A job title is a compression algorithm. It takes a complex bundle of responsibilities, judgment, technical depth and organisational context, and reduces it to two or three words that fit on a business card. Like any lossy compression, it discards whatever the format cannot carry — which, in practice, is most of what determines whether someone will succeed in a different seat.
This has always been true. What has changed is the rate at which the discarded information matters. The World Economic Forum's Future of Jobs Report 2025 puts it starkly: employers expect 39% of workers' core skills to change by 2030. If two-fifths of the substance behind a title turns over inside five years, then the title is not merely lossy. It is describing a role that no longer exists in the form the label implies.
WEF Future of Jobs Report 2025
What breaks when you search on titles
Two failures follow, and they pull in opposite directions.
The first is false positives. A Head of Supply Chain at a ₹200 crore regional distributor and a Head of Supply Chain at a national FMCG business share a title and almost nothing else — not scale, not systems maturity, not the political complexity of the role, not the decisions the job actually requires. A title-matched long list treats them as interchangeable. The screening burden this creates is then handed to the client, disguised as choice.
The second, and more expensive, failure is false negatives. The person best equipped to run your supply chain through its next phase may currently hold a title in operations, in manufacturing, or in a consulting practice serving the sector. Title-based search never surfaces them, and their absence is invisible — you cannot review a candidate the process never found.
False positives waste the client's time, which is visible and irritating. False negatives cost the client the best hire, which is invisible and far more expensive.
Decomposing a mandate into vectors
The alternative is to stop treating the role as a label and start treating it as a set of requirements that can be held independently. In practice, decomposing a mandate means separating at least four things that a title fuses together:
- Technical competencies — the specific, teachable capabilities the seat requires, at a stated level of depth rather than a binary present/absent.
- Scale and complexity context — headcount, revenue, geographic spread, matrix depth. Someone who has run ₹50 crore has not run ₹500 crore, whatever the title said.
- Situational archetype — a turnaround, a scale-up, a stabilisation, a market entry. These demand genuinely different temperaments, and experience in one predicts little about the others.
- Organisational behaviours — how someone builds alliances, handles a board, inherits a team they did not pick, and behaves when the plan stops working.
Once a mandate is expressed this way, the search space changes shape. You are no longer looking for people who held a title. You are looking for people who have demonstrably carried each requirement, in whatever container their career happened to put it in.
Where the machine helps, and where it stops
This decomposition is where data earns its place. Matching structured candidate information against competency vectors, rather than against job titles, genuinely widens the addressable pool — and it does so in a direction that is hard to replicate by hand at scale. LinkedIn's own research found that 60% of recruiters using AI report surfacing 'hidden gem' candidates they would have missed in manual search, precisely because the system can look at skills rather than labels.
But it is worth being exact about what has happened at that point. The pool has widened and been ranked. Nothing has been decided. The attributes that separate a strong shortlist from a hire — whether someone will own an outcome when it stops being fun, how they read a room they have never sat in, whether they can hold a position with a board that outranks them — do not appear in structured data. They surface in conversation, and in what former colleagues say when asked properly.
A note on internal application
The same decomposition is what makes succession planning tractable. Organisations that map internal capability at the skill level rather than the headcount level can model supply against demand before a gap becomes a hiring emergency. Most cannot, because different functions define the same capability differently and no shared vocabulary exists. Building one is unglamorous work with a long payback — which is exactly why it tends not to get done until a critical seat empties unexpectedly.