What’s wrong with the annual review process… and how to fix it.

One thing I don’t miss about big enterprise environments is the annual performance review cycle. When and where I’ve participated in this cycle, it goes something like this:

Your manager spends weeks writing up the annual review document for you and all of your team members. You are asked to self-evaluate and bring your own annual review document. You have an open and honest meeting with the boss to discuss any discrepancies between the two assessments, celebrate your achievements, talk about your career goals, and assess any growth opportunities that will prepare you for the next level.

You work together to achieve some fair, final number that represents how you’ve performed over the year. I’ve seen three-level scales (Needs Improvement, Meets Expectation, Exceptional), five-point scales, scales where the score is a convoluted weighted average of a whole bunch of measurable factors, and many methods in between.

Based on your review, you receive the raise or promotion that your review deserves, measured against your performance against last year’s goals, usually accompanied by a profit-sharing bonus based on how well the company did that year. Sometimes your entire team gets a huge bonus for the amazing work you all did.

This is how it should go. If you’re shaking your head at this point at my naivete, I don’t blame you.

The Problem:

Getting a raise or a promotion, especially in a deep and structured organization, requires being measured at the top of the scale of your organization’s review metrics. Organizational budgets need to be set in advance for the next year, however, so how does an organization handle all the raises and promotions it might be doling out the following year based on reviews that haven’t happened yet?

While I feel I’m giving away how the magician’s trick is done, I don’t think it’s a huge secret: most of the time, the organization budgets a particular number of promotions and raises that will be available. By the time the review occurs, the die has already been cast, and every manager knows what the budget is for raises and who can be promoted — long before they start the formal review process.

From the top down, the budget for the next year has already been set by the executives of the organization, and the fourth quarter is spent figuring out how to allocate that budget. Managers lobby for their good people to be considered for advancement and reward. They already know roughly who is going to be promoted, or at least roughly how many. Then the different departments are given allocations and divide them up between directors, senior managers, and managers.

Then the review has to happen, but because the big decisions have been made, the reviews are written to justify the data, not to drive it.

Consider the case study of a rockstar team in this environment, because it is possible to have an entire team of superstars. It’s actually rather common. Good managers gather or grow talented folks into an effective team. It’s possible that every member of a great team performs better than the best member of an average team, often because a smoothly-functioning team becomes more than the sum of its people.

For sake of discussion, let’s examine a simple three-level system. A manager can give one of three ratings. “Needs Improvement” indicates that the employee is not meeting the minimum basic requirements of the job, requiring a remediation plan to get on track. “Meets Expectations” indicates that an employee is doing a fine job. Nothing amazing, but solid work that can be more or less relied upon. The last rating of “Exceeds Expectations” indicates the employee has done amazing work, stepped up, and should be called out for promotion or other acknowledgement.

Let’s pretend that the business believes every team internally has it’s own Gaussian distribution of talent, and that each team is sufficiently small that every manager gets one Exceeds Expectation to give out, one Needs Improvement, and the rest get “Meets Expectations.”

So the managers of each team have a single Exceptional review to give out, and have to give the rest Meets Expectations. Maybe your exceptional team member gets a promotion or a raise, but that’s the only one that can. Because the HR policy the reward system is designed around assumes a bell curve distribution of employee talent. But the talent in the organization is not a bell curve.

When you can’t recognize an entire team as exceptional for political reasons, what have you done? You’ve just demoralized an entire high-performance team. You’ve not only not recognized them for their contributions, you’ve actively denigrated their performance by asking them to sign off on a “meh” review for great contributions. Why? To fit some predetermined assumed distribution of talent? We know it’s not right, but we don’t fight it. Why not?

Ultimately, the entire annual performance review is cheap chicanery to back-justify management promotion and salary increase decisions that are at best predetermined, and at worst capricious.

Employees know it, too. I’ve been through the process enough times to understand that it’s a show. A sham. And to be honest, it makes me upset to know that it’s not only a sham, but that managers think employees don’t see it’s a sham.

Stop it. Just stop it.

So let’s rethink the problem space just a little bit. What are the true requirements that the annual performance review cycle is supposed to fulfill?

• Ensure the employee and the manager agree on their perception of the employee’s performance
• Motivate salary increases, if needed
• Provide evidence for promotion, if needed
• Reward employees for exceptional contributions through performance bonuses
• Provide company performance-based bonuses (profit sharing)

The traditional annual performance review fails to meet these goals in a number of ways:

1. A year is a long time to wait for feedback. For positive reinforcement to work, the reward has to quickly follow the behavior. If an employee does something amazing in Q1, a reward nine months later doesn’t have the same effect on future behavior.
2. A year is a long time for goals to be achieved. Often by the time you do coaching and reach the goal you had planned for the year, the goal is no longer relevant.
3. Salaries need to stay in step with the employment market or the organization risks losing their best people.
4. Waiting on the annual review cycle for promotion leaves gaps in the workload as positions are held open for the annual restructuring.
5. The de-motivational effect of applying curve-based ranking to teams of great people can cause top-tier performers to look elsewhere.
6. Managers are forced to look exceptional performers in the eyes and call them average, when they both know that it’s part of a charade.

The process is painful, time-consuming, and anti-productive for everyone involved.

What can we do instead?

How could you achieve these goals without annual reviews?

Most of the problem with the annual review cycle comes from that disconnect between the need of the business to make small, predictable increases to annual costs and the goal of every worker to be the best at what they do. Use of the bell curve to dole out the best reviews achieves the business need, but grading employees on a curve can be devastating to all but your top folks.

So with that, I have a few ideas to improve the process for you to consider. I know that not everyone has the ability to change their processes, but it never hurts to seed the ideas and rationales into the business at every level. Good ideas find their champions.

Decouple job performance and salary for that skill set. The salary range for the industry for a given position is very difficult in technology. As an industry, we do not have great common definitions for what constitutes a Software Engineer II or a Security Analyst III. That creates a situation where each individual has to be assessed against what they actually do, not some role that isn’t well-defined. But the value of an individual’s skills is at least estimable, and the company has to provide at least adequate compensation for those skills or it runs the risk of losing good people.

The performance on the job is a completely separate issue from the salary range. Obviously the company’s best Senior Developer IX may be paid more handsomely than its worst, but that shouldn’t affect the review. Salary adjustments and industry standards and gratitude are all fine to discuss, but the only three performance assessments should be:

1. You’ve got to step up or we’ll have to part ways.
2. You’re doing well, so we’ll continue your employment.
3. You’re doing amazing! Are you challenged enough? Is there more that you’d like to do so that I may help you get there?

The most sensible approach to adjusting salaries is to re-index the salary to what the employee could get on the open market, to prevent that employee from leaving the company for avoidable reasons. If the company can’t afford to do that, then it’s time for the employee to move along. If his or her skills could earn the employee more, adjust that salary, promotion or not. If the employee has stagnated, and the company thinks it can get better on the open market than that employee (after adjusting for organizational knowledge value), say so and let the employee start looking. If you think you can find better, cheaper, you owe it to the business to do so.

Separate profit-sharing from the individual performance review. How well the company performs overall is due to the contributions from all employees, to the best of their ability, thinking like owners. This is a collective goal and is more a measure of how the organization functions as an entire organism, not an individual thing. Sharing the profit provides a strong feedback loop that encourages owner-thinking

Tie employee bonus to company priority. — The point of every employee being a proxy for the owner means that the company should dole out bonuses based on how well the employee makes decisions as the owner would. That is, how motivated they are to meet the owning entity’s goals. You do know what the goals are for every employee, right? Bonus monthly based on some simple but critical top-line driver, instead of some multivariate calculation. Employees can then see immediately how their actions are lining up with company growth.

Review continuously. Reviews shouldn’t be annual. No employee should receive any performance-based surprises. Continuous coaching, mentoring, and growth all involve constant feedback, and it is one of any manager’s primary responsibilities to assess and give feedback on the performance of each team member. Frequent, positive feedback is touted as a method to improve morale. It’s free, it’s easy, and every employee needs it to perform at their best. Ensure all employees know exactly where they stand through consistent feedback, and the annual review becomes a thing of the past.

THAT Conference encourages alignment of great workers with great companies. We enthusiastically endorse people finding their best gigs and living their best lives. Lots of people find their dream jobs on our Job board every year.