Calibration Guide — Design Career Framework

Most important thing to communicate before this process begins: The assessment is a growth tool, not an evaluation. A Developing rating does not mean someone is failing. The framework only works if people are honest, and people are only honest if they trust the process will not be used against them.

Self-assessment guide

Share this section with employees before the assessment cycle begins.

How to assess yourself honestly

Read each skill description carefully before choosing a level. The descriptions are the calibration anchor, not your general sense of how good you are at something.
Think about evidence, not intention. Ask: can I point to specific examples of doing this? Not: could I do this if asked?
Developing is not failure. Every person at every level has skills they are developing. Marking something as Developing is the honest and useful thing to do.
Shaping is rare. Most people, even at senior levels, will have very few Shaping ratings. That is expected and correct.
Do not optimize for the role expectation. Assess where you actually are, not where you think you should be.

Common self-assessment traps

Averaging up because you have done this skill once. Doing something once under guidance is Developing. Doing it independently and consistently is Practicing.
Rating on potential rather than practice. The question is not "could I do this?" It is "do I do this?"
Deflating due to imposter syndrome. If your manager and colleagues regularly come to you for guidance on a skill, and you coach others through it, that is Leading.
Inflating because the category feels important to your identity. The skills that feel most central to your identity are the ones most prone to inflation.

What to bring to the calibration conversation

Two or three specific examples of work that informed your highest-rated skills
One or two skills where you are uncertain about your rating and want your manager's perspective
The one or two skills you most want to develop in the next 6–12 months
Any skills where the framework description doesn't quite match your experience or role

Calibration conversation: step by step

Plan for 60–75 minutes. Complete your independent assessment before seeing the employee's.

Critical: Complete your independent assessment without seeing the employee's self-assessment first. Once you have seen theirs, your judgment is influenced. Do not skip this step.

Step 1 — Set the frame5 minutes

Establish that this is a growth conversation, not a judgment. Reduce defensiveness before it has a chance to start.

"We both completed this independently, which means we probably see some things differently. That's not a problem — those differences are the most useful part of this conversation. My goal is that you leave with a clearer sense of where your strengths are and where there's a real opportunity to grow. There are no wrong answers and nothing here feeds directly into your compensation review."

Do not open the matrix before the frame is set. Do not reference performance reviews, comp cycles, or promotion timelines.

Step 2 — Share assessments simultaneously5 minutes

Open the matrix together with both assessments visible. Give 2–3 minutes of silence for both of you to scan the full picture before speaking. Then ask: "What stands out to you when you look at this?"

Watch for immediate defensiveness about gaps — note it, do not address it yet. Employees often expect more disagreement than exists.

Step 3 — Explore areas of strong alignment10 minutes

Validate genuine strengths with specificity. Do not rush past this to get to the gaps. Employees need to hear that their self-perception of their strengths is accurate before they can engage productively with gaps.

"We both rated you at Leading on systems thinking. I want to make sure you know what specifically I'm seeing that puts you there."
"You rated yourself Practicing on this and I rated you Leading. Tell me what makes you uncertain about this one."

Common mistake: Generic validation ("You're really strong across the board") signals you haven't thought carefully and makes the gap conversation feel like it's coming out of nowhere.

Step 4 — Address gaps carefully20–25 minutes

Two kinds of gap require completely different approaches.

When employee rates themselves lower than you

Do not just tell them they are wrong — give specific evidence for the higher rating
Explore what is making them uncertain: "What would it take for you to feel more confident rating yourself at Leading here?"
Acknowledge imposter syndrome directly if relevant: "What you're describing sounds like imposter syndrome. The evidence I see is that you are already doing this at Leading level."

When employee rates themselves higher than you

Never lead with "I disagree." Lead with curiosity: "Walk me through what you were thinking when you rated yourself at Leading here."
Listen fully before responding. Often you will hear that the employee is describing something adjacent to the skill, not the skill itself.
Do not capitulate under social pressure: "I want to hold this rating for now. Let's revisit in 6 weeks with specific examples."

Phrases that shut down the conversation

"That's just not how I see it" — too blunt, triggers defensiveness
"I think you're being too hard on yourself" — dismisses their self-knowledge without exploring it
"Let's just split the difference" — destroys the integrity of the assessment

Step 5 — Identify 1–2 development priorities15 minutes

Leave with a concrete, agreed focus. Not a laundry list.

"If you could move the needle on one skill in the next six months, which one would make the biggest difference to your work and your career?"
"What would Leading look like in practice for you on this skill? What would you be doing differently?"

Before the conversation ends, document: the 1–2 agreed priorities, specific opportunities to practice those skills, how you will both know progress is being made, and a specific check-in date.

Step 6 — Close with the big picture5 minutes

"Stepping back from the matrix — how does where you are right now compare to where you want to be in two to three years? Does this conversation change anything about how you're thinking about that path?"

The conversation should end with the employee feeling more clear, not more measured. If the employee has a legitimate critique of how a skill is described, that is valuable framework feedback, not resistance.

Calibration failures to watch for

Read this section before each cycle, not just at onboarding.

Leniency biasHigh risk

Rating everyone higher than their actual performance to avoid difficult conversations. Creates grade inflation that disadvantages high performers who are correctly rated by rigorous managers.

Signals

Almost everyone on your team is at Leading or Shaping
You feel uncomfortable marking anything as Developing
Your team's ratings are consistently higher than peers' teams at equivalent levels

Correction

Ask: "If I showed this rating to a peer manager, would they agree?"
A Developing rating is supportive, not punitive, when delivered with a development plan
Cross-calibrate with another manager before finalizing

Recency biasHigh risk

Rating based on the last 4–6 weeks rather than the full period.

Signals

Your justifications are almost entirely from the last quarter
Ratings shifted significantly after a recent high-profile success or failure

Correction

Keep a running doc of notable observations throughout the year
Ask: "If I removed the last 6 weeks, would this rating change?"

Halo and horn effectHigh risk

Allowing one strong skill (halo) or one weak skill (horn) to color the entire assessment.

Signals

All ratings cluster at the same level across very different skills
Your ratings feel like a general impression rather than skill-by-skill evidence

Correction

Complete each skill independently before reviewing the full picture
For each skill ask: "What specific evidence do I have, independent of how I feel about this person overall?"

Social pressure capitulationHigh risk

Changing your assessment during the conversation because the employee pushed back or expressed disappointment. The most common failure mode in the calibration conversation itself.

Signals

Ratings changed during the conversation without new evidence being introduced
You felt relieved when agreement was reached, rather than confident the rating was accurate

Correction

New evidence justifies changing a rating. Discomfort does not.
Changing a rating under pressure tells the employee that the process is negotiable — which undermines every future conversation

Small team scope inflationMedium risk — research and design ops specific

Rating someone higher because the team's small size has required them to operate above their level by necessity.

How to handle it

Separate the assessment from the acknowledgment — the person deserves recognition for stepping up, but that belongs in the development conversation
Rate the skill at the level the evidence supports, not the level the work required
Document the scope inflation explicitly: useful context for the promotion conversation

The equity risk

Inflated ratings create a painful correction when the team grows and calibration tightens
That correction feels like a demotion even though nothing changed
Honest rating now, with documented context, protects the employee and the framework

Cross-manager calibration

Runs annually after individual conversations are complete. 90 minutes, all managers present, SVP facilitates.

Prepare calibration anchors. Identify 2–3 individuals whose skill levels are broadly known and agreed upon. Use them as reference points to recalibrate standards before individual reviews begin. Never made public to the broader team.
Start with the anchors. Each manager shares ratings for anchors on two or three key skills. Surface disagreement and discuss until alignment is reached.
Focus on promotions and significant gaps. Do not review every person's full matrix. Focus on level changes and significant self-vs-manager gaps.
Name outliers explicitly. If one manager's team has significantly different ratings, name it as a calibration conversation.
Document any rating changes with rationale. Managers then communicate changes to employees directly — never silently updated.

Annual cadence

Time	Activity
Q1	Assessment cycle opens. Employees and managers complete independent assessments in the same window.
Q1	Calibration conversations completed within two weeks. Development priorities documented.
Q1	Cross-manager calibration session runs before cycle closes. Development plans finalized.
Q2–Q3	Development priorities surface in regular 1:1s. No formal matrix update between cycles.
Q3	Mid-year development check-in (30–45 minutes). Progress on priorities only — not a full re-assessment.
Q4	Framework review. Competency descriptions, level expectations, and AI fluency rubric assessed for updates. Living skills addendum updated.

On promotion: A strong matrix profile is evidence for promotion readiness. It is not a guarantee. Promotion decisions run on the company's standard cycle, separately from the assessment cycle. Managers should be explicit with employees about this distinction.