How Your Job Title Gets Normalised

Finding your job title in the document is one step; deciding what that title means is a separate one. Many systems map the string they extracted onto an entry in a standard title taxonomy, so that a search for one title can also return people whose employers called it something else. When your title is unusual, invented, or internal, that mapping is where it gets lost — and the remedy is to state the plain-language equivalent alongside the official one.

This post picks up where job titles, dates, and the fields parsers look for stops. That one is about extraction: getting the right string out of the page. This one is about what happens to the string afterwards.

Why normalisation exists

Employers name roles inconsistently. The same work is called Software Engineer II at one company, Developer at another, Member of Technical Staff at a third, and Ingénieur logiciel at a fourth. A recruiter searching for one of those strings would miss the other three.

So systems maintain a canonical list — a taxonomy of standard occupation titles, sometimes with levels attached — and try to place each extracted title onto it. The stored record then carries two things: the raw string as you wrote it, and a normalised code or label. Searches and filters may run against either.

The mapping itself is usually string similarity plus a set of rules about seniority words, occasionally with a model doing the matching. Like everything in parsing, the quality varies by vendor, version, and how much the employer has configured it.

What normalisation gets right

Give it a conventional title and it works well. Sr. Financial Analyst maps to a senior analyst entry. RN maps to registered nurse. AP Clerk maps to accounts payable. Abbreviations, seniority prefixes, roman numerals, and department suffixes are the ordinary cases and are handled about as well as you would hope.

It is also the mechanism that helps you when a posting uses a different word than your employer did. If you were a Programme Manager and the posting says Project Manager, a decent mapping puts you in range without you doing anything.

Where it fails

Invented internal titles. Companies love these. Chapter Lead, Growth Ninja, Chief of Staff to the CTO, Level 4 Associate, Customer Happiness Architect. There is nothing on the canonical list that resembles them, so the mapping either fails or lands somewhere strange. Architect in a title can pull a customer-support role toward software architecture; Ninja maps to nothing at all.

Titles that describe the company’s structure rather than the work. Band 6 Officer, Grade 12 Specialist, Associate, Group Function. Common in government, banking, and large institutions. Accurate and completely opaque.

Titles where the seniority word is doing unusual work. Principal means a very senior individual contributor in software and the head of a school in education. Fellow can be a postdoctoral researcher or a distinguished engineer. Director is a board member in some countries and a middle manager in others. Partner means owner in a law firm and a shop-floor employee in some retailers.

Very short titles. A one-word title like Consultant or Manager gives the mapping almost nothing to work with, and without a domain it can land in an unrelated occupation.

The fix: a plain-language gloss

You cannot change your title. It is a factual claim about what an employer called you, and altering it on a resume is misrepresentation — the kind that employment verification is specifically designed to catch.

What you can do is put the recognisable description next to it, in the same line, clearly labelled as a description rather than a title:

Chapter Lead (engineering manager, 9 reports) — Northwind Ltd.

Band 6 Officer (data analyst, transport policy) — County Transport Authority

Chief of Staff to the CTO (technical programme manager) — Calder Group

Every one of those is true. The official title stands first and unmodified; the parenthetical explains it. A human reader gets clarity immediately. A machine gets a second string to match against, and a mapping that would have failed on Band 6 Officer may succeed on data analyst.

The wording matters slightly. (engineering manager, 9 reports) is a description of scope. (Senior Engineering Manager) in title case starts to look like a second, grander title, which is where an honest gloss shades into inflation. Keep it lower case and factual.

Levels, and why you should not invent them

Seniority is the part of normalisation people are most tempted to game, because a “senior” prefix demonstrably shifts where you land.

Don’t add one you weren’t given. But do include the level markers you were given, in the form your employer used, because they carry real information:

  • Software Engineer II is more useful than Software Engineer if that is what your offer letter said.
  • Analyst → Senior Analyst (promoted Mar 2022) documents progression and gives the record two titles honestly.

If your title genuinely understated your work — a common situation in small companies, where everyone is a Coordinator — the honest tool is scope, not a title change. Coordinator (managed the 4-person warehouse team) says what happened.

Multiple roles at one employer

Stacked entries under a single employer heading parse well and normalise well, because each entry gets its own title string:

Northwind Ltd. Lead Analyst — Mar 2022 – present Senior Analyst — Jan 2019 – Mar 2022

Compressing the two into Senior/Lead Analyst, 2019 – present gives the mapping a compound string that matches neither cleanly, and loses the promotion date.

What normalisation does not do

It does not read your bullet points to decide what your job was. A mapping works from the title string and sometimes from nearby context; the substance of the role lives in text that gets stored and searched separately.

It also does not resolve industry. Analyst in a hospital and Analyst at a hedge fund may normalise to the same entry. If your domain is a qualification for the role you want, name it in the entry — Analyst, clinical operations — rather than assuming the employer name conveys it. Nobody outside your sector recognises your former employer.

Check it

Look at each title line in the plain-text version of your resume — the copy-paste output described in how to test your resume like a parser would — and read it as a stranger would. Ask: if I knew nothing about this company, would I know what this person did?

If the answer is no, that title needs a gloss. If the answer is yes, leave it alone; normalisation will handle the ordinary case without your help.