In this blog post, we share our approach to the performance review process at Spartez Software. It is not a comprehensive guide, but it highlights the key aspects we pay attention to. In the chapters describing process details, there is a brief summary of how performance reviews help our people grow and give managers a good overview of how we are doing as a team. I hope you can use it as inspiration to improve the performance review process in your organisation.
In March 2020, we asked each team member for the top 3 things they like most about the company and things they would like to see improved. The unclear performance review process stood out as the #1 thing that should be improved. In particular, the team complained about:
- No clear criteria for promotion & salary raise;
- No well-defined career path and/or career development plans;
- Poorly organized process of peer feedback and its quality.
Let me take a step back here and share some important facts. We have public salary ranges at Spartez Software. This means that every employee knows a minimum and maximum salary for each role and seniority level in the company. Such an approach makes a lot of things simpler: we provide salary ranges when hiring, everyone knows how much they can earn at their current level, what to expect when they are promoted, everyone is in the same remuneration model, and there are no exceptions to remember. But it also comes with a huge responsibility - everyone’s salary has to match their actual contribution and be aligned for everyone in the company. Like it or not, this is the time when the “performance review” slogan should come to mind.
We acted fast. Three months later, we had a freshly designed performance review process that we trialled with three developers; then, we went company-wide by running the first review cycle two months later.
What stage are we at now?
Now, almost a year later, we are ready to share our findings on how performance reviews can embrace employees growth and help you align expectations company-wide. We want to share this with you to help your company and teams to develop. Why? Because:
- We have already completed forty performance reviews (note this ≠ number of people going through the process as some of us have already gone through two or even three cycles already);
- Eight people have successfully finished their probation periods passing the evaluation;
- One promotion happened (with a few more approaching in the next cycles);
- We decided to say “goodbye” to two colleagues who did not perform well, and could not improve according to the agreed plan. (New performance review framework helped us to uncover and manage these situations in an objective way.).
We are actively adjusting the process, focusing on the details, but its skeleton appears to be working well for us to share the details with you confidently.
Keep it simple
One of our company values is Simplicity. Another is Great Place to Work. As we take every value seriously, we wanted a framework that will be simple while helping the employees to grow. That’s how we made the key decisions:
- We apply the same rules for everyone in the company - developers, marketers, designers, office team, HR, …;
- Only those things that are different across roles require their own rules;
- Great Place to Work:
- Rules must be clear and non-ambiguous;
- Everything is based on specific examples and measurable goals;
- It’s a live process that constantly evolves, and everyone can suggest changes.
In the spirit of “Open company, no bullshit” value, I must mention that while the process is fully followed for engineering and support teams (together, they are the biggest part of our company), it’s not fully enrolled to other teams yet.
So, how does the new process look in practice?
- There are 5 performance review areas across the company + 1 extra area for people managers. 4 out of 5 areas are shared across the company, meaning they are exactly the same for everyone.
- Shared ones:
- Role-specific ones:
- For each of these areas, we have an evaluation matrix with specific behaviours expected at every level (Junior → Middle → Senior → Principal).
- This is definitely the most complex part of the framework: how to balance the expectations so they are realistic but challenging; achievable, but ambitious at the same time;
- We are still iterating by running performance reviews and fine-tuning the expectations. And we already defined some goals where we’d like to be:
- No more than 5 examples of desired behaviours per level;
- The expected read time of a page describing all expectations for the specific area should be under 5 minutes (we use Confluence analytics to measure this).
- Each area is evaluated as GREAT, OK or TO IMPROVE- based on the rules defined for each area separately. So the formal outcome of the performance assessment is 5 (or 6 for people managers) scores.
Let’s check how many growth vs. visibility aspects are present here:
Keep it regular, keep it balanced
A good performance review should be an ongoing process. Everyone should share feedback with everyone at any moment - we use the Small Improvements tool for that. But like any good framework, a performance review should have checkpoints, and for us, those are:
- Quarterly peer feedback:
- Every team member receives peer feedback from two selected colleagues (from inside and outside their immediate team). Different people provide feedback every quarter. It is provided to the employee (with some anonymisation if needed) to react immediately and improve before the next review cycle. Additionally, people may get voluntary feedback from other colleagues;
- Feedback is crucial to set goals and iterate on them.
- Quarterly “summary” talk:
- Manager to convey feedback;
- Review and set goals for the next period collaboratively.
- Formal performance review following the defined cadence:
- For Juniors & Mids - every 2nd quarter; for Senior+ people - annual (high expectations dictate long cycle for Senior+ people - some goals may take months before the long-lasting effects of the expected behaviours are observed);
- It’s when we go through the areas and behaviours in details, write a self and managerial assessment, discuss and iterate on goals;
- Possible outcomes are salary review, promotion or improvement plan.
Key growth/visibility considerations here:
Keep it realistic, keep it actionable
And now, the key part - execution. The greatest framework will fail if you apply it wrongly (an allusion to Story Point Concept. A Friend or Foe?). That’s why performance discussions should be based on specific examples. Our internal rules for feedback giving:
Feedback can be called a “good” one if it is:
1. Specific - references to a specific situation that took place/achievement/duty.
2. Explains an impact - why is this important? How did it impact you/team/company?
3. Time-bound - the situation described should happen during a period covered by the review cycle.
This applies to peer feedback, self-assessment and managerial assessment. We have even collected examples of good and poor feedback (from the performance reviews we did, of course!)
With examples in mind, it makes the discussion and setting the goals as objective as possible. You are discussing specific situation(s) and behaviour(s) so it’s an in-depth conversation. It leads to expectations calibration, and, as the result, - you can set realistic goals that help you score better in the next round.
First of all, goals are formulated for the behaviours scored lower. We are not prescriptive on the goals - the main idea is that they should help you to progress and improve by the following review. Goals should be discussed regularly in 1:1 meetings.
As our employees grow professionally, they must be fairly remunerated for their work. We take a proactive stance here: performance review outcomes automatically translate into salary. For myself, as a manager, it’s so great to see people being surprised how easy salary review is (well to be honest non-easy situations also happen, but they are an exception from the rule). We also plan to build an auxiliary mathematical model around salary change as we have quite granular metrics - scoring for every behaviour/area. Currently, it’s a less sophisticated model like:
- All areas scored OK - you are in the mid of salary range for your level;
- Three GREATs and no TO IMPROVE means you are on the promotion track (need to repeat that result twice) and are in the upper range of the salary band;
- TO IMPROVE is not a drama unless it repeats consistently :). They show what to focus on first, and put you in the lower range of the salary band.
Human touch is still important - we reserved a space for “extra achievements”. When someone does extraordinary work (e.g. executes an innovative 20% time project which impacts the whole company) it could positively impact their performance review or even result in faster promotion.
Key growth/control considerations here:
A visibility or growth instrument?
I am pretty sure you guessed the answer by now. It’s both:
- GROWTH is the goal, and
- VISIBILITY is the result of the goal’s measurement.
And remember that only dead processes are not iterated on!