Talent mapping: what it is, and how to map a competitor's org from its job postings

Recruiters use “talent mapping” for two things that share a name and little else. One is internal: a grid of who in the company could step into which role. The other is external: a picture of how a competitor is organised, what it hires for, where, and at what level, built before a requisition exists. The second one can be done from public data, and this guide shows how, with the requests and the fields that come back.

What is talent mapping?

Talent mapping is the practice of building a picture of where the people you need sit, inside your own organisation and at competitors, before there is a role to fill. It covers which teams a company has, how they are structured, what they pay, where they are located and what skills they hire for, so recruiting can be proactive rather than reactive.

Internal talent mapping draws on performance and skills data you already hold and produces succession plans. External talent mapping draws on whatever a competitor makes public, and the richest public source by far is its job postings. A posting states the function, the seniority, the location, often the tools and sometimes the salary, and a company’s full set of postings over a few months is a draft of its org chart with the growth areas highlighted.

How to map a competitor’s org from its job postings

The example uses the company from the JobsPipe API documentation, Acme Robotics, so the requests run unchanged against your own key. Swap in any company name or domain. The whole sequence is five requests to POST /v1/jobs/search plus one to the company endpoint.

Step 1: pull every open and recently closed role

Ask for everything, not only what is open today. Closed roles show what the company has already filled, which is the part of the org chart that exists rather than the part being built. status accepts active, closed or any, and the partial-match filter catches subsidiary and regional naming variants.

curl -X POST https://api.jobspipe.dev/v1/jobs/search \
  -H "Authorization: Bearer $JOBSPIPE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "company_name_partial_match_or": ["Acme Robotics"],
    "status": "any",
    "limit": 100
  }'

Each record returns the fields the map is built from:

• job_title and normalized_title, the posted title and the cleaned one used for grouping.
• job_function, seniority and is_manager, which place the role in a function, at a level from entry to exec, and mark whether it manages people.
• location, country_code, cities and remote, which place it on the map.
• technology_slugs and keyword_slugs, the tools and skills the posting names.
• min_annual_salary_usd and max_annual_salary_usd where the posting states pay, and the estimated benchmark fields where it does not.
• date_posted, discovered_at, status, closed_at and closed_reason, the lifecycle.
• company_object with employee_count, headquarters location and founding year where known.

The company record itself is one more call. It accepts a domain or a name and returns headcount, headquarters, description and the LinkedIn page, which is the header row of the map.

curl https://api.jobspipe.dev/v1/companies/acme.com \
  -H "Authorization: Bearer $JOBSPIPE_API_KEY"

Step 2: build the grid

Group the records by job_function across the columns and seniority down the rows. The cells with open roles are where the company is growing; the cells with only closed roles are teams that exist and are stable; the empty cells are functions the company either does not have or fills without posting. Count is_manager per function to estimate team count, since a company posting for two engineering managers is building at least two teams.

Step 3: place the teams

Group the same records by cities and country_code. A function that posts only in one city is a hub; a function that posts as remote everywhere is distributed. A first posting in a new country is an expansion signal in its own right, and the trigger events post covers how early postings show it.

Step 4: read the stack

Union the technology_slugs across engineering postings and you have the stack the company hires for, which is close to the stack it runs. To check a specific technology across several competitors at once, flip the query around and filter on the skill.

curl -X POST https://api.jobspipe.dev/v1/jobs/search \
  -H "Authorization: Bearer $JOBSPIPE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "company_name_or": ["Acme Robotics", "Globex", "Initech"],
    "skills_or": ["kubernetes", "rust"],
    "status": "any",
    "limit": 100
  }'

Step 5: read the base rates

A map of one company needs a scale. Across companies with an active posting on the ten boards JobsPipe collects, 41.5% have exactly one role open, 10.3% are hiring in three or more functions, and only 2.2% have three or more engineering roles open at once. A competitor with five open engineering roles is not merely busy; it is in a small minority of hiring companies by that measure.

Step 6: keep the map current

Rerun the first request on a schedule with discovered_at_gte set to your last run and you get only the postings JobsPipe has seen since, which is the change to the map rather than the map again. Or save the filter in Signals and have new postings from the mapped companies arrive by email, Slack or webhook.

Talent mapping vs talent intelligence

Talent mapping is an activity with an output: a map of one organisation or a handful of them, built for a specific hiring plan. Talent intelligence is the category of data and software the activity draws on, from profile databases to postings feeds to the platforms that package both. Every talent mapping exercise uses talent intelligence; most talent intelligence is used for things other than mapping, such as sourcing, compensation and workforce planning. The talent intelligence platforms comparison covers the software side, and the glossary entry the definition.

What postings cannot tell you

A map built from postings is a map of change, not of state. It shows the roles a company opened and closed in the collected window, not the people already in seat, so a stable team that has not hired in months is invisible. Some companies fill roles through agencies, and the employer_type field separates agency postings from the employer’s own. And coverage is the ten boards JobsPipe collects, so a company that posts only on its own careers page outside those systems is under-mapped. Read the postings map beside a profile source for headcount, and treat it as the leading edge rather than the census.