Review Design II: Determining Assessment Types and Questions

Welcome to LarkApp's How to Run a Great Performance Review series! In this post 3 (of 6), we’ll continue designing your review by helping you pick assessment types (self, peer, manager) and create your questions.

Assessment types

Assessment types determine what perspectives are represented in your review. We recommend always including self-assessments, which are crucial for employee development, and manager assessments as a manager’s job is to evaluate performance and coach their team. As for peer assessments, they’re useful when your employees work cross-functionally and collaboratively. However, they do make for more work, so it’s important to decide whether your priority is being thorough or less time commitment.

Number of questions

We recommend keeping the number of questions small to ensure that the workload from running this review process isn’t onerous. The number of questions depends on the mix of qualitative vs. quantitative responses. We find that asking 2–3 questions for qualitative feedback and 2–3 questions for a quantitative gauge (i.e. multiple-choice, ratings) is a great mix.

Question content

Great questions are specific, concise, and not leading. Below are just a few examples for each assessment type:

For a self assessment…

  • What were your top achievements during this time period? Please list 3–5.
  • What are 3–5 areas in which you want to improve?
  • What are things that you would like to do more of?

For a peer assessment…

  • What are your peer’s strengths?
  • What could your peer improve upon?
  • How would you like to work with this peer in the future?

For a manager assessment…

  • What are your report’s performance highlights during this time period?
  • In what ways could your report improve?
  • Are there areas that you would like your report to spend more time on/grow into?

Rating scales

Not quite this kind of rating.

Rating scales are commonly used in performance reviews, and they have implications that can significantly impact your process and results. With rating scales, your 3 major variations are:

(1) No ratings

They’re good when:

  • Your top priority is your employees’ comfort — people don’t love being rated.
  • You want to minimize prep time for the review — you have to get everyone on the same page about what rating scales mean, which means more prep work.
  • Your focus in on learning and development, and your primary goal is facilitating performance conversations.

(2) Numerical ratings

They’re good when:

  • Your org skews older, or has a more formal culture.
  • You want to compare data across employees. Though it’s possible to compare employees and make management decisions without a ratings structure, using one does make it easier. This is helpful when making personnel and/or compensation decisions. However, it’s important to consider that it’s hard to get consistent ratings across managers and across departments, which may make comparative data less accurate — so you want to invest time in training everyone on what each rating means.

(3) Emoji scales

Emojis are changing our way of communicating, and these scales are a novel approach to ratings. This will probably require additional training, as emojis haven’t been traditionally used, so plan accordingly.

They’re good when:

  • Your org is full of millennials, or the culture is more informal — i.e., people communicate mostly via Slack.
  • You want a form of ratings, but you feel that numbers come off as too transactional.

HR-only questions

In addition to determining response options, you may also choose who views certain responses. Should all parties involved see them, or should some answers be sent to HR only? HR-only questions solicit responses that are not seen by employees or managers — only by HR. HR-only questions are useful when:

  • You want employees to send upward feedback that they may not be comfortable sharing with their manager.
  • You want managers to fill out ratings not shared with their direct reports, for personnel decision purposes.

However, you may not want to use them when:

  • You want to create a culture of total transparency.
  • Your review will be used primary for learning & development.

In tying all this together, we’ve proposed some review designs that are aligned to your desired outcomes.

Note: for all options, you may choose to include peer assessments if your employees spend a lot of their time working cross-functionally.

We hope that this blog post helped you create your assessments! Next up: Review Implementation I: Preparing for the Review.

If you want to learn more about the best way to put this into practice, please email me at .

Don't miss these stories:

No items found.