a reading order, not a verdict.
Each applicant gets a match against the job they applied to: a percentage, a Strong, Good or Partial band, and the must-haves they meet and miss. It sets the order a company reads in, and people at the company decide who moves on.
one job, one applicant to follow.
Kavya applies to a frontend engineer job that asks for React, TypeScript and Accessibility and 3 to 6 years of experience. The rules that place her on the list are the real ones.
The job
Frontend engineer
Experience 3 to 6 years
- React
- TypeScript
- Accessibility
The applicant
Kavya M.
Frontend developer
Pune · 2 years
- React
- Accessibility
- JavaScript
- CSS
Frontend developer · 2 years · Built accessible checkout forms in React
her whole profile, against the whole job.
The match starts from one comparison of everything in Kavya's profile with everything the job describes. How closely her roles and work line up counts, not just whether skill names match.
From the profile
- Headline and professional brief
- Total years of experience
- City
- Skills
- Roles, with title, company, years in each and the main highlights
- Highest education
Email, phone and links: removed first
From the job
- Title and what the role involves
- Experience range
- Location and work mode
- Employment type
- Must-haves and nice-to-haves
- Responsibilities and requirements
Benefits, about the company, why join, how to apply: left out
a gap lowers it. nothing caps it.
Kavya is missing TypeScript and has 2 years against a range of 3 to 6, so each lowers her match a little. Neither removes her, and a missing must-have never caps the match at a set number.
Kavya starts where the comparison places her. TypeScript missing moves her match a little lower, and 2 years against a range of 3 to 6 moves it a little lower again. She lands in Good.
- Met
- Named in the headline, professional brief or work history, on the skills list, or shown through something close. Postgres counts for PostgreSQL, and k8s for Kubernetes.
- Listed only
- A skill only on the skills list counts for less than one the work history backs up.
- Missing
- Lowers the match a little, in proportion to how many must-haves the job lists. A near miss still earns some credit.
- Years
- Inside the range, or a little either side, changes nothing. Further out lowers the match, up to a limit.
- None set
- With no must-haves, the match comes from the comparison and the years alone.
three applicants, one list.
Kavya and Imran each meet two of the three must-haves, yet land in different bands. Open a row to see why.
01Tara S.Frontend engineer, design systems5 yearsStrong match3 of 3 must-haves metMetReactTypeScriptAccessibility, shown through WCAGWhy this band
A frontend role throughout, five years inside the range, and all three must-haves met, Accessibility through WCAG.
02Kavya M.Frontend developer2 yearsGood match2 of 3 must-haves metMetReactAccessibilityMissingTypeScriptWhy this band
The same kind of role with two gaps: TypeScript is missing and two years is under the range. Each lowers the match a little.
03Imran K.Backend engineer, Java services7 yearsPartial match2 of 3 must-haves metMetReactTypeScriptMissingAccessibilityWhy this band
React and TypeScript are on the skills list, but the roles are backend services, so the profile as a whole describes a different job. Skills the work history does not back up count for less.
strong, good and partial.
Every match falls in one of three bands, and the company sees both the percentage and the band. Tara, Kavya and Imran land in one each.
- Match 75 to 100
Strong
The profile as a whole matches the job and the must-haves are shown.
- Match 50 to 74
Good
A credible fit with gaps: an adjacent stack, a few must-haves missing, or experience off the range.
- Match 0 to 49
Partial
An adjacent role at best.
where it can get it wrong.
The match is a reading order with known blind spots, and we would rather you know them. It is one reason every applicant stays in your pipeline.
- 01
Close equivalents
Some are not caught, such as Kafka shown through Redpanda or Python through pandas. They earn some credit and show as missing, so read the profile.
- 02
Thin profiles
A profile with very little text gives the comparison less to go on, so it leans on the headline, professional brief and skills.
- 03
Engineering first
The scoring was tuned on engineering roles. Matches for other kinds of roles have not been validated yet.
people decide who moves on.
Does the ranking reject or hide anyone?
no.
Every application goes into the company's pipeline whatever its match, and only a person at the company moves, hires or declines anyone. A declined application can be restored.
- Sort by best match or by newest.
- A minimum match filter is off unless the company sets one. While it is on, applications still being scored are not shown.
- An application still being scored shows Scoring and is in the pipeline from the start.
- What the company sees
- The match percentage and its band
- How many must-haves are met, out of how many the job lists
- Each met must-have, and what it was shown through when a close equivalent met it
- Each missing must-have
- What the candidate sees
- The round each application has reached, with an email once a move settles
- When the company can see their contact details
The match itself is not shown on the candidate's side. Email, phone and resume file stay private until the company moves the candidate to its final round or marks them hired, unless the candidate shares them sooner.
post one role. read it in this order.
One posting is ₹2,000$39, refundable until you use it to publish. Joining the talent pool is free for candidates.
See pricingQuestions about ranking? hello@oorik.in