Most Singapore companies approach the Workforce Development Grant (WDG) and its Job Redesign (JR+) pathway the wrong way round: they design an AI project first, then ask a consultant whether it is fundable. The grant does not work like that. WDG(JR+) does not fund technology. It funds evidenced workforce redesign — and the technology only qualifies insofar as it produces the evidence.
That distinction is the whole game. This post sets out what WDG and JR+ actually pay for, what the co-funding range looks like on a real engagement, and the four evidence requirements that decide whether a submission succeeds or dies in assessment.
What WDG and JR+ actually fund
WDG(JR+) co-funds the costs of redesigning work: the job redesign consultancy itself, the job redesign workbook and role documentation, the training plan and delivery for affected staff, and the productivity measurement that proves the redesign changed anything. Eligible activities are the ones that change what people do — task audits, role redesign, reskilling, and the systems that make the new task mix stick.
What it does not fund, in practice, is a software licence bought on its own merits. An AI tool purchased with no redesigned role around it, no training cohort, and no baseline is a capital decision, not a workforce development programme. Assessors see dozens of these a quarter. They are rejected not because the technology is weak but because the submission is not a workforce programme at all.
The 50–80% co-funding range, on a real engagement
Co-funding sits between 50% and 80% of eligible costs, depending on company size and the strength of the redesign case. Take a concrete engagement: a 12-week AI workforce redesign at a Singapore digital media platform, scoped at S$78K across consultancy, build, and training. At the higher co-funding tier, the company's residual is roughly S$15,600–S$39,000 — the exact figure depends on the approved percentage.
Read that again: a full AI redesign of four commercial roles, five production AI systems, and a workforce analytics dashboard, for a residual that is often less than one month of a single commercial hire. The grant is not a subsidy tacked onto the project. For a well-evidenced submission, it is the difference between a project the board approves and one it defers.
The four evidence requirements assessors test
Grant managers assessing an AI workforce proposal are not evaluating your technology stack. They are evaluating four artefacts:
- Role redesign documentation. Which roles change, which tasks move where, and what the before-and-after task split looks like per role. Vague descriptions fail here.
- A training plan and cohort. Who is trained, on what, over what period, with what assessment. "Staff will be trained on the tools" is not a plan.
- A productivity baseline. What the task mix and time allocation were before the intervention. Without a baseline there is nothing to measure a delta against, and the submission has no evidentiary spine.
- Measurable outcome metrics. Which numbers move, by when, and who reports them. Percentage claims without attribution do not survive assessment.
Each of these is a document or a dataset that must exist before submission — or, better, must be generated by the engagement itself from week one.
How a Qlabs engagement builds the evidence trail from day one
Qlabs is a pre-approved delivery partner for WDG(JR+), which means the engagement is structured to produce grant artefacts as a by-product of the work rather than retrofitting them at submission. In week one, 90-minute task-audit interviews with each role-family lead and a two-week time-logging exercise produce the task-frequency matrix — that is your baseline, captured before anything changes.
The Job Redesign Workbook — the redesigned job descriptions, the AI-fit mapping per task, the skills matrix, the training curriculum — is the core grant artefact, and it is client-editable from the first draft. The workforce analytics dashboard goes live in week eight and reports the three metrics assessors accept: hours reallocated from rote to Anchor work per role, skills-assessment closure rate across the training cohort, and revenue-per-FTE delta at quarterly re-audit. By the time you submit, the evidence is not a promise. It is a dashboard with eight weeks of history in it.
The grant consultant and the delivery partner are different jobs
A common point of confusion: the pre-approved grant consultant and the delivery partner are separate roles. A grant consultant such as Cadence HR Consulting handles the application itself — the submission mechanics, the funding tier, the paperwork and liaison with the agency. The delivery partner — Qlabs, in our engagements — delivers the redesign and the systems, and produces the evidence the application cites.
The application is only as strong as the evidence behind it. The consultant writes the case; the delivery partner has to have built it.
Companies that try to collapse these roles — asking their systems vendor to "handle the grant" — typically produce a submission with no baseline and no training cohort, because the vendor was never scoped to produce them.
Start the evidence trail before you start the application
The practical sequence is therefore: scope the redesign, capture the baseline, begin the engagement, and let the grant application cite evidence that already exists. Companies that reverse the order — apply first, build later — end up retrofitting a baseline after the fact, which assessors can usually tell.
If you are considering WDG(JR+) for an AI workforce programme, the cheapest mistake to avoid is starting the application before the evidence trail exists.
Eligible versus ineligible: a working checklist
Because the boundary matters in practice, here is the checklist we apply when scoping an engagement that is intended to be grant-fundable. Eligible: the task-audit diagnostic and baseline capture; the job redesign consultancy and workbook production; redesigned job-description drafting; the skills matrix and training curriculum design; training delivery for the named cohort; the workforce analytics dashboard insofar as it measures workforce outcomes; and the re-audit cycles that certify the redesign held.
Ineligible, or at best contested: general-purpose software licences bought without a redesign rationale; infrastructure upgrades untied to a redesigned role; marketing spend; and headcount costs themselves. The test we apply is simple — if the cost would still exist in a world where no roles changed, it is not workforce redesign, and an assessor will read it that way too.
The residual, in cashflow terms
Finance directors ask a question the grant literature rarely answers: when does the money move? Co-funding is typically reimbursed against claimed eligible costs, which means the company funds the engagement first and claims after. On a S$78K engagement the residual at 80% co-funding is about S$15,600 — but the company must carry the full S$78K through the engagement and the claim cycle. That cashflow shape, not the residual percentage, is what surprises boards.
The mitigation is sequencing: phase the engagement so that early gates produce the evidence the claim requires, and keep phase sizes inside the company's working-capital comfort. A grant-funded programme that stalls for cashflow reasons produces neither the redesign nor the claim.
Common submission mistakes we see as delivery partners
Three mistakes recur. First, the baseline captured after the build began — assessors date artefacts, and a reconstructed baseline reads as one. Second, training described as an intention rather than a cohort: no names, no curriculum, no completion metric. Third, technology described in vendor language — "an AI platform to streamline operations" — rather than in task language: which task moves where, with which human confirmation point. The workbook exists to prevent all three, which is why it sits at the centre of the submission.
How WDG interacts with SkillsFuture and other funding
Companies frequently conflate WDG(JR+) with SkillsFuture schemes, and the confusion costs sequencing errors. SkillsFuture supports individual training — course fees, credits, training awards for staff development. WDG(JR+) funds the redesign of work itself: the consultancy, the role documentation, the measurement. In a properly structured programme the two stack: the training cohort inside a WDG(JR+) engagement may itself draw on SkillsFuture support for course fees, while the redesign and baseline sit under the grant.
The practical rule: training that exists because a role changed belongs in the WDG(JR+) training plan; training that exists because an individual needs a skill regardless of any redesign belongs to SkillsFuture. Assessors read a submission that double-claims the same course fee under both as a control weakness, so the split is documented in the workbook at design time. Your grant consultant — Cadence HR Consulting, in our engagements — owns that boundary in the application; the delivery partner owns the artefacts that make the split auditable.
Frequently asked questions
What costs are grant-eligible?
Job redesign consultancy, role documentation and the job redesign workbook, training plan and delivery for affected staff, and productivity measurement. Standalone software licences without a redesign programme around them are generally not eligible.
Can a small team (under 20 staff) apply?
Yes. Co-funding percentages vary by company size, and smaller companies often qualify at the higher tier. What matters is the quality of the redesign evidence, not headcount.
How long does grant approval take?
Timelines vary with the agency's queue and the completeness of the submission. Submissions with a captured baseline and a concrete training plan move faster because assessors spend less time requesting clarification.
Contact Qlabs before you submit your grant application — we build the evidence trail in from day one. Email hello@qlabshq.com.