For startup leaders: We built a career development system to retain top talent and grow our team. Here’s how.
Early on, zach holmquist and I were aware of the rare opportunity we had working at Teem. It was a small startup making a massive impact by defining an industry. We wanted Teem to be a springboard of career growth for everyone on the team. The issue was that we were a “flat” organization which didn’t have title advancement (jr, mid, senior, etc.). This caused a problem: Our engineers couldn’t demonstrate career advancement on their resumes. Teem acting as a springboard for career growth is a great idea — but we were limiting that because our team couldn’t show upward movement in their careers. It also made it hard for us to articulate where each individual should focus to improve their impact, increase their experience, and grow their career.
Goal: Create a system that helped us:
- Explicitly show individuals what they need to work on to improve their career
- Help each person articulate and demonstrate career advancement
- Allow leadership to track an individual’s progress over time (to remove recency bias from yearly reviews)
- Compare individuals to the collective group to identify mentorship opportunities and reduce team churn
I’ve said it before, and I’ll say it again. Not everyone can handle the rapid rate of failure and rebirth that startups expose you to. This project was intense — it was done on top of our daily responsibilities because we couldn’t afford enough people, but we knew the team needed it.
We kicked this project off by creating the values we wanted our team to embody. We wrote up a document that explained what they were, what they meant, and how we expected each other to put them into practice. We wanted these values to shape our culture.
Here’s the document which we made mandatory reading:
https://docs.google.com/document/d/1BbApQ5G6kEvKkPij90IC8JZhsw_NABtZTJT0IoO5IIQ/edit?usp=sharing
A career development framework is a tool you can use to measure how good you are at your job, and where you need to improve. Ideally it outlines the company and your team’s expectations for you, and then measures you against those expectations so you have explicit actions to take. To create it we borrowed a ton of learnings from other companies and sculpted it into something we thought would work for our unique team. This took a ton of research, pondering, and re-writes.
For the full 20-page framework document:
https://docs.google.com/document/d/1--jQTMBSkrIRSqTBX8KYsDmWpQdJ-WAFlbSaUtjBFQU/edit?usp=sharing
Career development is not passive
Startups will provide employees with 3x to 5x the experience in a given time range when compared with employees at large organizations. For example, 8 months ago we hired Dave Fryer as a new product manager. Since then he has shipped 3 major products and drastically improved his engineering team’s development consistency through implementing new process and growing the team. He did all this without a project manager, technical recruiter, or official mentorship. That’s the value of working at a startup.
But startups don’t help you evolve and grow your career in the way you want unless you harness the startup for that purpose. If you ignore your personal development you’ll eventually wonder why you aren’t where you think you should be. This framework was meant to help our teammates accelerate their personal career development.
Nerd matrix
First, we came up with a matrix defined by Scope and Skills. Scope is what, and how much responsibility you own. Skill is how good you are at it, which also influences how much scope you are responsible for. I named Skill after the Jedi order, because I am unabashedly a giant Star Wars fan. As proof, one of my son’s middle name is Jedi — true story.
Progression
We progress within each matrix element before transitioning into a new row or column. Progression is defined as first Learning the skills and responsibilities, then Matching them or the expectations from that level, and finally we Exceed the expectations as those skills and responsibilities become second nature to us. Once we begin to develop competencies outside of what is expected of us we progress to another matrix square. So you progress to Jedi, once you’ve already begun performing at a Jedi level. Progression is key. Time is not. The higher the level, the longer it will take to match the expectations consistently.
Peer reviews
Quality feedback can save us from complacency. Managers, and especially the executive team, won’t always recognize the impact that you’re making. You and your peers should, and that needs to be made explicit with self and peer evaluations. The intent of frequent feedback is to avoid bi annual reviews being subject to the availability heuristic, or the serial position effect. Frequent feedback helps you continuously improve, as opposed to yearly course corrections.
After writing out the framework, we put it into practice through peer reviews and a detailed report delivered to each employee.
Every 6 weeks we collected peer reviews and provided individuals with a personalized report. It’s time consuming, but this is one of the most valuable things we did for the engineering team.
We first let them know we cared
We started the report with a sincere, personalized message. Well…it’s a personal message I wrote and then automatically filled in each person’s name — so technically they all got the same message. But I was sincere, and we needed to start every review from a place of trust and caring to mitigate emotion and help our team objectively review their performance. It worked.
Skill progress
In the reviews, we tied a numerical value to each successive step (see Progression above). When you rate someone’s skill progress, you would see a number, and then the expectations of that step. The values were tied to the Fibonacci sequence to demonstrate the non linear increase in difficulty as you progressed.
Communication, reliability, and overall impact
Next, the reviews had people rate each other on their communication, reliability, and overall positive impact. It took a lot of research and iteration to understand that these three measures were the most important for our team. Over time, we found we could consolidate most measurements of performance into these three.
Anonymous comparison
The next section helped each person get a sense for how they compared to the rest of the department. This is important because people tend to let their thoughts wander to extremes. Either they think they’re better, or much worse, than everyone else. This helps correct that without invading privacy.
Comments
Peer reviews also had comments/justification for scores. We provided these comments, unfiltered, but anonymous to the individuals so they could understand sentiment and review specific advice for improvement. The comments were not self guided either — we asked reviewers to answer specific questions about the employee to maximize the value of the comment.
I firmly believe that leadership is a service role to those you lead. For that reason we used this data to identify those on the team that needed our focus and help.
If you notice, person U received the lowest peer reviews for Episode 4 (below). We made sure they knew what they could do to improve, and by the end of the next 6 week cycle, they had joined the rest of the pack and were perceived by their peers as adding value to the team. For us, this was a huge win.
People matter. This took a lot of effort but improved morale across the entire team, kept the team motivated, and helped us openly discuss personal contributions.
We did this back in Q4 of 2015