As outlined in the general methodology, we use information from professional networks as the basis for Workforce Intelligence.
We map the raw job titles from members' profiles to standardized categories, which we call job groups and job subgroups.
Taxonomy of job (sub)groups
We use the following taxonomy for job groups and subgroups:
Job group | Job subgroup | Example roles covered within the group |
Manufacturing & Production | Production | Line workers, operators, plant managers |
| Engineering & Maintenance | Technicians, mechanics, maintenance staff, facility staff, industrial engineers |
| Quality control | Quality control, health & safety roles, hygiene, food safety |
Sourcing, supply chain & logistics | Supply chain management | Demand planning, supply planning, materials planning, inventory management, supply chain management |
| Logistics & distribution | Warehousing, logistics, transport, storage, fleet, import/export |
| Sourcing & procurement | Procurement, purchasing, buyers, sourcing, contract management |
Sales | Account management | (key) account managers, category development, client managers |
| Direct sales & business development | Business development, sales, country/segment sales, country managers, regional managers |
| Sales operations | Sales planning, commercial excellence, aftersales, sales enablement |
Marketing | Digital marketing | Digital marketing, social media marketing, content managers, ecommerce |
| Sales marketing | Trade marketing, customer marketing |
| Strategic marketing | Brand management, marketing management |
| PR & Communications | Public relations, events, corporate communication |
Research & Development | Product development & innovation | Product management, product development, innovation |
| Basic and applied research | Laboratory research, new technologies, food science, trials & testing |
Corporate and support staff | Corporate strategy & development | Strategy, M&A, corporate development, market intelligence |
| Finance & accounting | Accounting, controlling, bookkeeping, revenue management, finance |
| Human resources | Recruitment, payrolling, training & development, talent acquisition, HR |
| Legal & compliance | Legal counsel, risk, compliance, regulatory affairs, intellectual property |
| Information technology | Developers, data science, IT administration, cybersecurity, ERP systems |
| Support staff | Administrative staff, customer support, secretaries, corporate staff without a direct functional mapping (e.g. 'Manager', 'Analyst', 'Assistant') |
Lastly, there are two remaining groups:
Irrelevant positions: We filter out a limited number of profiles because we know for a fact they are not relevant to the workforce. This mostly concerns people that state they are retired, but still have an active position on the associated company's professional network
Unmapped positions: This contains the profiles that we cannot map to a particular job (sub)group. Read more on how we carry out the mappings below.
Mappings
We use keyword mappings to map raw job titles to the job (sub)groups described above. We make use of several types of mappings:
Exact matches: A one on one match of the raw job title, e.g. mapping 'Key account manager' to the job group 'Sales' and job subgroup 'Account Management'
Start of word matches: Raw job titles that start with a particular keyword, such as 'R&D', meaning we map both 'R&D engineer' and 'R&D manager' to the job group 'Research & Development', job subgroup 'Basic and applied research'
Boolean matches: Keywords that occur within the raw job title, such as mappings using the keyword 'Human Resources', meaning we map the raw job titles 'Human Resources Manager - South Africa', 'Human Resources Analyst' and 'Senior Vice President & Head of Human Resources' to job group 'Corporate and support staff', job subgroup 'Human Resources'
Given the sheer variety of raw job titles and the fact that it is based on user-inputted data (i.e. meaning it can contain spelling errors, acronyms that aren't widely used, or even gibberish), not all raw job titles can be mapped. The ones that we can't map based on the steps described above end up in the 'Unmapped' category.
Note that we continuously review and add to our mappings, aiming to provide a picture that is as complete as possible. Feel free to make additional suggestions that we can add to our data cleaning process.