Acceleration is possible, but without increasing the amount of resource (eg by hiring), it requires optimising task choice for impact and implementation time, which has its limits.
4. Finally tasks are delegated. An ideal breakdown would give every team member a healthy mix of task priorities. It is unhelpful to load individuals with all the business critical “Must Completes”. People should feel positive about the sprint ahead, without being overburdened.
5. We check-in on tasks mid-Sprint. Inevitably in start-ups unforeseen distractions occur, but we keep 20% of actual resource capacity free to dampen these. Once a sprint has ended, we will have an open discussion about what we did, how it went, and look to apply any learnings to the next one.
Despite only implementing this practice in the last few months we have noticed clear benefits. Team members are better able to manage their time independently. A fixed set of tasks helps focus the mind and optimise delivery. The opportunity for reflection at a sprint-end is a helpful and mindful practice. Transfering learnings to the next sprint provides regular opportunities for redemption.
No task management system is perfect, and we are iterating on the practice as we go. Start-up life can often feel chaotic, but organised chaos sparks joy!