How can we maximize the efficacy of peer-to-peer performance evaluation for team collaboration?

Recently so many papers and blogs have highlighted the importance of team collaboration. Indeed, it is very crucial topic for innovation in VUCA world. What the effective performance evaluation under the context of team collaboration is seems still a conundrum. Evaluating individual performance based on the extent to which each individual achieved his or her planned target at the end of the year has severely been challenged to enhance team collaboration.

Anecdotal evidence means many C&B experts are shocked about the rise of egalitarianism for holacracy or team spirit and are suspicious or skeptical on the validity of the paradigm change from individual performance orientation to team one. I was not exceptional as a HR professional who have developed and implemented various compensation solutions for past 10 years or so.

Working in HR Lab- developing experimental HR solutions for agility- for last months I had focused on developing a new compensation solution for agile organizations, one of whose new features is peer-to-peer contribution (performance) evaluation, and tested it with some teams. I learned important aspects about effective ways of compensation in agile teams. I will introduce them as series of my blog later on. First things first. Here is my reflection of effectiveness of peer-to-peer performance evaluation. I hope it is helpful for your cases.

Why do we need peer-to-peer evaluation?

The echoing arguments that superiors should no longer evaluate employees in order to bestow autonomy or draw on intrinsic motivation of individuals couldn’t convince me much. Many associates who I interviewed said their autonomy and intrinsic motivation could be more secured if their managers more accepted bottom-up goal setting practices. It is not necessary to shift the ownership of evaluation from their superiors to them. "Why not simply training our managers for transformational leadership rather than leaving the big burden of peer evaluations on our shoulders?"

Some self-organization proponents may argue that those responses are conditioned – the associates have learned for most of their career time that evaluations are made by superiors. So their natural desire to be independent from superiors- autonomy- is tamed into strong dependency to their superiors, which is similar to Pavlov's Dogs or Skinner’s rat. From their point of view, associates are in the conditioned status of behaviorism culture – ‘do it to satisfy bosses’. Therefore, it would be very urgent and important now to phase out manager-to-associate evaluation practice to dissolve previous cognitive conditioning and relieve associates from dependency to their managers for the sake of autonomy and intrinsic motivation of humankind.

But I want to say that the anti-behaviorism approach is not convincing enough to pull a change impulse in practice. After I had developed the first prototype of peer-to-peer contribution evaluation (actually, I used the term of ‘contribution’ instead of ‘performance’ with some purpose. The reason will be explained in coming blog.), I firstly made use of the anti-behaviorism argument to get buy-in for the solution from organizations. But it was too weak to convince them. I should have had more empirical  arguments.

Thanks to the first failure, I observed ways of working more precisely, interviewed more associates, and revisited the reasons. Eventually, I concluded a pragmatic reasoning why superiors should seriously take peer-peer reviews. It is simply because they can’t assess individual performance of subordinates more accurately than the peers of associates in a collaborating working model of empowered and flat organizations. I had deeper discussion on that with product development teams where various functional experts cohesively collaborated to develop innovative products. They confirmed that they frequently gave and took supports among themselves and they had heuristic evidences on individual contribution to their team performance more accurately than their superior did.

They need peer-to-peer evaluation to demonstrate the holistic picture of individual contribution given that the feedbacks help them put every pieces of contribution credits together. For those teams, peer-to-peer evaluation is not an enabling tool for anti-behaviorism but a pragmatically essential tool for visualization of individual contributions.

Why do we still need manager-to-associate evaluation?

Even if peer-to-peer evaluation has very positive impact to improve the quality of the performance evaluation thanks to so called ‘the wisdom of crowds’, it is hard for me to say that it can replace manager-to-associate evaluation practice completely. There are two reasons.

The first reason is that the peer-to-peer evaluation is rather bilateral observation between individuals under the shadow of benevolence of supporting relationship. It substantially neglects the opportunity costs of the individual support- while he or she is supporting someone, he or she should postpone or give up some tasks at the expense of the supporting activities. If the individual capability is high, the opportunity costs can be minimized through effective prioritization or efficient problem solving and the total utility of human resource in the team can increase accordingly. But as far as I observed, peer-to-peer evaluation has a shortfall to neglect the extent to which the individual capability affects the utilization of human resource of their team. The phrase of ‘he is good for me, at least’ prevails.

Secondly, peer-to-peer evaluation is blunt to distinguish business impact of results that each task generates. As knowledge workers, we produce decisions similar to production workers in a factory. As if the impact of production workers’ job is different among each production process- for instance, a final quality inspection may have higher impact than a part assembly, the knowledge workers’ job has different level of impact to final business results. For instance, developing A product is far more critical to improve customer satisfaction or to market share acquisition than developing B product. However, the participants of B projects show strong tendency to see their value contribution comparable to the contribution of project A participants.

What should we do to maximize efficacy of peer-to-peer performance evaluation practice?

Firstly, check which working model you want to have. If you have very personally specialized projects, you might want to maintain the order-and-delivery working model between superiors and subordinates- the superiors order tasks and the subordinates return their results back to their superiors. In that case, the peer-to-peer evaluation might not be proper. I interviewed some experts in R&D, they denounced peer-to-peer evaluation arguing that each individual has so deep expertise that the others can’t understand the job contents well. It would be exhausting for those experts to share every nitty-gritty with their peers and get their evaluations for their personal research projects. 

Secondly, you have to blend the peer-to-peer and manager-to-associate evaluation results for individual performance assessment. Even if peer-to-peer evaluations are absolutely catalyst for the team collaboration given that individual contribution to peers are visualized and recognized, it has substantial shortfalls that it is rather dominated by a bilateral benevolence and neglect the impact differentiation among the assigned tasks. Thus, peer-to-peer evaluation should take place on the evaluation framework where leaders can embed balanced evaluation criteria such as impact of the results, completion status of project, peer satisfaction etc.

Lastly, if management wants to change their working culture to cohesive collaboration, they have to challenge current working behaviors that are blocking peer-peer evaluation, not to try to delay the implementation until they think, associates are ready for that. Actually, I think the time would never come. The first step in collaboration is that associates have to understand what others are doing and be engaged in supporting their peers as soon as possible. In this case, ‘I don’t know what my peers are doing’ are not a convincing argument against the peer-to-peer performance evaluation. It is a threat to the proper change.

In coming blogs, I will share my viewpoints on a bell curve performance distribution rule and a team bonus. You will see why they are inhibitors for team collaboration.