Talent Mapping in 2027

Talent mapping is the practice of researching and organising the people who could fill your future roles before those roles open. A talent map lists target companies, the people in the relevant functions at each of them, how those organisations are structured, and where people tend to move next. The goal is to start a hiring decision with a shortlist instead of a blank search.

Most guides present talent mapping as a template: a spreadsheet with columns for company, name, title and notes. The template is fine as a view. The problem is that a map built by hand starts to go stale the week it is finished. For 2027 the better approach is to build the map from queries that can be re-run, so refreshing it is a job for a script and not a project for a researcher.

Hiring teams are working harder for each hire. A survey by Censuswide of 6,554 HR professionals, fielded in November 2025, found that 66% say it has become harder to find qualified talent over the last year, 42% are under pressure to fill roles faster, and 39% are asked to uncover candidates other teams would miss. Those pressures all favour teams that already know the market when a requisition is approved.

A map also changes who you talk to. Job boards reach people who are looking. A map covers people who are employed and not searching, which makes first contact a different kind of conversation and makes the quality of your research visible in the first message.

Start with the organisations where the people you want already work. A company search returns the set that matches your criteria, and each result carries headcount, industry and headquarters location, which lets you drop companies that are too small or in the wrong region before spending time on them.

For each company, filter people by title, seniority and location instead of paging through a full roster. The filters combine, so one request can return senior engineers at a named company with at least a year and a half in their current role. Results are previews. Fetch the full profile only for people who make the shortlist.

A map that only lists current employees misses a second pool. People who used to work at a company know its tools and vocabulary, and a rival's former employees already know how that competitor operates. The past-employees endpoint returns them with their earlier roles and dates, and the alumni endpoint returns graduates of a named institution by school identifier.

A static list says where someone works. Timing says whether the approach is likely to land. The job-changes endpoint returns people who recently changed employers, optionally limited to those who previously worked at one company. Someone who started a new job last month is usually a weak target for a recruiter, while an agency prospecting clients can read the same event the other way, since the employer that person left now has a gap. The date reflects when the dataset last observed the person at the employer, so allow for a lag of days or weeks.

Re-run the searches on a schedule and store the results with the date, so you can see who arrived in or left your target set since the last run. For a short list of priority people or companies, a monitor sends a signed webhook when a tracked field changes, such as person.job_changed, and removes the need to poll them. Each record carries a fetched_at timestamp, and a live lookup at $0.001 per call is available to re-check a shortlist before anyone writes a message.

The output is a set of tables you can load into an applicant tracking system or a warehouse.

The API returns professional profile and company data and no email addresses or phone numbers. A talent map built here is research and prioritisation, and outreach goes through channels you already hold: your applicant tracking system, a licensed contact provider or a warm introduction.

You are the controller for how you use the data. That means a lawful basis for contacting people, usually legitimate interest, a notice to them, and deletion when someone objects. A lookup that returns HTTP 451 means the person has asked to be withdrawn, so delete the record from your lists instead of retrying. If you use automated filtering to rank candidates, check whether the EU AI Act treats that use as high-risk recruitment, where obligations include human oversight.

Volume hiring for roles filled from applicants does not need a map, since candidates come to you. Public profile data also under-represents people whose work rarely produces a profile, so it suits knowledge roles better than many skilled trades. And a map is research, not a pipeline: nothing here tracks consent, stages or outreach history, so keep that in your applicant tracking system.