An Occupation Is Not a Career Path, and a Title Is Not Either

Somebody exports your internal mobility report.

There are 1,622,400 Software Developers in one column.

There are 209,400 Software Quality Assurance Analysts and Testers in another.

And between them, a cheerful arrow marked career_path = next.

That arrow is doing a great deal of work.

It has decided that a statistical occupation, a local title, a person’s probable capability and a future vacancy are all the same kind of thing.

They are not.

An occupation describes work. A career path describes possible movement. A title does neither reliably enough to be the bridge.

This is not a complaint about orderly data.

Orderly data is lovely.

It is why you can ask how many people do a kind of work, what that work involves, or which of your roles need a particular learning programme, without convening the archaeology team.

The problem starts when an orderly category turns into a personal destiny.

Then Software Quality Assurance Analyst stops describing validation work and becomes a reason to hide a worker from a development opportunity.

Senior becomes evidence of scope.

Manager becomes evidence of people leadership.

An occupation code becomes a prophecy.

Your HR system thinks an arrow is a fact.

An occupation has a proper job

Occupations are population-level descriptions.

They let labour-market systems compare work that would otherwise arrive as a small pile of locally invented nouns.

The O*NET Content Model, sponsored by the Department of Labor, splits its descriptors into worker-oriented and job-oriented domains.

Worker characteristics, requirements and experience on one side.

Occupational requirements, workforce characteristics and occupation-specific information on the other.

That is a serious attempt to make work legible.

It lets a system compare task content, cognitive demand, physical conditions, interpersonal conditions and consequence of error, without pretending that everyone who ever held a title is interchangeable.

It also keeps different kinds of evidence apart.

Incumbents report tasks, work activities and context.

Trained occupational analysts rate abilities and skills against behavioural anchors.

Which is why occupational data is useful.

It gives your workforce planner something more defensible than a title string and somebody’s strong feelings about what “senior” means in Product.

It does not tell that planner what any one of your people should do next.

The boundary is a clue, not a fence

Look at three adjacent software occupations.

Software Developers are 15-1252.

Computer Programmers are 15-1251.

Software Quality Assurance Analysts and Testers are 15-1253.

Those are not redundant labels.

Developers translate requirements into system models, database schemas, microservices and production software.

Programmers implement defined designs and maintain codebases.

QA analysts build verification protocols, automate tests, trace defects and validate performance.

Architecture.

Implementation.

Validation.

The classification is saying something real about the centre of each occupation.

It is not saying that a QA analyst has never designed software.

Or that a programmer cannot reason about systems.

Or that a developer has never written a test after the release train caught fire.

In fact the boundary tells your mobility system what to examine.

Does the person have evidence of architectural synthesis?

Have they built or reviewed production code?

Can they meet the destination role’s reliability and design requirements?

Is the gap a missing capability, a missing chance to demonstrate it, or just a missing phrase in skills_export_AUGUST_final.csv?

That is a career-move question.

An occupation code cannot answer it, because an occupation is an aggregate description of work, and a move is a judgment about a person, a destination and a moment of opportunity.

A career path is not a column

Most career-path diagrams are very fond of columns.

Associate Engineer becomes Professional Engineer becomes Senior Engineer becomes Staff, Principal and Distinguished Fellow.

IC1 through IC6.

Alongside it, M1 through M6 climbs from Team Lead to Vice President.

This can be excellent design.

It lets you reward technical scope without requiring every capable engineer to become a manager, which has saved many meetings from an avoidable outcome.

But your diagram describes one employer’s progression architecture.

It does not describe the whole possible career of everyone assigned to an occupation.

Still less a universal order of titles.

An employee may enter through a different port.

Build adjacent capability in a temporary assignment.

Move sideways into a new context.

Leave and return with more experience.

Or discover that the next vertical rung was removed during a reorganisation called Operating_Model_Simplification_2.0.

All of those are careers.

Only some of them fit in a column.

The ladder is worth keeping, which is the difficulty

Ladders give people a visible account of level, expected scope and evaluation.

Calibrated impact rubrics and equivalent technical and management bands make progression less dependent on one manager’s memory and personal vocabulary.

Occupational classifications do something equally important.

They orient labour-market analysis, make task patterns comparable, and help a learning team see where an occupation commonly calls for particular knowledge or conditions.

A nurse, a heavy truck driver and a financial analyst should not collapse into the same talent-marketplace record because all three clicked “interested in advancement.”

Keep the classifications.

Keep the ladders.

Keep your local titles too, where a local title helps people find work and understand the place.

The failure is not the artefact.

The failure is making an artefact answer a question it was never built to answer.

The path needs ports and lateral exits

An internal talent marketplace is one way to make movement visible without inventing a promotion.

It matches people to short-term projects, cross-functional gigs and lateral developmental assignments, using skills information.

Enterprise versions allocate 10 to 20 percent of contracted time to rotational work lasting three to six months, without changing the employee’s primary reporting manager.

That is not a detour from a career path.

It is often the evidence that makes the next move credible.

Large multinational employers running these marketplaces report internal vacancy fill rates up by about a quarter, and cross-skilling across adjacent technology stacks accelerating by around 40 percent.

Those figures are not a promise that a marketplace is magic.

They describe why a project assignment is a different object from a title sequence.

It creates the opportunity to do new work before a new label exists.

Entry ports make the same point at national scale.

Dual vocational education and training in Germany, Switzerland and Austria absorbs roughly half of secondary school leavers into standardised occupational tracks.

Three-year contracts.

One or two days a week in vocational school, three or four in accredited workplaces.

A training allowance at a fraction of a qualified journeyworker’s wage.

That entry route is structured.

It is not identical to a local corporate title history.

Which matters to your mobility model.

It has to record where somebody demonstrated the capability, not merely whether they arrived by the route on last year’s succession slide.

The missing rung is still missing

Career paths also have to admit their structural limits.

When a reorganisation removes 30 to 50 percent of middle-management positions, span of control can move from one-to-five to one-to-fifteen or worse.

The middle rungs do not stay developmental because the talent team kept their labels in a framework.

They are gone.

In forced-ranking systems, cross-team knowledge sharing falls by a fifth to a third once the promotion prize spread becomes steep enough.

A path that rewards workers for making their neighbours less successful is not a path.

It is a procurement process for awkward calibration meetings.

The scale limit is plainer still.

Small and medium enterprises are more than 98 percent of employer businesses.

An employer with fewer than a hundred people typically has a low single-digit number of senior openings across five years.

A very good employee can develop substantially without a corresponding title move, because the position does not exist.

The worker has not failed to progress.

The organisation has run out of boxes.

Which is why a credible path states your constraints.

Available roles.

Entry requirements.

Possible lateral contexts.

Evidence required at the destination.

It never guarantees that a title sequence will materialise on schedule.

AI will call the old sequence natural

This used to be an awkward problem for whoever ran the spreadsheet.

Now it is an input problem for systems that recommend learning, match people to projects, identify successors and suggest careers.

The model sees tidy historical sequences.

QA Analyst becomes Senior QA Analyst becomes QA Manager.

It sees far fewer cases where somebody moved through testing, built production experience in a short assignment, and became a developer or a platform specialist.

History is a record of what your organisation made possible.

It is not a catalogue of everything workers could have done.

A model that treats the observed sequence as the natural one will make an old opportunity structure look like capability.

It will call the lateral move unusual, because you made it rare.

Then it will recommend more of the path it already knows.

So your recommendation needs a different contract.

Store the source occupation and its version.

Store the local role and display title separately.

Store task, skill, work-activity and context evidence separately.

Record the destination capability, the expected level, the relationship type, and the evidence still required.

Mark whether the move is lateral, diagonal or vertical, and whether a real opportunity exists.

Then the model can propose a move without pretending it has discovered destiny in a string column.

Which is less cinematic than a career-path arrow.

It is also less likely to tell a capable person that the next job is whichever noun happens to sit directly below the current one.