Employee turnover in USA was at its highest—400%— in the first half of the 20th century: This is before the Great Depression, before the 8-hour workday, before HSE. We surmise that had performance appraisals been as much a part of an employee’s career as they are today, that turnover would have been even higher. There’s a reason performance management –and each framework that governs it—draws such bad press. In fact, there are several. And most of them seem to start with assumptions about the relationship between people and productivity. The advantage of data-backed performance framework is that it calls out many of these erroneous assumptions, and takes a wider, more transparent look into the nature of work.
Assumption #1: Productivity And Performance Are The Same: It is desirable to think that productivity and performance are the same. Unfortunately, they’re not. Productivity is linked to output; performance is linked to results. Ideally, output should be the desired result. In lean organizations, it is. But in organizations where hierarchy pervades, productivity is just one of the raw materials that feeds into results. Market reputation, competitor activity, the organization’s ability to influence the course of regulation, all have a bearing on results.
In such cases, it is easy (and likely) that individual productivity gets lost in the massive assembly line of top-down processes and the chain of command. So someone who is closer to final results will have his/her performance evaluated more positively than someone who’s productivity directly enabled the results in the first place.
Take the example of regulation. For the latter half of the 21st century, multinational organizations struggled to penetrate protectionist markets. That didn’t mean there was a lack of effort or preparation in their home offices, or that the team leading feasibility studies had fallen short. But the person who used personal influence—either by way of industry networks, or more questionably, bribes to officials in such markets—was often the one most rewarded. Performance management frameworks that focus overtly on what was achieved, and not enough on how and what cost will always be skewed negatively towards workers whose efforts cannot be directly connected with profitability. And employees who focus on being productive, will feel undervalued until they leave. Because that is when the gap between output and results will be highlighted. Organizations that base their performance management on the foundations of frameworks like MBO, without considering its pros and cons might be susceptible to disproportionate award distribution. A data-driven performance management tool will call out these gaps before the company has to endure the attrition of a valuable resource.
Assumption #2: Competition Is Good For Performance: Studies fare favorably when it comes to the relationship between performance and competition. Competition acts as an antidote to monopolistic and hegemonistic behavior—both in the organization, and in the industry. It keeps individuals competitive and responsive, and it’s an effective guard against complacency.
But there’s a catch. Decisionmakers assume that competition is in the interests of the company. That employees will vie to outdo each other, to prove oneself better than the other. Such decisionmakers believe the trajectory is upward. It isn’t. A highly competitive culture creates more problems than it solves: Interest groups and lobbying, infighting, politicking, abusing resources to drag counterparts down instead of pulling them up. Often, these negative workplace dynamics occur in the same team. All in all, the ingredients of a toxic workplace.
Does this mean performance frameworks should not make employees compete? They should. To address the aforesaid issues, the reward structure needs to be distributed. Instead of the traditional ‘winner takes all’ approach, awards need to be categorized. Mohan Thite’s book addresses this via “Healthy Hierarchies”, in which he proposes the RACI approach: Reward high performers with things they actually value. Not everyone wants a big team to lead. Not everyone wants a corner office. By aligning rewards with the desired growth of the individual performer, it is easier to maintain morale, without compromising either output or the competitiveness that spurred the high performance. With a data-driven structure, this becomes easier: Pattern detection in individual and team milestones identifies what individuals are motivated (and demotivated) by. Data can be helpful in flagging the ‘seasonal workers’—those employees who slouch all year, but whose productivity shoots up dramatically around appraisal period. Likewise, it protects employees whose performance reflects burnout at the end of the year.
Assumption#3: Growth Cascades: Since the past twenty years at least, most appraisal programs are strictly top-down. For example:
A department has an overall annual sales target of 5X%.
Which means its five sales leads have individual sales target of X% each.
Assuming each sales lead has a team of 10 people, each leader’s team will be responsible for (X/10) % of that sales target.
While this sounds perfectly logical, in practice, it is messy. Because growth doesn’t always cascade. If there isn’t enough distance between two hierarchical roles, targets cannibalize instead of cascading. If goals are too similar, or the reporting authority cannot add more value to the output, there will be competition between the manager and his/her report. Creating lean structures isn’t always possible, nor does it always solve the problem. But what performance management tools need to do here is check the tendency for internal (unhealthy) competition. It is not realistic to revise sales targets based on this issue. However, to ensure that the manager doesn’t compete, the performance management tool can reward him/her for qualitative leadership skills—like mentoring and problem-solving. Chances are that the cannibalistic behavior originates from a place of insecurity. By rewarding behavior that encourages collaboration, insecurities are appeased, and collective team performance—and the targets—go in the same direction.
Assumption#4: Benefits Trump Culture: Does a well-paid workforce translate into a better workforce? Do companies that pay their employees above market rates, and/or provide attractive benefits always succeed in attracting and retaining best talent? Theoretically, yes. The belief is that when a business pays above-the-market rates, it also gets above-the-market performance. Or in other words, employees will be required to justify their generous compensation with high quality work. (“If you’re being paid that well, you must be good.”)
But this approach can be a double-edged sword. A generous reward system will appeal more strongly to people with extrinsic motivation. These people may or may not be exceptional performers. But if there are too many of them in one place, they will crowd out high performers with intrinsic motivation. The result—performance becomes a hostage to extrinsic rewards. Leaders in such organizations become hostage to a few employees, who are not adding as much value as they should, but who’ve created such a hegemony that it would be even riskier to do without them. It’s not uncommon to find lobbying behavior dominate in such organizations, following the creation of visible “in” and “out” groups.
If an organization has this kind of culture, tools like the bell curve provide a fatal setback to fair and honest evaluation. This is because the bell curve philosophy takes for granted, that:
60% of the workforce consists of ‘average’ performers, deserving a safe “3” rating.
20% are below average, or unsatisfactory performers, deserving a rating of “4” or “5”
20% are above average and exceptional, deserving a rewarding rating of “2” or “1”
The numbers may follow a range outside the given 60/20%, but the practice is the same: For the majority to be graded a safe average, someone must become an underperformer. Likewise, even if an entire department has performed very well, only a chosen few will be given the coveted ‘1’ rating, even if more people deserved it. GE is a case in point of why applying the bell curve approach is a bad idea. Given how much human bias feeds into these decisions, can we be sure that controlled attrition is a fair outcome in these cases?
It’s not just about morale and competence. It’s about limiting people’s contribution to being safe, instead of being innovative. Approaches like these force people to toe the line instead of being honest, loyal critiques when the company needs people like that the most.
Assumption#5: Performance Is Profitability: Support functions—such as HSE and Administration— often complain about their “invisibility” in the corporate ladder: “No one notices when we do our job. People only notice when we don’t.” And a more hopeful spin, “Our job is to make it possible for you to do your job.”
Either way, they’re right. Cost centers are generally underrepresented. The problem is, they’re often undercompensated as well. Someone in the administrative department is not responsible for adding revenues. But he/she is responsible for managing costs. At some point—and in organizations’ continuous bid to rationalize resources, cost containment plateaus. And it is simply unrealistic to expect further cost containment. So it’s unfair to measure performance of functions such as administration and HSE targets via tools like the Balanced Scorecard, which place a 25% weight on financial goals. Performance isn’t always profitable. But that doesn’t mean it can’t achieve the company’s goals. Learning to measure what matters, (where the OKR approach comes in), is a positive step forward in this direction.
Assumption #6: All Output Is Quantifiable: The Balanced Scorecard divides all performance in four goals: Processes, Financial, Customer and Learning & Growth. Not all employees have customers. And if Learning & Growth targets are subject to budgetary approvals, or follow a highly regimental process, who is to be held accountable if an individual’s learning and growth targets are missed?
How would a data-driven approach change this? By expanding the measurables to include factors like timeline, number and quality of initiatives, resource management, budget management, and adding weights to each. For instance, if two managers are producing an equal volume of output, but one has a team of 10 resources, and the other has a team of 2 resources, it is easy to calculate qualitative aspects like commitment, initiative and synergy, and assign a quantifiable award structure to them. You can measure soft skill components via SMART goals via exception reporting.
Performance Appraisal Vs Actual Performance: Closing The Gap
Performance appraisals have been at the center of so much work-related stress, there’s been debate around abolishing them altogether.
More workable, however, is to narrow down the gap between performance appraisals and actual performance.Data-backed programs are effective in this regard because they offer transparency and are less at the mercy of personal assessments. To ensure that this advantage is sustained, decisionmakers need to ensure their framework reflects the following characteristics:
Qualitative and quantitative goals: To quote Peter Drucker, “What gets measured gets managed.” Qualitative goals have historically been difficult to measure, which is perhaps why they rarely are managed effectively. As mentioned above, a way around this is to factor in exception reporting. When measuring qualitative goals, consider what would happen if these goals didn’t exist. What would the tolerance threshold be in such a case?
Wide angle: Creativity is another realm where high performance is not often recognized. Worse, if mediocre creativity results in a positive market response, creative practices generally follow a downward trend. So it is important to measure both vertical and horizontal creativity when assessing performance.
Reflective of the agile approach: The Agile Manifesto advocates a consultative and reiterative approach to meeting project objectives. While performance frameworks themselves cannot be continually fluid, they should reflect an agile approach. This means being open for periodic reviews; revising targets in the line of changed market activity (instead of reactively responding to poor annual performance) and so on.
Beyond the checkbox goals: Performance should be about productivity. But productivity does not need to be inhuman. Unfortunately, the ruthless pursuit of goals tends to do that to corporate citizens. A sound performance management framework will identify and reward the components of workplace behavior that don’t necessarily form a checklist. For instance, while many organizations have reward and recognition programs, few document employee testimonials and citations in the long term. A data-driven system could absorb this unstructured data to produce results that would reflect the emotional values that really mattered to employees. For instance: Kindness, courtesy, fun. These don’t show up in Annual Performance Reviews, but employee acknowledgements are full of descriptions citing specific behaviors and responses.
Workplaces are supposed to be about people and not the other way around. A good performance management framework has the power to make work more meaningful, keeping humans at the core of human resources.