How the IMAbled Ability Match Algorithm Works

Traditional job matching is keyword matching: your CV mentions "Python" and the job requires "Python" — you match. This works fine if every candidate's path, conditions, and work needs are identical. They are not. IMAbled's ability match algorithm was built to capture what keyword matching misses: the full picture of what a candidate can contribute and what environment they need to do it.

Here is exactly how the algorithm works, what it weighs, and how to work with it rather than against it.

The Five Matching Dimensions

Dimension 1: Skills Match (Weight: 40%)

The largest weight goes to skills — because skills determine your actual ability to do the work. The algorithm matches your skills to the skills required for each job posting using:

• Exact match: You list "SQL" and the job requires "SQL" — full score for this skill
• Semantic match: You list "MySQL" and the job says "database querying" — partial score, because these are functionally related
• Complementary skills: You have skills that are commonly paired with required skills — e.g., Python with data analysis — which signals likely proficiency even if not explicitly listed
• Skill depth indicators: Years of experience with each skill, certifications, and work samples where provided

Profile optimisation tip: Be specific and complete about your skills. "Data analysis" as a skill is worth less than listing "Microsoft Excel (pivot tables, VLOOKUP), SQL (MySQL, PostgreSQL), Python (Pandas, NumPy), Tableau." The more granular your skill listing, the more accurately you match to relevant roles.

Dimension 2: Experience Level Alignment (Weight: 20%)

This dimension prevents mismatches in both directions — a senior professional being matched to entry-level roles, or a fresher being matched to director-level positions. The algorithm uses:

• Total years of work experience (all types — full-time, part-time, freelance, NGO placement)
• Complexity of roles held (individual contributor vs team lead vs manager)
• Sector-specific experience depth

Crucially: employment gaps do not reduce this score. The algorithm measures what experience you have accumulated, not how recently or continuously it was held.

Profile optimisation tip: Include all experience — every role, every project, every NGO placement, every internship. Part-time roles count proportionally. Do not leave relevant experience unlisted because you think it "does not count."

Dimension 3: Work Preference Compatibility (Weight: 20%)

This dimension is unique to ability-first matching. Your stated work preferences are matched against the employer's job posting to prevent placements that look good on paper but fail in practice:

• Work mode: Remote, hybrid, or in-person — if you need remote and the job is fully in-person, match score drops significantly
• Location: Geographic match between your locations and the job location (for in-person or hybrid roles)
• Work hours: Full-time, part-time, flexible hours — matched against the role's requirements
• Sector preference: Your preferred industries matched against the employer's sector

Profile optimisation tip: Be honest about your work preferences, not aspirational. If you genuinely need remote work due to mobility constraints, mark that clearly. An in-person job match will fail at the interview stage — and an accurate preference saves time for everyone.

Dimension 4: Accommodation Compatibility (Weight: 15%)

This is the dimension that makes IMAbled fundamentally different from every general job board. The algorithm matches your stated accommodation needs against what the employer has listed they can provide.

• If you need a screen reader-compatible environment and the employer has listed this as available → high compatibility score
• If you need sign language support in meetings and the employer has not listed this → lower compatibility score → the job appears lower in your matches, reducing the chance of a mismatch
• If the employer offers flexible hours and you have listed this as a need → positive boost to the match score

This dimension works to protect both parties — you do not waste time interviewing for roles where your accommodation needs cannot be met, and employers do not onboard candidates whose workplace needs they cannot accommodate.

Profile optimisation tip: List accommodation needs accurately and specifically. Vague entries like "accessible workplace" are less useful than "screen reader-compatible software, screen magnification 200%, ground floor or lift access." Specificity improves match precision.

Dimension 5: Application and Engagement History (Weight: 5%)

A smaller weight goes to behavioural signals that suggest relevance and fit:

• Types of roles you have applied to previously
• Which job types you engage with in the platform (click, save, share)
• Employer feedback on past applications (structured feedback from employers who have reviewed your profile)

This dimension helps refine your matches over time — the more you use IMAbled, the better it understands what actually suits you.

What the Match Score Means

Each job in your feed shows a match percentage. Interpret it as follows:

• 85–100%: Excellent match — strong skill alignment, work preference compatibility, accommodation compatibility. Prioritise these applications.
• 70–84%: Good match — some skill or preference alignment gaps. Worth applying if the role interests you; check what the gaps are before applying.
• 55–69%: Partial match — notable gaps. Apply if you are actively stretching toward this role type; be transparent about the gaps in your application note.
• Below 55%: Weak match — listed for awareness, not as a primary recommendation. May reflect aspirational interest that is not yet supported by your current profile.

How NGO Assessment Improves Match Quality

When a candidate comes through an NGO partner, the NGO provides an additional layer of assessed capability data:

• NGO's formal assessment of the candidate's functional skills
• Training completion records with competency levels
• Job-readiness assessment (not just skill scores, but soft skills, communication, punctuality, teamwork)
• Recommended role types based on the NGO's placement experience

This enriched data significantly improves match quality for NGO-sourced candidates — and is a major reason why NGO-placed candidates have higher retention rates on IMAbled compared to direct applicants without this assessment layer.

How Employers Can Improve Their End of the Match

Match quality depends on the quality of both sides of the profile. Employers can improve their match results by:

• Listing specific skill requirements rather than broad job descriptions
• Being detailed about which accommodations they can provide — not just "accessible workplace" but "screen reader-compatible software, ground floor office, flexible hours 8AM–7PM core anytime"
• Marking the interview formats they can offer
• Specifying their NGO partner preference (some NGOs train for specific skills that match the role)

The Match Algorithm Does Not See Your Condition

To be explicit: your condition type, condition percentage, and medical details are not inputs to the matching algorithm. The algorithm is ability-first — it matches on what you can do, not on what you have. Accommodation needs (how you work best) feed into one dimension, but your condition type itself is entirely invisible to the match engine and to employers at the application stage.

Ready to see your matches? Create or complete your IMAbled profile to activate your ability match score across all available roles.

Frequently Asked Questions