Skip to content
Culture & Coaching

A Career Ladder Is a Calibration Tool Before It Is a Promotion Tool

Most people open a career ladder to find out what a promotion needs. The Etsy ladder is more useful read the other way round: as the shared language managers use to agree on where someone already is.

Culture & CoachingEngineering Management Academy6 min read

Etsy published its Engineering Career Ladder as two pages. One lists the levels, Engineer I through Principal Engineer. The other lists five competencies, Delivery, Domain Expertise, Problem Solving, Communication and Leadership, each described at five stages from Beginner to Leading Expert. Most readers go straight to the levels page, find their title, and read the bullet points under it as a promotion checklist.

We think that is the less useful way to read it. The design choice that makes the Etsy ladder worth studying is the separation, and the separation is there to help managers agree with each other before any promotion is on the table.

Three columns for every level

Each level on the Etsy ladder has up to three parts. Expectations describe the scope of problem the engineer takes on, from "narrowly-defined tasks with clear instructions" at Engineer I to "significantly complex, unscoped problems" at Staff. Track record describes evidence over time. For Senior Staff Engineer the ladder says your work "has been validated as a success over a significant window of time, typically at least a year". Competencies describe a profile across the five areas, and the wording is deliberately hedged. Staff Engineer I reads "Engineers at this level will generally be Experts in one competency or Advanced in all competencies". From Senior Engineer II upward, engineers "should not be Beginners in any competencies".

The competencies page adds the rule that holds the whole thing together. "Mastery of a competency is cumulative." An Advanced engineer is expected to still do everything the Intermediate column describes.

Three columns is more work to write than one. The reason to do it shows up in the room where managers compare notes.

What goes wrong with a one-column ladder

Picture four managers, eight promotion cases, one afternoon. With a ladder that is only a list of expectations, every case becomes a scope argument. "She led the migration" meets "he led the re-platforming", and the discussion turns on whose project sounded bigger. Scope is real evidence, and it is also the easiest column to inflate, because a manager controls which projects their reports are given.

The track record column slows that down. A migration that shipped in March and has not been touched since is a different fact from one that has run a year of peak traffic. Etsy's wording for the higher levels, "led numerous projects with consistently high value", asks for a pattern rather than a highlight.

The competency column does something the other two cannot. It lets you compare an engineer on the payments team with one on search. Neither manager knows the other's codebase, and both can read the Advanced column for Delivery, which asks for reliable estimates of "time, effort, and risk" and for taking responsibility for plans "both when they're on track and when they aren't". The question in the room becomes "which column is this evidence for?" rather than "whose story is better?"

Calibrate first, decide second

The practice we recommend uses the ladder before a promotion cycle rather than during one. Once a quarter, each manager places each of their engineers on the five competencies using the ladder's own language, with one or two pieces of evidence per placement. Then the managers compare. The purpose of the meeting is to find the places where two managers would have placed the same engineer differently, and to resolve those with evidence rather than advocacy.

Make the evidence concrete. The Intermediate column for Delivery says "You ship complete solutions, but release them in small, safe patches". An engineer who planned a schema change that three teams read from, migrated it behind a flag in four pull requests and wrote the rollback has evidence for that line. An engineer who was on the project while it happened does not, however good their standups sounded.

The Advanced column for Leadership says "You take action when there is a problem without a clear owner, and you don't shy away from being the accountable party". The on-call rota that nobody owned until someone rewrote it is that line, with a date. A manager who cannot point to a date should place the engineer one column lower and say why.

After two or three of these sessions, managers start using the same words for the same behaviour. That is calibration. Promotion decisions become a consequence of it rather than the occasion for it, and the engineer hears nothing in the promotion conversation they had not already heard in a 1:1.

What the manager refuses

Three refusals keep the tool honest.

Refuse to promote on expectations alone. Scope without track record is a project that went well once.

Refuse to let the hedge words do extra work. The Senior Engineer II track record says you "regularly demonstrate at least one of the Staff Engineer expectations". "Regularly" does not mean once, in the quarter before the cycle, on a project you were handed for the purpose.

Refuse to let the ladder drift into describing your best people rather than the level. Etsy's Leading Expert column for Leadership includes "You make hard decisions in the face of uncertainty". That is a description of a level, and any engineer who meets it should be placed there, whether or not they resemble the person the column was first written about.

One more thing worth noticing. Etsy's Leadership competency, even at Leading Expert, describes an individual contributor. There is no manager track on this ladder. The same calibration logic applies to managers, and most organisations have never written the columns for them.

Why this matters beyond one company

Calibration is also where fairness lives or dies. If two engineers doing the same quality of work land at different levels because one manager writes a better packet, the ladder has become a measure of the manager. A shared competency language, with evidence per line and a regular cross-team comparison, is the cheapest protection against that we know of.

It is also why the Academy assesses capability against a public framework rather than years in the role. Tenure is the one-column ladder of the industry, and it has the same failure mode.

Where this sits in the standard

Equitable distribution of opportunity, including who gets the projects that generate promotion evidence, sits in the Culture & Coaching domain of the EMA Competency Framework. The Body of Knowledge unit on inclusive practices and equitable opportunity distribution covers the calibration meeting itself, and the performance management unit covers how evidence is gathered through the year rather than assembled at the end.

To start from shared lines instead of a blank page, the career-ladder template sets out every competency at team, organisation and executive scope, free to copy and adapt.

EMA-I tests this at team scope, usually as a scenario in which two managers disagree about the same engineer and you have to decide what evidence would settle it.

Filed undercareer laddercalibrationpromotioncompetenciesengineering management
This sits under Culture & Coaching in the Competency FrameworkThe Body of Knowledge chapter covers the full set of competencies the examinations assess in this domain.Read it →

Test your judgment, not your recall.

EMA-I is a scenario-based examination in the five domains of the framework. There is no course to buy and no attendance to prove. Sit a sample first, or read what the credential covers.

Keep reading