A while back I had a conversation with a good friend, Santi Trench (VP of Sales in NetRivals) regarding motivation, high-performance teams, and KPIs for measuring them. He was leading a webinar for discussing these topics and asked me to join and explain our approach here at TravelPerk.
I don’t believe there’s one unique approach for driving motivation since it’s quite dependent on every particular context: the mission of the project, the industry, the team itself, their background. But in general, I try to evaluate all the possibilities from a theoretical perspective using gamification techniques. These techniques are based on the self-determination theory and how to focus on intrinsic motivators.
So, how do we at TravelPerk motivate our teams? How do we manage our engineering team and drive each individual’s performance?
What are the factors for motivating a team?
Each team is context-dependent, so I usually like to focus on the theory for evaluating the different actions you need to do according to your particular context.
To explain the reasoning behind the different actions we do in Engineering at TravelPerk, I will compare them in this post against Paloma Medina’s BICEPS concept.
B – Belonging / Relatedness
One essential aspect of high-performance teams is trust. So any action you can do to build trust and accelerate it will be crucial for success.
In our case, we have very strong core values at both the company and the engineering levels, and we align these core values with our brand-building activities and with our hiring process. Hiring the right people is key for adopting and cultivating company culture and generating a feeling of belonging between team members.
More technically, the configuration of the different teams also matters a great deal. We’re currently working with ”squads”, following the Spotify model. We have around 20 squads, and on principle, we never create a new squad with only new people. We always allocate people with a previous track record in the company as foundational members, to help reduce the forming and storming phases as much as we can.
Team size also plays a fundamental role. In our case, we operate on Dunbar’s number of between 7 and 9 multidisciplinary members. We only exceed this number if we’re considering splitting the team soon. In this way, we’re trying to transform one “performing” team into two “norming” teams in a relatively short time.
We’ve seen a dramatic impact on this because of COVID-19. Previously, our company was optimized for face-to-face interaction. That’s why both during the lockdown and afterward, it’s been a lot more challenging to find creative and safe ways to keep the team atmosphere and connection alive.
Finally, the company and team missions are other important aspects for building relatedness, since they are directly connected with team members’ intrinsic motivation.
ICE – Improvement/Progress + Choice + Equality/ Fairness
To progress, you need feedback—the sooner the better. At TravelPerk, we’re a continuous deployment company, so by design, we can provide automatic feedback on any person’s engineering contribution almost immediately.
There is probably no point in explaining our code lifecycle approach in detail here, but basically, the minimal unit of work is a PR (Pull Request). We have several automatic status checks for the team (coverage, testing, style, static code revision…), so whoever violates one of these statuses is not able to continue, and needs to change this approach to be able to proceed with the contribution.
After that automatic check, we also have a manual revision from other colleagues, which again provides direct and quick feedback about the engineering approach.
During this SDLC, we’re able to gather visibility about the behavior of the team. We’re able to collect metrics at the squad and individual level, and we provide those metrics to team members to empower their self-awareness.
We are also a continuous feedback organization, we have weekly/biweekly 1:1s with all team members, to gather insights about how the team members feel, their ambitions, their blockers, and we work together on their development plan.
We dedicated a lot of time and effort to support the engineering team’s career development in 2020. We decided to adopt Medium’s snowflake model, based on levels of competency in certain areas. We don’t expect each one of our team members to be an expert in each area of competence. Rather, we empower them to develop only in the areas they are inspired by.
Progress at TravelPerk isn’t about writing perfect code. It’s about being pragmatic, embracing and understanding change, and ensuring everyone is respected and can be successful by being themselves. We’re happy about the team adoption so far, about the impact we’ve generated, although we acknowledge we are still crawling—we are still learning to walk.
P – Predictability
As a company, we work with OKRs, so every quarter the whole company goes through a self-reflection process of where we are and where we want to be. It starts with the leadership team, and then the departments, and then each particular team commits to the changes they want to generate to accomplish the overall goals defined by the leadership team. That gives everyone in the organization an understanding of what we want to accomplish in the following quarter, and why.
We also provide predictability regarding career development. On the one hand, we are a continuous feedback organization. But a side effect of this approach is that people miss the predictability of when they will reach the point of getting a promotion or a raise. As a result, we’ve now adopted a bi-yearly performance review process. Together with our engineering grading, it gives a clear framework for setting up expectations for everyone in the team about what they should expect, and what we expect in return.
This has, of course, been made more challenging by COVID-19. There is only one recipe for building trust with your team, which is transparency and communication. And in truth, that’s easy when you stick to your core values as a company. We are really proud of our leadership team for how they lead us during this global pandemic.
S – Significance
This aspect is not part of the self-determination theory, and it’s not part of the intrinsic motivation. In my opinion, we need to treat this very carefully. The engineering community is quite particular, and not everyone feels or reacts the same way when it comes to feedback and public praise.
It’s each manager’s responsibility to identify, especially during the first 1:1s, each person’s preferences regarding feedback and praise—and to respect it.
Obviously, everyone has a clear career path, and every role comes with status. However, we let each member manage it the way that makes them feel most comfortable.
What are the main KPIs for driving performance?
At TravelPerk, we designed a director plan for scaling up the engineering team. We wanted to almost triple it—from around 40 engineers at the end of 2019 to more than 100 by the end of 2020.
For me personally, it was essential to have metrics and monitoring in place (both qualitative and quantitative) that could bring visibility that we were not degrading the quality of our impact.
So, we decided to adopt a customer lifecycle framework and use it as an employee lifecycle framework. We defined KPIs in every area of the journey of an engineer. We decided to adopt the PIRATES metrics and we defined the following:
- Brand-building (Awareness): Activities regarding how our engineering company is addressing the talent.
- Acquisition: How we get engineers in the funnel, how our hiring process performs.
- Activation: Once an engineer is part of our organization, how do we speed up their contribution (onboarding)?
- Revenue: For us, the revenue of the PIRATES metrics is correlated to the engineering contribution. Probably the most complex area, so I will develop it below.
- Retention: Do they stay? ****Monitoring the retention of the engineering team.
- Growth: Do they progress? Some of our KPIs are NPS over the training in our e-learning platform, ratios of promotions, and raises.
Measuring the contribution of an engineering team is a very controversial topic. We should measure our impact based on the customer impact. While we have a lot of influence, we don’t have the final call on what to build, and still, we need to see that we are building in a high-performance way. So we decided to define metrics that are meaningful for us, and actionable on a daily basis for members, technical leaders, and managers.
In fact, we wrote a whole article about this: Engineering analytics: Understanding how our teams work to make proactive decisions
This structure not only gave us the perfect foundation for monitoring the team performance but also provided tools for taking direct actions daily—both as a team and as a member.
And, obviously, it was also vital for monitoring the growth of the team and how the COVID-19 situation affected our outcome.