Career assessment for PhDs and postdocs: what each option does, and what it does not

You are searching for a career assessment because a date is approaching. A thesis submission, a fellowship that ends, a contract with a known last day.

The question behind that search is rarely "who am I". It is narrower: which of these options fits my case, and is it worth the money?

This article answers that with the options side by side. We sell one of them, so we name our sources, and we also name the cases where paying a human is the better decision.

Two things get confused here. A career assessment measures your dispositions, interests, values and skills. A career decision is what you do with that measurement. Most PhDs already have more material than they think. What is missing is the wording an industry reader can use.

1. What anglophone PhDs actually reach for

Four options come up in almost every conversation. Each does something real, and each has a boundary worth knowing before you spend.

The university careers service. Free, already paid for by your institution, and staffed by people who talk to employers. A good one will read your resume, run a mock interview, and open doors through alumni. What it usually cannot do is go deep on a research profile. Its frameworks were built for undergraduates and taught master's students. You may be the only postdoc that adviser sees this term.

Paid PhD-to-industry programmes. These are the paid, structured option, usually several sessions over several weeks. They do one thing that no questionnaire does: they put a human in front of you who makes you say things out loud. A blocked decision about leaving research is a conversation, not a scoring exercise. They also hold you to a schedule. Prices are set per provider and we have no source for a market average, so we publish none. Ask for the total, the number of sessions, and what you keep at the end.

Free online typology quizzes. They take a few minutes and they are entertaining. The four-letter output is the problem. It sorts you into a category, has no researcher-specific vocabulary, and rests on instruments you often cannot trace to a publication. A scientist cannot act on a result they cannot check.

ChatGPT and similar assistants. Genuinely useful, and underrated for this. Paste a job description and your thesis abstract, and you get a first pass at translation in seconds. It is also patient at rewriting. What it does not give you is measurement. It has no scored profile of you, only the story you told it, so it will agree with your framing and reflect it back. Ask it for your top values and it produces plausible values, not yours.

The LabShift assessment sits in a fifth place. It is free, takes ~10 min, and returns scored results from four published instruments. It has no human interlocutor and no sessions. It is a framing step, not an accompaniment.

2. The four questions a PhD has to settle

Whichever option you choose, a useful output answers four questions. Check that a provider handles them before you pay.

1. In which working environment am I durably functional? Established organisation or startup, deep work or coordination, long project cycles or short ones.
2. What are my drivers, and which ones did academia satisfy without my naming them? Autonomy, security and contribution do not carry the same weight from one person to the next.
3. Which of my skills are transferable, and under what name do employers search for them? "Co-supervised two master's students" and "led a project team" sometimes describe the same fact.
4. Which role titles match that combination? Not fifty leads. A short, ranked list, with the reasons attached.

Four answers are worth more than a thirty-page synthesis document. Those answers are what you carry into an interview.

Frame your profile before you pay for anything. The LabShift assessment takes ~10 min, needs no account, and shows your results immediately.

Take the free assessment

You get an OCEAN radar, a RIASEC hexagon, your top 3 values and 5 recommended roles with a compatibility score. No hiring promise.

3. The instruments, and why we name them

A researcher does not act on a typology they cannot verify. So the assessment uses only published, citable instruments. 48 items in total.

• Big Five — IPIP-20 (Goldberg, L. R., 1992). Five continuous dimensions: openness, conscientiousness, extraversion, agreeableness, emotional stability. Scored dimensions, not categories. The five-factor model is the standard of contemporary personality psychology.
• RIASEC Holland — 18 items (Holland, J. L., 1973). Six poles of vocational interest: realistic, investigative, artistic, social, enterprising, conventional. Doctoral profiles often score very high on investigative. The useful question becomes which pole comes second, and which roles combine the two.
• Schwartz Values Survey (Schwartz, S. H., 1992). Ten universal values. Useful for anticipating where a role will actually create friction, which is rarely on the technical side.
• Vitae RDF (Vitae, 2010), the Researcher Development Framework. It is the framework that describes researcher competences explicitly, which no general careers framework does. It lets you name your transferable skills without inventing them.

Naming the instrument is not academic decoration. It is what makes the result auditable. You can read the original publications before you answer a single item. You can see which dimension a score belongs to, and what it does not claim to measure. A four-letter label offers none of that, and neither does a paragraph a model wrote about you.

None of the four is proprietary or opaque. What they do not do: predict a hire, guarantee a salary, or take the decision for you. They frame a decision that stays yours.

4. An honest comparison

Read the "human interlocutor" row without minimising it. A person in the room makes you say things a questionnaire never will. If that is what you need, a paid programme provides it and we do not.

One aside for context, because readers sometimes arrive from the French market. France has a state-funded equivalent, the bilan de compétences, capped at 24 hours and €1,600 of public funding (Code du travail art. L. 6313-4 / R. 6313-4; moncompteformation.gouv.fr, 2026). There is no anglophone equivalent of that entitlement.

5. Three cases where paying a human is the right choice

We would be dishonest to claim otherwise.

• You need a third party who makes you talk. If the block is the decision to leave research, that is resolved in conversation, not in a questionnaire.
• You need accountability against a date. Some people run applications well alone. Others need a scheduled session and someone expecting a draft.
• Your target is already outside research and unrelated to your thesis. The advantage of a researcher-specific framework disappears, and general career support is a better fit.

In the other cases, especially when the deadline is short and the difficulty is wording your experience, start with fast framing. You will then know whether you need several paid sessions, and you will arrive at the first one with material instead of a blank page.

6. Decide this week

An order that works, in this precise order:

1. Book the free option first. Email your university careers service and ask for the next appointment. It costs nothing and the queue is often long.
2. Frame your profile so you know what you will discuss. The LabShift assessment is free and takes ~10 min.
3. Write three role titles you will target, each with two elements of your doctoral work that support it.
4. Only then decide whether paid sessions add anything to what you have already written.

Step 3 is the one most people skip, and it is the one that changes an interview.

Take the free assessment

The translation engine goes further, for readers who want the translation written out. You paste a real job ad, and it reads your repertoire of doctoral achievements together with your psychometric profile to produce a cover letter shaped as an offer of a solution, an outreach message, a prepared answer to "tell me about yourself", and questions to ask in the interview. Every claim about your track record carries the identifier of the achievement that supports it, and the engine also names what you cannot claim. It refuses to generate, and explains why, when your profile does not answer the ad. The full assessment report is included and is what calibrates the engine.

00 — Context

Start here: the only two details asked

Free · ~10 min · no account, no email required to see the results.

They shape the recommended roles. The rest of the assessment is 48 items across four published instruments — Big Five (IPIP-20), RIASEC, Schwartz values, Vitae RDF.

Your PhD field-- Select --Biology / Life sciencesChemistry / BiochemistryPhysics / MechanicsComputer science / AI / DataMaterials scienceMedicine / PharmacyEnvironmental scienceSocial sciences / HumanitiesOtherWhere are you now?-- Select --1st or 2nd year of PhDFinal PhD year or dissertation writingPost-defense / recent PhD (< 1 year)Current postdocActively transitioning

Continue to step 01 →Select both fields to continue.

Or open the assessment directly  →

Frequently asked questions

Is there a career assessment designed for PhDs and postdocs? Yes, and the distinguishing feature to look for is the competence framework. General assessments describe functions, management and commercial targets. The Vitae RDF describes researcher competences, which is why the LabShift assessment is organised on it.

How much does PhD career support cost? It depends entirely on the provider, and we have no source for a market average, so we give no figure. Ask any provider for the total price, the number of sessions and the deliverable in writing before you sign.

Can my university careers service help a postdoc? Often yes, and it is free, so book it. Expect general frameworks rather than research-specific ones, and expect an adviser whose cohort is mostly students.

Are free personality quizzes useless? Not useless, but not actionable. A four-letter category cannot be traced to a scored dimension or a publication, and it says nothing about which roles fit a research profile.

Can I just use ChatGPT for this? For rewriting and for drafting, yes, and it does that well. For measurement, no. It has only the account of yourself that you gave it, so it cannot tell you your top 3 values or a compatibility score.

Does the LabShift assessment replace paid sessions with a human? No. It is not an accompaniment and there is no interlocutor. It gives psychometric framing and 5 recommended roles in ~10 min, free, which is useful before you decide to spend.

Do I need an account? No. No account and no email address are requested before the results.