Difficulties in tracking employment outcomes,skill gaps, and the impact of skilling initiatives
Government Of Maharashtra · Smart Education · Software
The hard part here is getting people to respond years later, which is not a software problem — you would be building analytics over data you invented, so take SIH26134 instead where the demand signal is genuinely real.
What it actually is
Training programmes record who enrolled and who was certified, but nobody knows whether trainees actually got jobs, kept them, or earned more. Trainees change phone numbers, employers do not report, and different programmes use different identifiers. The ask is a system that follows outcomes over time without burdening anyone, and while respecting consent.
What to build
A longitudinal outcomes system built on a consent-based trainee record with a stable identifier that survives phone number and location changes, linking training completion to employment signals from multiple sources, running automated and assisted follow-ups at defined intervals through low-effort channels with escalation to a human caller only when automated contact fails, capturing self-employment and apprenticeship outcomes that formal payroll signals miss, validating employer-reported placements against independent signals, and producing cohort, course, provider, district and demographic analytics that surface not just placement rates but wage progression, retention and the stated reasons for non-placement.
Smallest thing that wins the room
Show one cohort tracked across eighteen months with placement, retention and wage progression on a single curve, then break it by provider to reveal one provider whose day-ninety placement looks strong but whose day-three-hundred-sixty-five retention collapses.
How crowded this one gets
A guess, projected from the 2025 statements — the last year where both the submission counts and the winners were published.
Quieter than 2% of the 226 · #221 of 226 by expected field · reaches the 500 cap
Forecast to blow past the 500-idea cap. Submissions close when it fills, so late teams may not get in at all.
Why: state governments drew the biggest crowds in 2025.
This is a guess, not a fact
Nobody has published 2026’s numbers yet. This is an analysed estimate from last year’s pattern, so please do not take it as the truth — check the live counter on the SIH portal before you decide anything. The range covers the middle half of likely outcomes, so one statement in two lands outside it. Entry closes at 500 ideas per statement, so no range goes past that — a statement that reaches the cap fills and shuts rather than drawing an unlimited crowd. The model reads only three things a team can see before choosing — software or hardware, the theme, and what kind of body posted it — and those explain about a quarter of the variation in last year’s field sizes (R² 0.25 on held-out statements). Trust the band more than the number, and the ordering more than either. It cannot see how good your idea is, which is the part that actually decides it.
The scores
The number is the shorthand. The line under it is the reason.
Acceptance potential
2/5The core difficulty the description names — trainees becoming uncontactable and employers not reporting — is a human incentive problem that no software feature solves, so a platform built on longitudinal data you synthesised demonstrates dashboards rather than the tracking that is actually hard.
Feasibility
4/5Consent records, follow-up scheduling, multi-channel contact and cohort analytics are all standard engineering, though the longitudinal data itself must be synthesised since no real outcome dataset is available.
Innovation scope
3/5The description enumerates the expected components, so your genuine room is in the follow-up strategy that survives contact loss and in how you validate an employer's claim against independent evidence.
Clarity
3/5The capabilities are listed clearly and the privacy constraint is stated, but no programme, cohort size or outcome definition is specified, and definitions are precisely what the description says vary across programmes.
Effort
HeavyConsent management, identifier resolution, multi-channel follow-up, employer validation and cohort analytics is five components, all conventional but collectively substantial.
Demo-ability
MediumThe retention-collapse finding is a genuinely arresting result, but it comes from data you generated, so the demo shows an analytical capability rather than a discovery.
In its favour
- Green flag: Distinguishing day-ninety placement from day-three-hundred-sixty-five retention is a genuinely important analytical distinction that exposes provider quality and is easy to visualise
- Green flag: The consent and revocation layer is real, buildable substance and directly answers the privacy constraint the description raises
- Green flag: Capturing self-employment outcomes, which formal payroll signals miss entirely, addresses a real blind spot in Indian skilling measurement
- Green flag: Probabilistic record linkage across programmes using different identifiers is legitimate technical work with mature libraries
Against it
- Red flag: The stated central difficulty is that trainees become uncontactable and employers do not report, and this is an incentive and outreach problem that no amount of software design solves
- Red flag: No real longitudinal outcome data exists, so your analytics run entirely on data you generated and reveal only the patterns you encoded
- Red flag: Employer validation requires an independent signal to validate against, and the description does not name one that is actually accessible
- Red flag: It overlaps SIH26134 from the same state closely enough that a judge may treat them as one problem
What you will be writing
- Consent artefact management with revocation
- Probabilistic record linkage across programme identifiers
- IVR / WhatsApp automated follow-up with escalation
- Cohort survival and retention analysis
- Wage progression time-series analytics
- React multi-level analytics dashboard
- Programme evaluation
- Longitudinal outcome tracking
- Skills policy
Prior art to read before you start
longitudinal trainee outcome measurement · consent-based follow-up systems · provider performance analytics
Analysed by Claude Opus. Every score above is a judgment call with its reasoning attached — kindly cross-check this against the official statement on the SIH portal before your team commits to it.