Over the last couple years, Vistaprint has transformed its career development system. We’ve moved from a very manager-oriented system to one where employees are empowered and gathering feedback from each other. It’s been a long journey already, and it’s far from over.
What follows isn’t a canned ceremony you can copy/paste into your organization. It provides a view into some of our company culture, the current state of this system, and how it’s evolved. It will tell you a little bit about one facet of working here. And hopefully you’ll pull out some insights that you can apply yourselves.
Understanding Vistaprint’s culture
Before we can describe Vistaprint’s peer review program, it’s important to understand the cultural context of the company in which it exists.
Vistaprint began to embrace the agile mindset several years ago. We’ve found it helps us get our best ideas to our customers as fast as possible. It’s a journey we’re still on today and has driven us to a culture of continuous learning and adaptation.
We’ve also started to think differently about our workplace environment. We’re encouraging people to be more authentic – to show up at work with their full selves, to understand each other’s perspectives, and to provide candid and kind feedback to each other. We’re seeking greater degrees of collaboration. We’re reimagining the role of leadership in the organization. Instead of a deep hierarchy built around command and control, our leaders are promoting a flatter organizational structure, pushing down decision making and rallying diverse teams around clear goals, supporting them as they bring their best ideas to customers (sound familiar?). This, too, is a path we’re still walking today.
Traditional systems for career growth and promotions relied on the role of an all-seeing, all-powerful manager. Your manager controlled not just what you were doing, but potentially how you were doing it. They developed a perspective on your performance (the only perspective that mattered). They defined a growth path for you. They decided when you were ready to be promoted, in accordance with a detailed list of criteria.
Such systems are clearly at odds with the culture we’re creating. We don’t want our leaders to control how people work – we want them to define goals and let employees find the best ways to achieve them. The perspectives that matter most are those of the people you’re working with every day. And your career growth is entirely unique to you – you’re the person best positioned to drive it.
We had to find something that met our needs and supported our culture.
The Peer Review Process
In that context, we developed a new way of thinking about promotions: the peer review program. As the name implies, it’s all about you and your peers. The following sections will elaborate on the different steps in the process. But at a high level, it goes something like this:
As an engineer seeking promotion, you begin by gathering feedback from your peers and reflecting on it. Those results are synthesized down into a self-assessment, which is then given to a panel made up of your peers. That panel will learn more about you, think about potential growth opportunities, and recommend whether a promotion is warranted.
Giving and receiving feedback directly (not through a manager), with the recipient’s best interest in mind (and not vindication), is a core expectation of all our employees. It should come as no surprise that a peer review program begins with getting feedback from one’s peers.
Ideally, you’re getting (and giving) a lot of that feedback day-to-day. For example, you’re recognizing at standup when someone has done a great job of helping you with a problem you were stuck on. But it’s also good to step back and to ask people not just to reflect on today or this week, but over the last several months. What can they always count on you to do, day in and day out?
Many of our employees have found that structured group feedback sessions are a great way to answer those questions. In many ways, they’re not unlike a regular team retro. Everyone is expected to reflect candidly on what went well and what could be improved, with a mindset not of blame but of recognition and continuous improvement. Instead of being about a team, the focus is on you.
One format we employ is sailboat feedback (based on the sailboat retrospective format). We get a group together to provide feedback to Alice (it could be a small dev team of 4 or 5 people, or an extended project team of 15+ people from across the organization). Those people are asked two basic questions. The first is “what is the wind in Alice’s sails?” What propels her forward and helps her get great work done? The second is “what are the rocks around her?” What slows her down or represents a risk she needs to watch out for? After some reflection time, people take turns going around the room and sharing their perspectives. Themes quickly emerge. And in the span of an hour or less, Alice has a great idea of her core strengths and areas for improvement.
Does it sound uncomfortable to be in Alice’s shoes here? It can be at first – but never for more than a few minutes. The adage “feedback is a gift” has never seemed truer than during these sessions. It’s rewarding to hear that things you do are appreciated. It’s cathartic to hear that something you were worried about wasn’t a big deal and life has moved on. And it’s fascinating to hear other people’s opinions about what they find valuable in you. People almost always leave these sessions feeling something ranging from pleased to excited (and I’ve never once heard an expression of regret).
These group feedback sessions aren’t the only channel available. The peer review program expects that people have solicited feedback, but does not mandate a specific vehicle for doing so. If you’re more comfortable in a 1×1 format, there are sample questions you can use to help start the conversation. We want you to leverage your strengths – if you communicate better in writing, then start there.
Regardless of how it was collected, peer review begins with this step of soliciting feedback from the full range of people you work with regularly.
Reflecting on yourself
Along with input from your peers, your own notions about yourself are another important part of the peer review process. Self-reflection can be both an input to feedback solicitation, as well as an output from it. If you have an idea of what your own strengths and weaknesses are, getting feedback from others is a great way to validate those ideas. Do they see you the same way? If not, why? And after gathering feedback from others, you will need to look at the aggregate picture that’s emerging. What messages are you consistently hearing about strengths and opportunities?
The peer review process requires pulling all this information together into a document. You’ll be summarizing the strengths and opportunities you found in collaboration with your peers. You’ll also be asked to provide a little bit of context on your work, such as what you’ve worked on and how you’ve contributed (e.g. coding, facilitating design sessions, etc.). The last thing you’re asked to do is compare yourself to our level definitions. Looking at everything you’ve found, do you think you’re ready for promotion? The ability to understand and reflect back what’s expected of you at the next level is an important part of demonstrating the maturity necessary for a promotion.
After submitting this information, you’ve now officially entered the process and are considered a candidate.
As we’ve evolved the way we work, our level definitions have had to follow suit. Previous iterations of our level definitions were still rooted in a command and control structure. Those aspects had to be changed.
While we were revisiting our level definitions, we had a major revelation about how they were framed. In the past, we had always struggled with legislating the entrance criteria for each level. It was an all-encompassing rubric that made it hard to accommodate people who wore many hats. It left little room for people to grow into a role. It was also extremely hard to change.
With the rollout of the peer review program, we’ve stopped thinking of levels in those terms. Instead of framing them as a minimum set of skills, we framed them as a median set of skills, paired with a strong focus on the size of the impact. We’ve also identified areas of breadth (skills most people will have) as distinct from areas of depth (skills that specialists will have).
It was much easier to align on what the levels really meant because we were talking about typical cases. We knew the flexibility of the process would give us a way to handle edge cases, without needing to codify every possible career path in a level definition. That let us fit a lot of different shapes and specialties into relatively simple level definitions.
Being assessed by your peers
After you finish your self-assessment, a panel will be formed consisting of three engineers from the next level. For example, if you’re a level 3 engineer who wants to be promoted to 4, your panel will consist of three people at level 4.
The panel has a relatively simple question to answer: do you believe this person will be successful at the next level? Why or why not?
What panels ultimately produce is a written recommendation. The recommendation includes a clear statement (“yes” or “not yet”) and an explanation of why. The panel reflects back the most important behaviors they found in the self-assessment. This helps the candidate ensure that what they said was properly understood, and also gives them an additional perspective on their own skills.
In keeping with how the rest of the organization operates, panels are told what to accomplish (clearly answering whether the candidate is ready for promotion) but have a great deal of freedom in how they accomplish it. If you’re on a panel and you need more information from the candidate, you’re encouraged to simply ask them. If you need another perspective on a story the candidate was relaying, then you should seek that person out and find it. Do you wonder why the candidate made a certain design decision? What better way is there to find out than by standing at a whiteboard and talking through it together? By giving the panels this much freedom, we can make the process highly personalized, for both the candidate and the panel members.
There’s a tremendous amount of responsibility resting on the panel. As such, composing the right panel for any given candidate is very important.
We always try to have one person on the panel with good line of sight into the candidate. We want someone who understands their technology stacks, the daily work environment of their team, etc. It’s important to have someone who intrinsically shares that context the candidate has been working in. It helps ensure that candidates feel understood.
We also try to have one person on the panel without such line of sight. We want someone who can bring in an external perspective and help relate what they’re seeing and hearing to what’s going on elsewhere in the company or in the industry. That’s important to help us stay calibrated across the company, and not develop pockets where it’s very easy or very hard to get promoted. It helps ensure that candidates feel like the panel is unbiased.
And we always try to ensure that panel members are volunteers. It’s an intensely personal process for the candidate. Panel members can sometimes spend several hours assessing a candidate. It’s critical that we have people who want to be there and provide a service to their coworkers. So far, we haven’t had trouble finding people who are willing (and even excited) to participate.
A recommendation for growth
In the spirit of continuous improvement, peer reviews also have a strong focus on growth.
By this point in the process, we have a candidate who has spent several hours reflecting on their core strengths and opportunities. We also have a group of three people (already at the next level) who have become intimately familiar with that candidate. This is an excellent opportunity for coaching the candidate ahead, and it’s one we try to capitalize on.
As part of the recommendation they produce, panels are asked to think of step function growth opportunities for the candidate. How could they apply their existing skills in new ways? Could changing their focus slightly dramatically increase their impact? What could a bigger, better version of this role look like in a few years?
This is one of the key aspects of the panel’s recommendation: it’s intended to be forward-looking. It isn’t simply about validating work that has already been done. It’s about identifying the skills and behaviors it took to accomplish that work and looking ahead at bigger opportunities and complementary skills. It makes the recommendation an invaluable tool in the hands of a motivated individual. Compelling career growth stories have emerged from these recommendations.
This forward-looking focus is also an important part of making candidates feel supported. Regardless of the specific decision a panel makes regarding a promotion, they’re making an investment in the candidate’s career.
Closing out the process
The final stages of the peer review process were crafted very carefully.
Once a panel makes their recommendation, they must deliver it to the candidate. After sharing it digitally, the panel meets with the candidate to discuss it. This gives the candidate an opportunity to ask questions of the panel and ensures everyone arrives at a mutual understanding. It’s important that the candidate and panel are direct and honest with each other, because that’s what we expect of all our employees. This process also happens before a manager is notified. If a candidate feels something critically important was missed or misunderstood, this gives everyone an opportunity to regroup.
Only at the very conclusion of the process is the panel’s written recommendation shared with the manager.
Checks and balances
Panels are given many degrees of freedom in how they assess candidates. They are expected to allow room for growth and account for a variety of different ways in which people can add value. This is a very powerful tool, but it requires some checks and balances to maintain its integrity.
One important check is for consistency. The recommendation of each panel is quickly reviewed by a central group. This lets us ensure that the recommendations being made are consistent across the company (i.e. we don’t have panels that set unreasonably high or low expectations). This also lets us ensure that candidates are being treated well (candidly and kindly).
Another important check is built into the structure of the panels. By having groups of people, instead of a single person, we ensure that no single person’s thought process dominates the decision. By balancing the composition of that panel with people who are both near to and far from the candidate, we also help prevent localized extremes from emerging (where there are easier or harder parts of the company to get promoted).
One final important check and balance is in the decision making structure itself. Our panels make a recommendation regarding promotion. If all other checks and balances have failed, that recommendation can be overridden by a manager.
Some of these checks and balances have never been invoked. A manager hasn’t yet overridden a panel’s decision. A panel hasn’t yet met with a candidate and subsequently reversed their recommendation. However, we’ve found a robust system of checks and balances is an important part of creating a system our employees can trust.
In part 2 we discuss how we incrementally built this program to make sure it meets the needs of those that use it.