The Target Level Claim, Scored by Ladder Axis Before the First Recruiter Call

This playbook helps you decide which level to sell yourself at (Senior or Staff, Manager or Senior Manager) by checking your own records against public career ladders, so the decision is made before an employer makes it for you. It is for experienced candidates rewriting their resume and profile ahead of a search. By the end you will have a written level verdict, a translation of that verdict onto each target employer's level names, a resume and profile top block that states your scope, and a scripted level question for the first recruiter call.

When does an employer decide your level?

Often before your first interview. Pragmatic Engineer writes that "leveling decisions have often been made before the interviews, based on the resume and perceived experience". The recruiter screen then calibrates that guess, and the loop and committee start from it.

No employer publishes official documentation of when the level gets set, so the picture below is pieced together from practitioner write-ups and community reports. Read it as a pattern, not as a rule for any one company.

The practical point: if your materials don't state a level, the reader fills one in, and the default isn't neutral.

Why the default is one rung low

Staff is a scarcity claim. In Refolk's index of professional profiles, there are far more Senior engineers than Staff engineers, so a recruiter who has seen thousands of profiles will assume Senior unless you show otherwise.

The US Staff count is about 15 times the UK Staff count, and 10 of the 25 UK Staff profiles I sampled were in London. US Staff employers in the sample include Databricks, GitHub, Vanta, Starburst, and Airbyte.

There is a second force pushing you down. As one practitioner account puts it, "candidates whom interviewers are unsure about would often get offered a level below." Uneven evidence across the axes is exactly what creates that doubt.

Ambiguity resolves downward. Uneven evidence across the axes is the ambiguity.

Getting it wrong is expensive. Pragmatic Engineer reports that staff-level engineers are "often offered senior positions when they meet but don't easily exceed the bar", and that getting back to the former level "often requires 2-3 years".

What do career ladders actually score?

Public career ladders score your work on a small set of dimensions, usually four to six according to one practitioner summary. Your title is not one of them. This guide uses five axes: scope, autonomy, impact, complexity, and decision rights. A ladder axis is a dimension the ladder describes separately at each level.

I could not find a published count of how often each axis appears across ladders, so the five-axis list is a working synthesis, not a measured standard. If you want to check it, progression.fyi catalogues 75 frameworks and you can tally the axes by hand. Here is what the ladders I used show:

• Dropbox runs IC1 to IC7 and separates levels by "extent of influence". The range goes from "working within the scope of my team with specific guidance from my manager" to "influencing department and company-wide strategy".
• CircleCI runs a 6-level ladder, E1 to E6. Its early levels are scoped "within task, within epic/project, within team".
• One ladder in swyx's public list breaks Leadership into "Autonomy, Judgement, Initiative, and Consensus Building".

What each axis proves, and what it looks like when it lies

For a Staff claim, add one more test. Will Larson describes four Staff archetypes: Architect, Tech Lead, Solver, and Right Hand. If your evidence doesn't make you clearly one of them, the claim will be hard to defend in a loop.

For the manager track, the same axes apply, but the public evidence is thinner. Dropbox publishes a manager track from M3 to M7. Where your target employer publishes a manager ladder, use its dimensions, and expect decision rights over people, hiring, and direction to carry more weight.

How big is the gap between Senior and Staff?

On posted base-pay midpoints, Staff pays about 20% more than Senior. In big-tech total compensation, the step is roughly 49 to 62%. The difference comes from equity and bonus, which "usually grows faster than the base".

Read the table with two caveats. The first row comes from posted ranges. The other three come from a third-party page that cites Levels.fyi, and that page's own Amazon level labels contradict each other, so treat the Amazon row as approximate. I found no posting-based figure for the manager to senior manager step, so if you are on the manager track, look up your target employers directly on a free level-comparison site.

Why the posted range can't tell you the level

If one range is as wide as the whole gap between two levels, the range can't tell you which level the role is. Worse, one pay-transparency write-up notes that "a single posting might cover L4-L6". The only reliable fix is to ask, which is covered below.

How do I map my job level across companies?

Translate your verdict through a published level map, then check it against at least one second source, because the maps disagree. A level name is useful only once you can say it on a call: "this is L6-equivalent at Google" lands more precisely than "I am pretty senior".

Apple is commonly listed as ICT4 for Senior and ICT5 for Staff. The Dropbox Senior placement is my inference from the ladder's title list and is not verified. One outlet warns that matching names "do not establish equivalent responsibilities or hiring standards". The map gets you into the right conversation; it doesn't prove you belong there.

Correcting for title inflation

Posting titles have drifted upward, so the title on a posting overstates its level just as your own title can. Datapeople found that:

• A quarter of tech jobs once considered junior now carry senior titles.
• Twice as many junior jobs include the word "Lead", and 20% more mid-level jobs include the word "Senior".
• Their experience bands are junior under four years, mid four to 10, and senior 10 or more.

Recruitment firm Walters People found that Lead and Manager titles asking for at most two years of experience were up a combined 53% in the UK and Ireland, and 48% in the US. The correction takes about 10 minutes per posting: ignore the title, read the years of experience required in the posting body against Datapeople's bands, and note whether the pay range is wide enough to span more than one level.

The level claim procedure, step by step

Seven steps, done in order and finished before the first application goes out. Steps 1 to 3 produce the verdict, steps 4 and 5 translate it, and steps 6 and 7 get it in front of the employer.

Step notes, with time and what good looks like

1. Gather records (2 to 4 hours, an unverified estimate). Will Larson's advice applies directly: "write the promotion packet for yourself. Run it, talk it through with some of your peers." Peers catch the two common distortions: underselling your own decisions and overselling the team's. If your history is scattered across old reviews, design docs, and a stale profile, Refolk drafts the resume from your own history, which gives you a dated project list to start the packet from instead of a blank page.
2. Pull the axes (1 to 2 hours). Sources disagree on the order of steps 1 and 2. Interview Copilot advises you to "start with the dimensions the company actually evaluates", which puts the target ladder before your records. If you have one clear target employer with a published ladder, do step 2 first. If you are searching broadly, gather your records first so the ladders don't steer what you remember.
3. Score each axis. There is no time estimate in the sources for this step, and it is the one worth slowing down on. Use the template below and the matrix after it to reach the verdict.
4. Translate (about 15 minutes per employer). Record both sources you checked. When they disagree, as on Meta, write down both names and ask the recruiter which one applies.
5. Correct for inflation (10 minutes per posting). Tag each posting with the level you think it is, and drop or re-pitch any posting more than one level away from your verdict.
6. Rewrite the top block (2 to 3 hours). Covered in the next section.
7. Run the ask. Covered in the section after that.

TARGET ROLE FAMILY:
LADDERS CHECKED (2 or 3, by name):

AXIS            | DATED EVIDENCE (month, year, project) | OUTCOME | LEVEL IT SHOWS
Scope           |                                       |         |
Autonomy        |                                       |         |
Impact          |                                       |         |
Complexity      |                                       |         |
Decision rights |                                       |         |

AXES AT TARGET LEVEL (count of 5):
WEAKEST AXIS:
STAFF ARCHETYPE, IF CLAIMING STAFF (Architect / Tech Lead / Solver / Right Hand):
VERDICT (one line): I am selling ______ because ______.
DATE OF VERDICT:

What should the resume top block and profile headline say?

They should state the target level and the scope that proves it in the first line, in wording any employer can read, not your past title. Exponent's advice is to put the target level "in the first line of your personal pitch". The same applies to your written materials.

The top block is the first three or four lines of your resume and your profile headline. It has to survive a few seconds of skimming, because that skim is where the pre-interview level decision is made. What it should carry:

• The level, stated plainly. "Staff-level backend engineer" or "Senior engineering manager", whatever your verdict says.
• The scope in numbers you can defend. How many teams your work touched, and which decision you owned.
• One outcome tied to your strongest axis. Pick it from the evidence sheet, not from memory.
• No raw big-tech level labels. Translate them or leave them out, for the Amazon L5 reason above.
• The same claim on every surface. If the resume says Staff and the profile headline reads like Senior, the reader picks the lower one.

The top block has to match the posting without changing the level. Each application stresses different axes, so the outcome line changes, but the level line stays the same. Refolk tailors the resume to each posting from the same underlying history, which keeps the level claim fixed while the supporting evidence changes for each role.

How do I ask a recruiter what level the role is?

Ask directly on the first call, in plain words, and ask for the answer in writing. One interview-prep source advises you to lock in "the approved role, internal level, and compensation band during the first recruiter conversation". After that point, changing the level means reopening a decision instead of shaping one.

OPENING
"I'm focused on roles scoped at [Staff / your translated level name]. For the last [N] years my work has been [one-line scope from your top block]."

ASK 1 - LEVEL
"What level is this role scoped for?"

ASK 2 - RANGE COVERAGE
"Does the posted range cover one level or several?"

ASK 3 - TIMING
"Is level decided before the loop, after the loop, or through a separate calibration?"

ASK 4 - SPLIT FEEDBACK
"If interview feedback is split across rounds, how does that affect the level?"

CLOSE
"Could you confirm the role, internal level, and band in a quick email so I prepare for the right loop?"

FOLLOW-UP EMAIL (same day)
"Thanks for the call. To confirm my notes: [role], scoped at [internal level], band [range], with level decided [before / after / at calibration]. Let me know if any of that is off."

Reading the answers

• A specific level that matches your verdict. Go ahead. Prepare stories tagged to the axes that level emphasises.
• A level one rung below. Restate your scope evidence once, then ask whether the requisition can be scoped higher. Community reports include candidates told they were "not fit for the level but offered lower IC positions", so it is better to find this out now than at the offer.
• "The range covers several levels." Ask where your profile currently sits within it.
• "Level is set after the loop." This is a real answer, not a dodge. Candidate accounts describe OpenAI this way. Write down the process and prepare the rounds that decide level, such as design rounds, which one Meta interviewer says decide level.
• A committee decides. One community report describes a committee choosing between "a weak Staff Engineer or a strong Senior SWE". Ask what the committee weighs.

How the level claim goes wrong

Most level losses come from seven predictable mistakes, and each has a quick check you can run before the call. They are listed below in roughly the order they bite.

1. Using a title as evidence. Inflated titles look like seniority, both yours and the posting's. A quarter of junior-era tech jobs now carry senior titles. Check: require scope evidence (teams affected, decisions owned) on the evidence sheet. A title alone never fills a row.
2. Under-titled big-tech history. Your scope is real, but the raw label reads low. Saying "L5 at Amazon" can downlevel the loop. Check: translate the level in step 4 before any call, and lead with scope.
3. Reading the range as a level. A $38K average range is as wide as the whole Senior to Staff posted-midpoint gap, so a range that looks like a Staff band may be Senior to Staff. Check: ask whether the range covers one level or several.
4. Senior stories in a Staff loop. One interview-prep course warns: "If all of your stories make you sound like a senior-level engineer, you won't get a staff-level offer." The usual cause is stories about execution inside a team that never reach decision rights or cross-team scope. Check: tag every prepared story with its axis and make sure the Staff-only axes are covered.
5. A weak single round. In one reported LinkedIn case, a single missed edge case cost the candidate Staff. Check: ask the recruiter what happens to the level if feedback is split, and prepare your weakest-axis round hardest.
6. Assuming the recruiter can confirm. At some employers nobody can confirm the level until the loop ends, and a confident-sounding recruiter answer may not bind anyone. Check: ask whether the level is decided before, during, or after the loop, and record the answer.
7. Mismatched level names. Translation tables disagree, most visibly on whether Meta Senior is E5 or E6. A wrong map means you ask for the wrong level with confidence. Check: cross-reference two sources for every employer, and when they conflict, say the scope and ask which internal level matches it.

A related policy risk: Meta reportedly withdrew a senior offer under "a new policy requiring at least six years of experience for senior positions". Hard minimums like that can override scope evidence, so if your years are near a threshold, ask about it in Ask 1 instead of finding out at the offer.

Before the first application goes out, and keeping it current

The level claim is finished when the verdict, the translations, the materials, and the ask all say the same level. Run the checklist below once before the first application, and again whenever your target role family changes.

To keep the claim current:

• Re-date the evidence sheet after any major project. The verdict is only as recent as its newest row.
• Re-check level maps per employer, not once. Maps change and disagree, so check the published ladder or a free level-comparison site when a new employer enters the pipeline.
• Re-score when the role family changes. Moving from IC to management, or between specialties, changes which axes carry weight.
• Log every recruiter's level answer. After a few calls you will see whether employers keep placing you one rung below your verdict. If they do, the problem is in your top block or your weakest axis, not in the market.

When you start applying, Refolk scores how well you fit each posting. Read that score next to the posting's corrected level: a strong fit at a level below your verdict means your evidence sheet and that employer's ladder disagree, and you should settle that before the screen.