Career Ladders That Don’t Trap People
How to design IC and management ladders that give people a real path, hold a consistent bar, and avoid becoming a filing system for politics.
A career ladder is a public answer to two questions: what does ‘great’ look like at each level, and what changes between levels? When the answer is clear and consistently applied, people grow and stay. When it’s vague or politicized, your strongest people leave.
Why ladders matter
- Make expectations explicit so feedback is about evidence, not personality
- Calibrate hiring decisions across teams (a Senior is a Senior)
- Defend compensation decisions with structure, not anecdote
- Give people a credible growth path that doesn’t require switching companies
- Force leadership to articulate what ‘scope’, ‘craft’, and ‘leadership’ actually mean
Anatomy of a good ladder
- 1DimensionsUsually 3–5 (e.g., Scope, Craft, Execution, Communication, Leadership). The same dimensions appear at every level.
- 2Level descriptorsA paragraph per dimension per level, written in observable behavior — not adjectives like ‘strong’ or ‘exceptional’.
- 3ExamplesReal, anonymized examples of work product at each level (designs, code reviews, projects, talks).
- 4Promotion criteria‘Sustained performance at the next level for ~6 months’ + the evidence packet rules.
- 5Calibration mechanismCross-team committee that reviews proposed promotions for consistency and bias.
IC and management tracks
A modern ladder has two parallel tracks at senior levels — individual contributor and management — with equivalent compensation and prestige. The tracks diverge above ‘Senior’ and re-converge at the most senior levels (Principal/Distinguished IC ↔ Director/VP).
- Technical / craft depth
- Scope of system or problem
- Influence without authority
- Mentorship and standards setting
- Team performance and growth
- Hiring and retention
- Cross-team coordination
- Organizational design
‘Equivalent’ does not mean ‘symmetrical’. The IC track usually has fewer slots at the very top — Principal-level work is genuinely rare. What must be equal is comp, voice in technical decisions, and respect.
A reference set of levels
| Level | Scope | Manager equivalent |
|---|---|---|
| L1 — Junior | Subtasks under guidance | — |
| L2 — Engineer | Owns features end-to-end | — |
| L3 — Senior | Owns a system; mentors others | Team Lead (player-coach) |
| L4 — Staff | Influences across multiple teams | Engineering Manager |
| L5 — Senior Staff / Principal | Sets technical direction across an org | Senior Manager / Director |
| L6 — Distinguished / Fellow | Company-wide technical leadership | VP / SVP |
Promotions and committees
- Manager nominates with an evidence packet (artifacts, peer feedback, ladder-mapping)
- Calibration committee (3–5 peer managers) reviews against the ladder
- Committee votes promote / not-yet / not-ready; rationale recorded
- Promo communicated with a development conversation, not just a comp bump
- ‘Not yet’ comes with a concrete plan to close the gap and a re-look date
If your promotion outcomes correlate with which manager someone has more than with the level descriptors, your ladder is failing. Run a calibration audit.
Pitfalls and anti-patterns
- Ladders written in adjectives (‘exceptional impact’) instead of behaviors
- No examples — every team interprets ‘scope’ differently
- Promotion-by-time (‘they’ve been here 3 years, time for Staff’)
- Manager track paid more than IC track at the same nominal level
- Hidden ‘shadow ladders’ for favored founders or early employees
- Levels created for retention without real scope change (‘Senior II’)
- Updating the ladder annually without updating people’s level assessments
- Will Larson — Staff Engineer & ladders — Lethain.com
- Patrick Kua — Engineering ladders — patkua.com
- Rent the Runway — Engineering ladder (open) — RtR
- progression.fyi — Public career ladders — Progression
- Camille Fournier — The Manager’s Path — O’Reilly
Read next
All playbooksThe system around the review matters more than the review itself. A modern approach to goals, feedback, calibration, and the conversation.
Spans of control, layer counts, and the operating signals that tell you a team needs to split, get a manager, or be reorganized — without burning trust.
Why most feedback fails, and the small set of frameworks that make it useful, specific, and bias-aware.