Skills-Based HRIS: Moving Beyond Job Titles in Pakistan

A skills-based HRIS records what each person can do, at what level and with what evidence, next to their job title. That record lets you match people to open roles, build learning paths and name successors by capability instead of seniority. Job titles still exist. They stop being the only thing you know about your people.

For a Pakistani company of 50 to 300 employees this is a modest project, not a transformation: a list of 30 to 60 skills, four proficiency levels and a yearly validation. The payoff is that you stop hiring outside for roles your own people could fill.

Because the work moves faster than the title. Deloitte's 2022 skills-based organisation survey of 1,021 workers and 225 business and HR executives in 10 countries found that only 42% of executives said job descriptions accurately characterise the work their organisation needs done. Just 10% of HR leaders strongly agreed they could anticipate the skills they will need over the next three years (Deloitte, 2022).

Localise that. In a growing Pakistani company, one person titled Accounts Officer may also run vendor payments, prepare the tax file and train new joiners. The title shows none of it. When that person resigns, you learn about the skills only after they leave.

The Deloitte survey also found organisations that embed a skills-based approach were 63% more likely to achieve results across 11 business and workforce outcomes, though fewer than 1 in 5 had adopted the approach to a significant extent. The survey is from 2022 and mostly covers larger organisations, so read it as direction, not a promise for a 100-person firm.

A taxonomy is a controlled list of skills with a shared definition of each level. Keep it small enough that a manager can rate someone in five minutes. My suggestion for a first version: 30 to 60 skills across four groups, which is my judgement and not a standard.

Use four proficiency levels, each defined by what the person does, not how long they have done it.

AI helps with the tedious parts: suggesting skills from a CV, a job description or a project history, spotting that two skill names mean the same thing and proposing role matches. It should suggest. The employee confirms and the manager validates.

Self-rated skills run high, and a skill nobody has checked is a claim, not a record. Treat unvalidated skills as drafts and show validated ones with a date. The data is personal, so apply the controls in our guide to ethical AI governance in an HRIS.

Each use needs a different report. Build the report first, then collect only the skills data it needs. That keeps the taxonomy honest.

The World Economic Forum's Future of Jobs Report 2025 expects nearly 40% of the skills required on the job to change by 2030, and 77% of employers plan to upskill their workforce (World Economic Forum, January 2025). Pakistan's National AI Policy, approved in July 2025, targets one million AI professionals by 2030 (The Express Tribune), so competition for digital talent will not ease. Developing your own people becomes cheaper than outbidding the market.

Picture a finance team of eight. Title by title they are an accounts officer, two payroll executives, a tax analyst and four others. Skill by skill, the picture changes. Only one person is rated at level 3 or above in tax filing. Two people can run payroll reconciliation independently. Nobody can teach new joiners. That is a hypothetical, but teams of this size usually show the same shape: a skill that matters held by one person.

Once you see it, the response writes itself. Pair the tax analyst with a second person for the next filing cycle, name a trainer for onboarding and add the gap to next quarter's hiring plan if internal learning will not close it in time. None of that was visible from the title list.

Pick one team and one question, for example, which skills would we lose if our two longest-serving people resigned? Build the taxonomy for that team in two weeks, rate everyone with managers in the room, then review the answers together. You will find gaps and duplicates that no title ever showed.

Then link the skills to your review cycle so the data stays current. Our post on performance reviews as a workflow shows how, and the buyer view in HRMS features Pakistani companies need covers what the system should store.

Workflow Engine is Bitsbuffer's HRMS for Pakistani companies, and we suggest holding it to this list. Bring your first taxonomy to a demo and we will say plainly what it supports today and what it does not.

Review each person's skills once a year with their manager, and prompt an update after any project, certification or role change. Tie the check to your review cycle so it does not become a separate chore. Show the validation date on every record, and treat anything older than 18 months as due for review.

Keep the employee in control. They see their own record, propose changes and see who validated each skill. A record people trust gets used. A record people cannot see gets ignored.

Do not expect a clean database in month one, and do not expect skills data to replace manager judgement. We found no Pakistan study on skills-based adoption, so every figure here is global. Keep the taxonomy small, validate it yearly and drop any skill nobody uses in a decision.