Help and more
How every number is worked out
No score on NexRecruit is a black box to the person it describes. Every number tells you what it measures, what moves it, and why it exists, and a person reviews every match. No score rejects anyone on its own. The exact weightings are ours and stay private.
The metrics
01Profile strength
Worked example: A CV and a LinkedIn profile together get you most of the way. Qualifications, identity and a licence lift it from there, in that order of impact.
Why it exists: The weights mirror what SA employers ask for first. A CV and LinkedIn profile decide whether you get shortlisted; verified extras decide how fast an employer can say yes.
Match readiness
Worked example: Profile strength counts for most of it. Switching your profile on is the single biggest jump available to you after that, and supporting documents add the rest.
Why it exists: Completeness is what screeners score, visibility is what makes you matchable at all, and extras cut verification time. It tells you exactly which lever to pull next.
Live demand
Worked example: If four companies have live role requests open and two of them are verified employers, that is what the number shows you. It is a count, not an estimate.
Why it exists: Real demand, not motivation copy. When this number grows, your odds grow, and you can see it.
Spec strength
Worked example: A role title and a location get a spec most of the way. Detailed must-have notes are what close the gap, and they are the part most companies skip.
Why it exists: A tight spec is the single biggest predictor of a fast, accurate shortlist. The score shows exactly what is missing before you submit.
Match potential
Worked example: It weighs how completely you described the role against how many candidates sit in the salary band you chose. A sharper spec and a realistic band both move it up.
Why it exists: It combines what you control (the spec) with what the market holds (the pool), so you know the odds before the clock starts. Pool numbers are sample data until launch; after launch they are live counts of consented, visible profiles matching your role.
Founding spots counter
Worked example: 8,214 registered members: 10,000 minus 8,214 = 1,786 spots left.
Why it exists: Early access is genuinely capped at 10,000 founding members. The counter is never inflated or faked; when it reaches zero, registration closes until the next intake. Your founding number is assigned in order and never changes.
Launch marketing list
Worked example: 12 job seekers and 3 companies opted in: launch list = 15.
Why it exists: POPIA s69 allows direct marketing to people who opted in. Only this list receives launch mail; everyone else stays untouched.
Role fit
Worked example: Every role has a checklist of the skills and terms SA employers actually ask for. Fit is how much of that checklist your CV already evidences, so naming the tools and tickets you actually hold is what moves it.
Why it exists: Your CV is read by software on our own server (it never leaves the machine) and compared against a fixed public checklist per role. The score only helps our team rank likely matches faster; it never rejects anyone, and a human reviews every shortlist before an employer sees a name. A scanned or photographed CV cannot be read this way, which is also true of most recruiter systems; a typed PDF always serves you better.
LinkedIn discovery score
Worked example: A profile with a photo, a location, a 220-character headline, 850 characters of About text, 3 dated roles with action verbs, 42 skills, 2 recommendations, and matric education scores about 78 of 100.
Why it exists: Each rule maps to LinkedIn published research: photos drive materially more views, 5+ skills gets up to 33x more recruiter contacts (LinkedIn own data), quantified experience is a documented InMail lift, All-Star completeness correlates with more inbound. We read a screenshot of your public profile on our server (it never leaves the machine), extract the text, and score against these levers. Every point of every section is visible in the audit, so you can see exactly which lever moves the number and act on it.
EEA gap per cell
Worked example: A gap is simply where a company sits below its own employment equity target for a level, using the aggregated EEA2 numbers it supplied. Nothing about any individual is used to compute it.
Why it exists: The 6 occupational levels and 11 demographic bands come straight from your annual EEA2 report to Department of Employment and Labour. We only use the aggregated numbers (Sections B and C) so no individual employee PII is stored. Clamping at zero matters: a cell where you are ahead of target does not create a negative gap that would penalise otherwise qualified candidates. It only means we are not actively looking to close that cell.
EEA match score for a candidate
Worked example: A candidate scores higher where the company has a larger shortfall against its own target at that occupational level. It only ever applies when the candidate has opted in.
Why it exists: This score is only ever a re-order, never a filter. A candidate who never shared demographics scores a neutral 50, so they still get ranked on skills. A company that never shared a workforce profile scores every candidate as 100 (no adjustment). Only when both sides have opted in does the number push the ranking.
Shortlist ranking signals and grades
Worked example: A candidate who fits the role on skills, lives where the job is, holds the licence it calls for and has real years behind them ranks at the top and earns an A. Each of those signals is listed by name next to the candidate, so a person can see why they ranked where they did.
Why it exists: Skills are settled first, so a location or equity tailwind can never lift an unqualified candidate into a shortlist. The other signals only separate candidates who already fit. Every signal carries a written reason a person reads before choosing anyone, and no signal ever subtracts points from a candidate.
Combined shortlist ranking: the base score
Worked example: Skills come first and set the base. Practical signals such as living in the right city then adjust it. A strong equity match never rescues a weak skills match, by design.
Why it exists: Skills are a floor, not a variable, and the maths is built so it stays that way: a weakly qualified candidate with a perfect equity match never leapfrogs a qualified one. An equity match lifts a candidate who already fits; it never creates one. Data freshness controls: a workforce profile older than 90 days shows an aging chip, and older than 365 days you must re-upload or re-attest before your next shortlist request.
Documents and uploads
02Right now: nothing. We are not collecting CVs or documents until our registration as a private employment agency is granted, so the only thing we ask for today is an email address. When we open: CV first, then a screenshot of your LinkedIn profile, with qualifications and supporting documents lifting your scores.
Formats: any document type. Up to 40 MB per file.
Big uploads: the tray shrinks to the corner and keeps running while you explore. You never have to wait on a page.
CV templates: the research behind each one
03Linguistic hard rules applied to every field before it prints.
Every string you type is sanitised on our server before it lands in a template. No em or en dashes (hyphen used, or the word "to" between years). Straight quotes only. No non-breaking spaces or soft hyphens (invisible copy-paste landmines). Sentence starts and the first letter of the field are capitalised. Phrases known to sink callback rates ("responsible for", "duties included", "in charge of" per Ladders 2020 study of 500 opening phrases) are auto-replaced with an action verb ("led", "delivered", "managed"). No registered mark ever appears.
Atlas: ATS-safe single column.
Modern ATS platforms filter primarily by keyword match, not by formatting alone. A 2024 Jobscan analysis found around 15% of resumes suffer critical parsing failures and 45% partial parsing failures on complex layouts. Atlas uses standard section headers (About, Experience, Education, Skills), Arial-family type, and no tables, text boxes or graphics. The popular "75% of resumes are auto-rejected by ATS due to formatting" claim is a myth traceable to a defunct company; the real risk is being filtered out of a keyword-ranked list.
Kopje: two-column with pine sidebar.
Source: Ladders 2018 eye-tracking study of 30 recruiters reading resumes. Recruiters spend 7.4 seconds on the first scan (up from 6 seconds in the 2012 study) and read in an F-pattern: horizontal across the top, vertical down the left. A left sidebar with contact/skills/education means the recruiter absorbs the essentials in the first fixation.
Vaal: executive minimalist.
Executive hiring managers scan fast. The specific "CareerBuilder 2020: 40% under a minute on senior CVs" figure that circulates online is a conflation of at least two separate CareerBuilder studies and cannot be verified. What is universally advised by senior recruiters is that dense pages that hide the value proposition lose the read. Vaal uses generous whitespace and a centred header so the reader absorbs identity and tagline before deciding to read on.
Drakensberg: deep chronological, 2 pages.
Sources: SHRM 2023 survey: 63% of recruiters prefer 2-page resumes for experienced candidates. ResumeGo 2024: recruiters were 2.3x more likely to prefer 2-page resumes for senior roles. The often-quoted "87% prefer reverse-chronological" Zety figure is misattributed; reverse-chronological is nevertheless the default preferred by most recruiters. Recent 3 roles get 5-8 bullets each; earlier roles compressed to one line.
Savanna: skills-first hybrid.
Source: Pin 2024 analysis of LinkedIn Recruiter searches: over 95% of searches include a skill filter; recruiters are 50% more likely to search by skill than by years of experience. For technical, digital, and skilled trades roles, a categorised skills matrix at the top expands surface area against Boolean skill searches.
Table Mountain: one-page concise.
The often-quoted "TopResume 2019: 63% prefer 1 page for under 5 years" figure cannot be verified. Recent evidence in fact favours 2 pages even for early career (ResumeGo 2024: recruiters 2.3x more likely to prefer 2 pages). Table Mountain remains one page for candidates whose story genuinely fits without padding; forcing 2 pages of thin experience reads as filler.
Protea: achievement vault.
The specific "HBR 2014: 8:1 callbacks" figure that circulates online lacks a verifiable primary source. What is universally endorsed by Harvard Business School career services and virtually every hiring manual is that quantified achievement bullets outperform duty statements for callback rate. Protea auto-selects your bullets containing numbers, percentages or rand values and places them BEFORE the role list.
Karoo: recruiter-scan F-pattern.
Source: Ladders 2018 eye-tracking (same study as Kopje). The recruiter's eye lands top-left on the name, moves right along the headline, then drops down-left to the current role. Karoo puts the current role in a bold bar directly under the name, matching how the eye actually moves.
Indlela: trade and technical.
Source: CompTIA/ITWeb 2026 South African survey: 97% of SA employers say industry-recognised certifications are important for validating technical skills. Indlela hoists licences to the top block with prominent typography, matching how site foremen, healthcare hiring, and technical recruiters actually screen.
Ubuntu: career-changer functional.
The specific "SHRM 2021: 34% callback lift" figure lacks a verifiable primary source. What is broadly agreed by career-change coaches: chronological listing of mixed-industry roles reads as noise, and a functional layout that groups by transferable strength improves signal for pivot hires. Ubuntu regroups your experience bullets by theme and presents chronology below.
LinkedIn playbooks: the research behind each lens
04Same audit, three lenses.
Your LinkedIn discovery score never changes. The three playbooks re-weight and re-sort the same eight sections so you see the biggest wins for your specific goal first. Every rule under every playbook is a research citation, not an opinion.
Recruiter Magnet.
Goal: get found in LinkedIn Recruiter search. Emphasises Skills, Headline, and Basics (photo, location).
Sources: Pin 2024 analysis of LinkedIn Recruiter: skill filters appear in over 95% of searches. LinkedIn own data: members with 5+ skills receive up to 33x more recruiter contacts and 13x more profile views. LinkedIn own data: profiles with photos consistently receive materially more views and messages than photo-less profiles. Location is a hard filter in Recruiter city search: no location, no result.
Career Momentum.
Goal: get warm InMails and inbound offers. Emphasises Recommendations, Basics (photo drives messages), About (contact detail).
Sources: LinkedIn Talent Insights: profiles with several recommendations correlate with higher recruiter InMail response. LinkedIn own data: profiles with photos consistently receive materially more messages. About sections that name an email/DM channel see materially higher outbound-to-inbound conversion (LinkedIn Marketing Solutions data).
Deep Specialist.
Goal: position as the expert in your niche. Emphasises About (depth), Experience (quantified), Featured (published content).
Sources: Edelman + LinkedIn B2B Thought Leadership Impact 2024: majority of decision-makers say a proven expert profile drives consideration. LinkedIn Creator research: profiles with Featured content are treated as active-user signals by the algorithm. Quantified experience in Experience is broadly documented as a stronger InMail signal than duty statements.
Straight answers
05Is this really free for job seekers?
Yes, and it stays free. If anyone asks you for money in your job search, it is not us; report it and we will act.
Who can see my details?
Nobody. All we hold today is an email address, and it is never shared or sold. When we open, only companies we have verified will see a profile, and only while you switch it on.
Does a score ever reject me?
No. Scores show you what to improve and help our team prioritise; a human reviews every match before anything reaches an employer.
Will you ask for a certified ID?
Not yet, and please do not send one. Once we are registered and open, identity verification is what employers need before an offer, and having it ready moves you from days to hours at the moment speed matters.
What is two-factor authentication?
A 6 digit code from your phone that stops anyone who guesses your password. Your documents deserve it; admins are forced to use it.
What does it cost a company?
Pre-launch, nothing: queued roles lock founding access. Paid screening switches on at launch and founding members go first.
Something unclear? hjr@nexbdm.com. NexRecruit improves how you are presented to employers. It never promises or guarantees employment.