{"id":338,"date":"2026-08-17T12:16:41","date_gmt":"2026-08-17T12:16:41","guid":{"rendered":"https:\/\/i10x.ai\/blog\/?p=338"},"modified":"2026-08-18T07:22:53","modified_gmt":"2026-08-18T07:22:53","slug":"ai-candidate-sourcing","status":"publish","type":"post","link":"https:\/\/i10x.ai\/blog\/ai-candidate-sourcing","title":{"rendered":"AI Candidate Sourcing: Scorecards and Outreach That Work"},"content":{"rendered":"\n<div class=\"i10x-article\">\n\n<p class=\"i10x-pill\">Guide \u00b7 August 2026<\/p>\n\n<p class=\"i10x-lead\">\nAI candidate sourcing is how talent teams find, rank, and open conversations with people who are not already in the apply pile. Used well, it expands passive reach, speeds first contact, and keeps criteria explicit. Used poorly, it sprays generic messages, overweights pedigree proxies, and burns employer brand. This guide covers hybrid search, the Sourcing Fit Scorecard, 3-touch and 5-touch outreach, channels, tracking fields, privacy high-level notes, free vs paid stacks, passive vs active talent, failure modes, metrics, and FAQs. Pair it with\n<a href=\"https:\/\/i10x.ai\/tools\/category\/business-management\/free-ai-recruiting\" rel=\"noopener\" target=\"_blank\">Free AI Recruiting on i10X<\/a>\nand the broader\n<a href=\"https:\/\/i10x.ai\/blog\/ai-recruiting-guide\">AI recruiting guide<\/a>.\n<\/p>\n\n<div class=\"i10x-highlight-stats\">\n<table class=\"i10x-table\">\n<tbody>\n<tr>\n<td><p>43%<\/p><\/td>\n<td><p>Organizations using AI in HR tasks in 2025, up from 26% in 2024 (SHRM 2025 Talent Trends)<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>37%<\/p><\/td>\n<td><p>Orgs actively integrating or experimenting with gen AI in hiring (LinkedIn Future of Recruiting 2025), up from 27%<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>~20%<\/p><\/td>\n<td><p>Average share of the workweek TA pros using gen AI report saving (LinkedIn Future of Recruiting 2025)<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>+9%<\/p><\/td>\n<td><p>Higher likelihood of a quality hire when recruiters use AI-Assisted Messaging most vs least (LinkedIn)<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>39 days<\/p><\/td>\n<td><p>Median time-to-fill for nonexecutive roles (SHRM 2026 Recruiting Executives Benchmarking)<\/p><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n\n<hr>\n\n<h2 id=\"what-is-ai-candidate-sourcing\">What is AI candidate sourcing?<\/h2>\n<p>AI candidate sourcing is the use of language models, matching systems, and search tools to identify people who may fit a role, draft personalized outreach, and keep sequences organized, while humans still decide who to contact and how hard to pursue. It covers passive talent (not actively applying), reactivation of warm past applicants, and smarter use of employee or community referrals.<\/p>\n<p>Sourcing is different from\n<a href=\"https:\/\/i10x.ai\/blog\/ai-resume-screening\">AI resume screening<\/a>.\nScreening ranks people who already applied. Sourcing builds the top of the funnel before or alongside inbound applications. Many teams mix both: AI suggests lookalikes and Boolean strings, then drafts first-touch messages that a recruiter edits and sends.<\/p>\n<p>In 2026, AI candidate sourcing usually means a combination of:<\/p>\n<ul>\n<li><strong>Search assistance:<\/strong> Boolean strings, semantic queries, and \u201cpeople like this profile\u201d suggestions.<\/li>\n<li><strong>Fit ranking:<\/strong> scoring public or CRM profiles against a scorecard, not a vague job ad dump.<\/li>\n<li><strong>Message generation:<\/strong> short, specific outreach and follow-ups grounded in real profile signals.<\/li>\n<li><strong>Workflow memory:<\/strong> who was contacted, what they said, and when to stop (often via ATS, CRM, or an agent).<\/li>\n<\/ul>\n<p>For multi-step automation that chains source, screen, and shortlist prep, see\n<a href=\"https:\/\/i10x.ai\/blog\/ai-recruiting-agents\">AI recruiting agents<\/a>\nand the end-to-end\n<a href=\"https:\/\/i10x.ai\/blog\/ai-recruiting-workflow\">AI recruiting workflow<\/a>\nmap in this series. Ethical constraints on ranking and notice live in\n<a href=\"https:\/\/i10x.ai\/blog\/ethical-ai-recruiting\">ethical AI recruiting<\/a>.<\/p>\n\n<hr>\n\n<h2 id=\"why-sourcing-ai-matters-now\">Why AI sourcing matters now<\/h2>\n<p>Adoption is rising across HR, not only screening. According to SHRM\u2019s 2025 Talent Trends research, <strong>43% of organizations used AI in HR tasks in 2025, up from 26% in 2024<\/strong>. LinkedIn\u2019s Future of Recruiting 2025 report found that <strong>37% of organizations were actively integrating or experimenting with generative AI in hiring, up from 27%<\/strong>, and that talent acquisition professionals using gen AI reported saving about <strong>20% of their workweek on average<\/strong>.<\/p>\n<p>That time usually lands on repetitive drafting and first-pass research, not on replacing judgment. LinkedIn has also reported that companies whose recruiters use <strong>AI-Assisted Messaging most are about 9% more likely to make a quality hire<\/strong> than those who use it least. Personalized, relevant outreach still matters; AI helps more people write it consistently when the inputs are real.<\/p>\n<p>Speed pressure is real too. SHRM\u2019s 2025 Recruiting Benchmarking cycle has been widely reported with a <strong>median time-to-fill around 44 days for nonexecutive roles<\/strong>, while SHRM\u2019s 2026 Recruiting Executives Benchmarking reports a <strong>median time-to-fill of 39 calendar days for nonexecutive roles<\/strong>. Cost pressure sits beside speed: SHRM 2025 Benchmarking Report averages (as reported in press coverage) put nonexecutive <strong>cost-per-hire around $5,475<\/strong> and executive around <strong>$35,879<\/strong> (averages; other SHRM releases use medians that can differ). Sourcing that fills critical seats faster can move those numbers, but only if quality and fairness hold.<\/p>\n<p>AI sourcing fails as a spray cannon. It works when fit is written first, proxies are banned, and contact, long sequences, and DNC stay human-owned.<\/p>\n\n<hr>\n\n<h2 id=\"passive-vs-active-candidates\">Passive vs active candidates (and why AI treats them differently)<\/h2>\n<p><strong>Active candidates<\/strong> already apply or update profiles; risk is volume and fair screening. <strong>Passive candidates<\/strong> need a reason to reply (specific problem, credible story, easy no); default shorter sequences and stricter caps. <strong>Warm reactivation<\/strong> (prior applicants, silver medalists) often beats pure cold; check DNC and recontact rules before AI drafts.<\/p>\n\n<table class=\"i10x-table\">\n<tbody>\n<tr>\n<th><p>Segment<\/p><\/th>\n<th><p>Primary goal<\/p><\/th>\n<th><p>AI role<\/p><\/th>\n<th><p>Human gate<\/p><\/th>\n<th><p>Typical sequence<\/p><\/th>\n<\/tr>\n<tr>\n<td><p>Active inbound<\/p><\/td>\n<td><p>Fair, fast shortlist<\/p><\/td>\n<td><p>Screen notes vs scorecard<\/p><\/td>\n<td><p>Reject \/ advance<\/p><\/td>\n<td><p>Application flow, not cold sequence<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Passive cold<\/p><\/td>\n<td><p>Earn a reply without brand damage<\/p><\/td>\n<td><p>Find + rank + draft<\/p><\/td>\n<td><p>Contact list + send edit<\/p><\/td>\n<td><p>3-touch default<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>CRM reactivation<\/p><\/td>\n<td><p>Restart a known conversation<\/p><\/td>\n<td><p>Summarize past stage + draft<\/p><\/td>\n<td><p>Eligibility check<\/p><\/td>\n<td><p>1-2 touches then stop<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Referral assist<\/p><\/td>\n<td><p>Help employees introduce well<\/p><\/td>\n<td><p>Suggest who + why in plain language<\/p><\/td>\n<td><p>Employee sends intro<\/p><\/td>\n<td><p>One intro + recruiter follow<\/p><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n<hr>\n\n<h2 id=\"sourcing-fit-scorecard\">Sourcing Fit Scorecard (no demographic proxies)<\/h2>\n<p>Primary backlink asset: a <strong>Sourcing Fit Scorecard<\/strong> for outbound ranking (not protected-class inference). Score 0-2 per dimension; multiply by weight. Any must-have at 0 means do not source this week unless a human logs an override. Publish weights on the requisition so sourcers score the same way.<\/p>\n\n<table class=\"i10x-table\">\n<tbody>\n<tr>\n<th><p>Dimension<\/p><\/th>\n<th><p>Weight<\/p><\/th>\n<th><p>0 (fail)<\/p><\/th>\n<th><p>1 (partial)<\/p><\/th>\n<th><p>2 (strong)<\/p><\/th>\n<th><p>Banned proxies (never score)<\/p><\/th>\n<\/tr>\n<tr>\n<td><p><strong>Core craft match<\/strong><\/p><\/td>\n<td><p>5<\/p><\/td>\n<td><p>Wrong function or no relevant craft<\/p><\/td>\n<td><p>Adjacent craft, weak evidence<\/p><\/td>\n<td><p>Direct craft with recent evidence<\/p><\/td>\n<td><p>School name prestige as a substitute for craft<\/p><\/td>\n<\/tr>\n<tr>\n<td><p><strong>Level and scope<\/strong><\/p><\/td>\n<td><p>4<\/p><\/td>\n<td><p>Clearly under or over level with no path<\/p><\/td>\n<td><p>Close level, unclear ownership<\/p><\/td>\n<td><p>Scope matches role (IC\/lead\/manager)<\/p><\/td>\n<td><p>Age, graduation year as age proxy<\/p><\/td>\n<\/tr>\n<tr>\n<td><p><strong>Domain or product context<\/strong><\/p><\/td>\n<td><p>3<\/p><\/td>\n<td><p>No transferable domain<\/p><\/td>\n<td><p>Partial domain overlap<\/p><\/td>\n<td><p>Relevant product, industry, or user type<\/p><\/td>\n<td><p>Employer \u201cbrand coolness\u201d as a proxy for skill<\/p><\/td>\n<\/tr>\n<tr>\n<td><p><strong>Stack or method evidence<\/strong><\/p><\/td>\n<td><p>4<\/p><\/td>\n<td><p>Missing required tools or methods<\/p><\/td>\n<td><p>Related tools only<\/p><\/td>\n<td><p>Required stack or methods shown in work<\/p><\/td>\n<td><p>Hobby lists or social interests<\/p><\/td>\n<\/tr>\n<tr>\n<td><p><strong>Location \/ work model fit<\/strong><\/p><\/td>\n<td><p>3<\/p><\/td>\n<td><p>Hard conflict with true constraint<\/p><\/td>\n<td><p>Unclear or needs confirmation<\/p><\/td>\n<td><p>Meets stated location or remote rules<\/p><\/td>\n<td><p>Neighborhood prestige or commute stereotypes<\/p><\/td>\n<\/tr>\n<tr>\n<td><p><strong>Work authorization (if truly required)<\/strong><\/p><\/td>\n<td><p>5 (gate)<\/p><\/td>\n<td><p>Known hard fail for the role<\/p><\/td>\n<td><p>Unknown (flag, do not invent)<\/p><\/td>\n<td><p>Clear match to posted requirement<\/p><\/td>\n<td><p>Name, accent, or nationality guesses<\/p><\/td>\n<\/tr>\n<tr>\n<td><p><strong>Signal of openness (optional)<\/strong><\/p><\/td>\n<td><p>2<\/p><\/td>\n<td><p>Public \u201cnot open\u201d and no warm path<\/p><\/td>\n<td><p>Neutral \/ unknown<\/p><\/td>\n<td><p>Open to work, recent job change, or mutual path<\/p><\/td>\n<td><p>Family status, photos, marital signals<\/p><\/td>\n<\/tr>\n<tr>\n<td><p><strong>Evidence quality<\/strong><\/p><\/td>\n<td><p>3<\/p><\/td>\n<td><p>Claims only, no artifacts<\/p><\/td>\n<td><p>Titles only, light detail<\/p><\/td>\n<td><p>Projects, outcomes, or public work you can cite<\/p><\/td>\n<td><p>Photo quality, \u201cenergy,\u201d or writing polish alone<\/p><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n<div class=\"i10x-callout\">\n<strong>Scorecard rules (print these next to the table)<\/strong>\n<p>1) Score evidence, not vibe. 2) Never ask the model to infer race, gender, age, disability, religion, or family status. 3) Do not use years-of-experience as a hard cutoff unless the role truly requires it for safety or regulation. 4) Version the scorecard (v1, v2) when hiring managers change must-haves. 5) A high total with a must-have zero is still a fail until a human override is logged. 6) Prefer \u201cunknown\u201d over invented certainty when profiles are thin.<\/p>\n<\/div>\n\n<p><strong>Operator notes:<\/strong> cite artifacts for craft; map level to ownership (not title inflation); lower domain weight when transferability is real; prove stack from work not headlines; use location\/authorization only when truly required. Paste the scorecard before Boolean generation (\n<a href=\"https:\/\/i10x.ai\/blog\/free-ai-recruiting-tools-2026\">free tools 2026<\/a>,\n<a href=\"https:\/\/i10x.ai\/tools\/category\/business-management\/free-ai-recruiting\">Free AI Recruiting<\/a>).<\/p>\n\n<hr>\n\n<h2 id=\"boolean-vs-semantic-vs-hybrid\">Boolean vs semantic vs hybrid search (deep decision table)<\/h2>\n<p>Sourcers still need Boolean. Models still invent confident matches. Hybrid is the default for most professional roles in 2026.<\/p>\n\n<table class=\"i10x-table\">\n<tbody>\n<tr>\n<th><p>Dimension<\/p><\/th>\n<th><p>Boolean<\/p><\/th>\n<th><p>Semantic \/ LLM assist<\/p><\/th>\n<th><p>Hybrid (recommended default)<\/p><\/th>\n<\/tr>\n<tr>\n<td><p>What it optimizes<\/p><\/td>\n<td><p>Precision, reproducibility, audit strings<\/p><\/td>\n<td><p>Recall across titles, narrative fit, ranking notes<\/p><\/td>\n<td><p>Precision gate + narrative rank inside a capped set<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Best when<\/p><\/td>\n<td><p>Exact skill, license, language, cert, stack token<\/p><\/td>\n<td><p>Titles vary, career paths are non-linear, lookalikes<\/p><\/td>\n<td><p>Almost all mid and senior professional searches<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Weak when<\/p><\/td>\n<td><p>Sparse profiles, synonym-heavy crafts<\/p><\/td>\n<td><p>Compliance needs string replay; thin text over-inference<\/p><\/td>\n<td><p>Team refuses to log either string or prompt version<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Must-have skill is exact<\/p><\/td>\n<td><p>Primary filter<\/p><\/td>\n<td><p>Only expand synonyms after Boolean<\/p><\/td>\n<td><p>Boolean gate, then semantic ranking inside the set<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Role language is messy<\/p><\/td>\n<td><p>Secondary (title OR clusters)<\/p><\/td>\n<td><p>Primary for title expansion and lookalikes<\/p><\/td>\n<td><p>LLM proposes title OR list; human locks it; Boolean runs it<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Compliance or audit trail<\/p><\/td>\n<td><p>Strong (reproducible strings)<\/p><\/td>\n<td><p>Weaker unless you log prompts and outputs<\/p><\/td>\n<td><p>Store both: string version + model notes<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Passive talent, sparse profiles<\/p><\/td>\n<td><p>Can over-filter empty profiles<\/p><\/td>\n<td><p>Can over-infer from thin text<\/p><\/td>\n<td><p>Loose Boolean, strict human sample of top N<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>High-volume commodity role<\/p><\/td>\n<td><p>Fast volume filter<\/p><\/td>\n<td><p>Helpful for ranking, risky for auto-contact<\/p><\/td>\n<td><p>Boolean volume, scorecard rank, human send<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Niche or senior search<\/p><\/td>\n<td><p>Precision strings for stack<\/p><\/td>\n<td><p>Narrative fit and career-path reasoning<\/p><\/td>\n<td><p>Small Boolean set, deep semantic notes, human shortlist<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Typical failure<\/p><\/td>\n<td><p>False negatives from synonym miss<\/p><\/td>\n<td><p>Confident wrong matches; pedigree flavor<\/p><\/td>\n<td><p>Hybrid without human gate becomes automated spam<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>When NOT to use<\/p><\/td>\n<td><p>As only method when titles are chaotic<\/p><\/td>\n<td><p>As sole auto-contact engine<\/p><\/td>\n<td><p>When no scorecard exists yet (fix intake first)<\/p><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n<p><strong>Practical hybrid workflow (15 minutes per req).<\/strong><\/p>\n<ol>\n<li>Lock the Sourcing Fit Scorecard with the hiring manager (must-haves first).<\/li>\n<li>Ask the model for three Boolean variants: tight, medium, wide. Edit for accuracy. Never paste unedited strings into a mass action.<\/li>\n<li>Run Boolean on your licensed sources. Export a capped list (for example top 50 to 100).<\/li>\n<li>Ask the model to rank that list against the scorecard with evidence bullets only. Ban demographic inference in the system prompt.<\/li>\n<li>Human selects the contact list for the week. AI drafts messages. Human edits and sends.<\/li>\n<\/ol>\n<p>This hybrid keeps reproducibility (Boolean) and narrative understanding (semantic) without handing the send button to a black box.<\/p>\n\n<hr>\n\n<h2 id=\"channels-playbook\">Channels playbook: LinkedIn, GitHub, portfolios, boards, CRM, referrals<\/h2>\n<p>AI multiplies what licensed channels allow. Choose channels by craft and evidence density, not tool fashion.<\/p>\n\n<table class=\"i10x-table\">\n<tbody>\n<tr>\n<th><p>Channel<\/p><\/th>\n<th><p>Best for<\/p><\/th>\n<th><p>AI assist<\/p><\/th>\n<th><p>Human gate<\/p><\/th>\n<th><p>Watch-outs<\/p><\/th>\n<\/tr>\n<tr>\n<td><p><strong>LinkedIn<\/strong><\/p><\/td>\n<td><p>Most professional roles; title clustering<\/p><\/td>\n<td><p>Boolean variants, lookalikes, InMail drafts<\/p><\/td>\n<td><p>Weekly contact list; edit T1<\/p><\/td>\n<td><p>Platform terms; InMail fatigue; prestige bias in headlines<\/p><\/td>\n<\/tr>\n<tr>\n<td><p><strong>GitHub \/ code hosts<\/strong><\/p><\/td>\n<td><p>Engineering, data, infra with public code<\/p><\/td>\n<td><p>Summarize repos vs stack must-haves; rank by evidence<\/p><\/td>\n<td><p>Verify license and real ownership of code<\/p><\/td>\n<td><p>Activity \u2260 job readiness; ignore stars as prestige proxy<\/p><\/td>\n<\/tr>\n<tr>\n<td><p><strong>Portfolios \/ personal sites<\/strong><\/p><\/td>\n<td><p>Design, product, content, research<\/p><\/td>\n<td><p>Extract case outcomes into scorecard rows<\/p><\/td>\n<td><p>Confirm role and level from case depth<\/p><\/td>\n<td><p>Scraping ethics; do not paste private contact data into consumer chat<\/p><\/td>\n<\/tr>\n<tr>\n<td><p><strong>Job boards \/ talent pools<\/strong><\/p><\/td>\n<td><p>Active seekers; niche boards<\/p><\/td>\n<td><p>Rank applicants and saved searches<\/p><\/td>\n<td><p>Fair screen same as inbound<\/p><\/td>\n<td><p>Do not double-message people already in ATS mid-process<\/p><\/td>\n<\/tr>\n<tr>\n<td><p><strong>CRM reactivation<\/strong><\/p><\/td>\n<td><p>Silver medalists; past finalists<\/p><\/td>\n<td><p>Summarize last stage, reasons, reopen draft<\/p><\/td>\n<td><p>Check DNC and recontact rules<\/p><\/td>\n<td><p>Stale notes; prior \u201cno\u201d still counts<\/p><\/td>\n<\/tr>\n<tr>\n<td><p><strong>Employee referral AI assist<\/strong><\/p><\/td>\n<td><p>Warm intros with context<\/p><\/td>\n<td><p>Suggest who might fit and why in plain language for the employee<\/p><\/td>\n<td><p>Employee owns the ask; recruiter follows<\/p><\/td>\n<td><p>Do not pressure employees to spam their network<\/p><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n<p><strong>GitHub:<\/strong> Prefer ownership signals (sustained commits, design docs) over stars. Cite real modules in T1; humans verify authorship. <strong>CRM reactivation:<\/strong> sort by last positive stage; open with what changed; ask permission; one thoughtful note beats a long sequence.<\/p>\n\n<hr>\n\n<h2 id=\"three-touch-outreach\">3-Touch Outreach Sequence (template with example copy)<\/h2>\n<p>The <strong>3-Touch Outreach Sequence<\/strong> is the cold-passive default: three value-bearing touches, then stop or switch to a warm path (referral, event, mutual intro).<\/p>\n\n<table class=\"i10x-table\">\n<tbody>\n<tr>\n<th><p>Touch<\/p><\/th>\n<th><p>Timing<\/p><\/th>\n<th><p>Goal<\/p><\/th>\n<th><p>Must include<\/p><\/th>\n<th><p>Must avoid<\/p><\/th>\n<\/tr>\n<tr>\n<td><p><strong>T1: Specific open<\/strong><\/p><\/td>\n<td><p>Day 0<\/p><\/td>\n<td><p>Earn a reply or soft no<\/p><\/td>\n<td><p>One real profile signal, one role outcome, clear ask (15-min chat or \u201cnot now\u201d is fine)<\/p><\/td>\n<td><p>Flattery about prestige, fake \u201cI came across your profile\u201d with zero detail, salary bait if you cannot stand behind it<\/p><\/td>\n<\/tr>\n<tr>\n<td><p><strong>T2: Value follow-up<\/strong><\/p><\/td>\n<td><p>Day 4-6<\/p><\/td>\n<td><p>Add new information<\/p><\/td>\n<td><p>One new fact (team problem, stack, impact, interview process honesty) plus easy exit<\/p><\/td>\n<td><p>Guilt language, \u201cjust bumping this,\u201d repeating T1 word for word<\/p><\/td>\n<\/tr>\n<tr>\n<td><p><strong>T3: Graceful close<\/strong><\/p><\/td>\n<td><p>Day 10-12<\/p><\/td>\n<td><p>Close loop, leave door open<\/p><\/td>\n<td><p>Permission to re-contact later, optional referral ask, thanks<\/p><\/td>\n<td><p>Fake deadline pressure, third-party spam CC<\/p><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n<p><strong>T1 skeleton (edit heavily):<\/strong><br>\n\u201cHi [Name], I noticed [specific project, talk, open-source, or product outcome]. We are hiring a [role] to [one outcome in plain language]. Your [skill\/evidence] maps to [scorecard line]. Open to a short conversation, or a simple no if timing is wrong?\u201d<\/p>\n<p><strong>T2 skeleton:<\/strong><br>\n\u201cQuick add: the team\u2019s main constraint right now is [problem]. Interview process is [stages]. If useful, happy to share the scorecard we use so you can self-select. Either way, appreciate your time.\u201d<\/p>\n<p><strong>T3 skeleton:<\/strong><br>\n\u201cLast note from me on this role. I\u2019ll close the loop on my side. If you ever want a look when [condition], or know someone who thrives on [problem], I\u2019m easy to reach. Thanks again.\u201d<\/p>\n\n<div class=\"i10x-callout\">\n<strong>AI drafting rules for outreach<\/strong>\n<p>Feed only non-sensitive public signals you would be comfortable showing the candidate. Require the model to cite the signal it used. Cap length (for example under 120 words for T1). Human review every first touch for executive, diversity-critical, or regulated roles. Log channel and consent constraints for your market. Never invent mutual connections, funding claims, or \u201cfast promotion\u201d paths.<\/p>\n<\/div>\n\n<hr>\n\n<h2 id=\"five-touch-outreach\">5-Touch Outreach Sequence (when longer is justified)<\/h2>\n<p>Use 5 touches only for hard-to-fill roles when each touch adds value. Not a default. Frequency caps still apply.<\/p>\n\n<table class=\"i10x-table\">\n<tbody>\n<tr>\n<th><p>Touch<\/p><\/th>\n<th><p>Timing<\/p><\/th>\n<th><p>Content job<\/p><\/th>\n<th><p>Example angle<\/p><\/th>\n<\/tr>\n<tr>\n<td><p>T1<\/p><\/td>\n<td><p>Day 0<\/p><\/td>\n<td><p>Specific open<\/p><\/td>\n<td><p>Cite one real project; one outcome; soft ask<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>T2<\/p><\/td>\n<td><p>Day 4<\/p><\/td>\n<td><p>Problem depth<\/p><\/td>\n<td><p>One technical or business constraint the hire will own<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>T3<\/p><\/td>\n<td><p>Day 9<\/p><\/td>\n<td><p>Process honesty<\/p><\/td>\n<td><p>Stages, who they meet, decision speed you can keep<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>T4<\/p><\/td>\n<td><p>Day 16<\/p><\/td>\n<td><p>Social proof without prestige worship<\/p><\/td>\n<td><p>Team craft story, not \u201ctop company\u201d flattery<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>T5<\/p><\/td>\n<td><p>Day 24<\/p><\/td>\n<td><p>Close + referral<\/p><\/td>\n<td><p>Stop cold sequence; invite future ping or referral<\/p><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n<p><strong>When NOT to run 5 touches:<\/strong> fragile brand; hard no already; spam complaints; legal\/frequency limits; weak scorecard fit (volume is not a substitute for fit).<\/p>\n\n<hr>\n\n<h2 id=\"copy-paste-sourcing-prompt-pack\">Copy-paste template: Sourcing Prompt Pack<\/h2>\n<p>Approved workspace only. Keep the ban list intact.<\/p>\n<p><strong>System intent:<\/strong><br>\n\u201cYou assist with candidate sourcing. Score only against the provided Sourcing Fit Scorecard. Quote public evidence. If evidence is missing, mark unknown. Do not infer race, gender, age, disability, religion, family status, or nationality from names or photos. Do not use school prestige or employer brand coolness as skill. Do not invent contact history. Output structured notes only.\u201d<\/p>\n<p><strong>User packet:<\/strong><br>\n\u201cRole: [title]. Scorecard v[N]: [paste]. Must-haves: [list]. Non-criteria: [list]. Channel: [LinkedIn\/GitHub\/CRM]. Task: (1) propose tight\/medium\/wide Boolean, (2) rank the pasted profiles 1-N with scorecard rows and evidence quotes, (3) draft T1 under 120 words citing one real signal. Flag any profile with must-have zero.\u201d<\/p>\n<p><strong>Recruiter checklist before send:<\/strong><br>\nSignal is real; claims about company are true; no sensitive data in prompt that was not needed; person is not mid-process elsewhere; DNC checked; message sounds like a human on your team would send it.<\/p>\n\n<hr>\n\n<h2 id=\"reply-rate-tracking-fields\">Reply-rate tracking spreadsheet fields<\/h2>\n<p>Track a small field set per touch and channel so you fix scorecards and copy, not vanity views.<\/p>\n\n<table class=\"i10x-table\">\n<tbody>\n<tr>\n<th><p>Field<\/p><\/th>\n<th><p>Type<\/p><\/th>\n<th><p>Why it matters<\/p><\/th>\n<\/tr>\n<tr>\n<td><p>req_id \/ role<\/p><\/td>\n<td><p>Text<\/p><\/td>\n<td><p>Attribute outcomes to a scorecard version<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>scorecard_version<\/p><\/td>\n<td><p>Text<\/p><\/td>\n<td><p>Detect criteria drift<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>channel<\/p><\/td>\n<td><p>Enum<\/p><\/td>\n<td><p>LinkedIn, email, GitHub, referral, CRM<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>segment<\/p><\/td>\n<td><p>Enum<\/p><\/td>\n<td><p>Passive cold, active, reactivation, referral<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>fit_score_total<\/p><\/td>\n<td><p>Number<\/p><\/td>\n<td><p>Compare reply quality by fit band<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>must_have_fail<\/p><\/td>\n<td><p>Yes\/No<\/p><\/td>\n<td><p>Should be No for contacted people<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>touch_number<\/p><\/td>\n<td><p>1-5<\/p><\/td>\n<td><p>Attribute reply to sequence step<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>sent_at<\/p><\/td>\n<td><p>Date<\/p><\/td>\n<td><p>Cadence and frequency caps<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>replied<\/p><\/td>\n<td><p>Yes\/No<\/p><\/td>\n<td><p>Raw reply rate<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>reply_type<\/p><\/td>\n<td><p>Enum<\/p><\/td>\n<td><p>Positive, neutral, negative, OOO, unsubscribe<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>positive_reply<\/p><\/td>\n<td><p>Yes\/No<\/p><\/td>\n<td><p>Primary quality metric<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>interview_booked<\/p><\/td>\n<td><p>Yes\/No<\/p><\/td>\n<td><p>Conversion past vanity reply<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>complaint_or_spam_flag<\/p><\/td>\n<td><p>Yes\/No<\/p><\/td>\n<td><p>Stop-the-line signal<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>human_edited_message<\/p><\/td>\n<td><p>Yes\/No<\/p><\/td>\n<td><p>Audit AI vs human quality<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>notes<\/p><\/td>\n<td><p>Text<\/p><\/td>\n<td><p>Why this person, what signal was used<\/p><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n<p><strong>Definitions:<\/strong> positive = interest, referral, or \u201ctell me more\u201d; neutral = maybe later; negative = hard no. Unsubscribe\/complaint ends the sequence and updates DNC. Report positive reply rate and interview conversion weekly.<\/p>\n\n<hr>\n\n<h2 id=\"privacy-frequency-consent\">Privacy, frequency caps, GDPR and CAN-SPAM high-level<\/h2>\n<p>Operational orientation only, <strong>not legal advice<\/strong>. Laws and platform terms vary. Use counsel for your facts.<\/p>\n<ul>\n<li><strong>Licensed sources only:<\/strong> platform terms for LinkedIn, GitHub, boards, CRM. Violating terms is not innovation.<\/li>\n<li><strong>CRM before cold:<\/strong> prior apps, DNC flags, referrals first.<\/li>\n<li><strong>Minimize prompt data:<\/strong> strip contacts when unneeded; use approved tools, not personal consumer accounts.<\/li>\n<li><strong>Retention and frequency:<\/strong> define note life for non-applicants; default 3 cold touches; no re-open for 90 days without new reason and human approval.<\/li>\n<li><strong>GDPR-oriented (high-level, not legal advice):<\/strong> know legal basis; honor access\/deletion; no secondary training use unless contract and policy allow.<\/li>\n<li><strong>CAN-SPAM-oriented email (high-level, US context):<\/strong> accurate sender, truthful subjects, opt-out where required, honor opt-outs. Platform messages still need platform rules.<\/li>\n<li><strong>Truth in advertising:<\/strong> no invented headcount, funding, or promotion claims.<\/li>\n<\/ul>\n<p>For high-level AI governance context (not legal advice): under the EU AI Act, Annex III treats AI used for recruitment and selection (including targeted job ads and systems that analyse or filter applications or evaluate candidates) as a high-risk employment use case. High-risk obligations continue to phase in through 2026-2027 depending on system type and role. Outbound ranking and automated filtering that affect access to work deserve documentation, human oversight, and counsel review if you are in scope. See\n<a href=\"https:\/\/i10x.ai\/blog\/ethical-ai-recruiting\">ethical AI recruiting<\/a>\nfor a practical checklist.<\/p>\n\n<hr>\n\n<h2 id=\"diversity-sourcing-without-proxies\">Diversity sourcing without illegal proxies (careful framing)<\/h2>\n<p>Widening who you consider is legitimate. Using protected-class proxies as scoring features is not. <strong>Do:<\/strong> widen skill synonyms; drop false must-haves; multi-channel search; score evidence not pedigree; sample exclusions; structured scorecards. <strong>Do not:<\/strong> score names, photos, or schools as diversity signals; prompt demographic preference; use zip\/hobby proxies; invent group-specific thresholds without counsel. Score the job, expand surface, audit exclusions (not legal advice).<\/p>\n\n<hr>\n\n<h2 id=\"free-stack-vs-paid\">Free stack options and when to pay<\/h2>\n<table class=\"i10x-table\">\n<tbody>\n<tr>\n<th><p>Choose free \/ freemium when<\/p><\/th>\n<th><p>Choose paid when<\/p><\/th>\n<\/tr>\n<tr>\n<td><p>You are piloting one role and can afford human review of every send<\/p><\/td>\n<td><p>Volume requires CRM sequencing, deliverability tools, or ATS-native rankers with logging<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Two model UIs + spreadsheet scorecard are enough for ranking notes<\/p><\/td>\n<td><p>You need SSO, DPA, training opt-out, and exportable audit logs<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Brand risk of mistakes is high and you want slower, careful outreach<\/p><\/td>\n<td><p>Multiple recruiters need shared memory of who was contacted<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Engineering can self-source via GitHub without a sales-style sequencer<\/p><\/td>\n<td><p>You must integrate calendar, ATS stages, and compliance fields<\/p><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p>Start free via\n<a href=\"https:\/\/i10x.ai\/blog\/free-ai-recruiting-tools-2026\">free AI recruiting tools for 2026<\/a>\nand\n<a href=\"https:\/\/i10x.ai\/tools\/category\/business-management\/free-ai-recruiting\">Free AI Recruiting on i10X<\/a>.\nPay when process discipline is ready. Features without human gates become shadow automation.<\/p>\n\n<hr>\n\n<h2 id=\"worked-example-sourcing\">Worked example: backend hire (before \/ after)<\/h2>\n<p><strong>Context (anonymized).<\/strong> 40-person B2B SaaS, remote EU backend engineer, distributed systems + Go\/Java. TTF pressure vs SHRM nonexec medians (~44 days 2025 reports; 39 days SHRM 2026 exec benchmarking). Cost context SHRM 2025 ~$5,475 nonexec average, not a spam justification.<\/p>\n<p><strong>Before:<\/strong> JD paste into a chatbot for \u201c50 ideal profiles,\u201d five near-identical touches, prestige flattery. Weak positive replies; two interviews failed stack must-haves; HM lost trust.<\/p>\n<p><strong>After:<\/strong> scorecard v1 with systems + Go\/Java must-haves; prestige banned. Tight\/medium\/wide Boolean; cap 80; evidence-only rank; human list of 22; 3-touch with real signals.<\/p>\n<p><strong>Illustrative outcomes (scenario, not a published study statistic).<\/strong> Positive replies improved with real-work signals. Prestigious-but-weak-stack profiles were not contacted. One strong non-linear Go candidate advanced. Criteria matched later\n<a href=\"https:\/\/i10x.ai\/blog\/ai-resume-screening\">screening<\/a>.\nStyle-risk awareness from the\n<a href=\"https:\/\/i10x.ai\/blog\/ai-cv-bias\">i10X AI CV bias study<\/a>\n(up to 42 pp hire-rate gap; 1,576 points; 100 profiles; 29 pt evaluator gap) kept notes evidence-first.<\/p>\n\n<hr>\n\n<h2 id=\"when-not-to-use-ai-sourcing\">When NOT to use AI candidate sourcing (or not yet)<\/h2>\n<ul>\n<li>No scorecard exists and the hiring manager cannot name must-haves in 30 minutes.<\/li>\n<li>You plan to auto-send hundreds of messages without human edit or frequency caps.<\/li>\n<li>The only \u201csignal\u201d you have is school or employer brand.<\/li>\n<li>Legal, works council, or platform constraints on outbound are unclear for your market and you have not checked.<\/li>\n<li>The role is confidential C-level with a closed search firm already running a sensitive process.<\/li>\n<li>Your CRM is a mess of DNC flags and you would re-contact people who asked to stop.<\/li>\n<\/ul>\n<p>Fix intake and data hygiene first. AI multiplies quality and garbage equally.<\/p>\n\n<hr>\n\n<h2 id=\"metrics-for-ai-sourcing\">Metrics for AI sourcing (baseline, 30 days, 90 days)<\/h2>\n<table class=\"i10x-table\">\n<tbody>\n<tr>\n<th><p>Metric<\/p><\/th>\n<th><p>Baseline (week 0)<\/p><\/th>\n<th><p>30 days<\/p><\/th>\n<th><p>90 days<\/p><\/th>\n<\/tr>\n<tr>\n<td><p>Scorecard signed before first contact<\/p><\/td>\n<td><p>% of reqs with v1 locked<\/p><\/td>\n<td><p>100% on pilot req type<\/p><\/td>\n<td><p>Standard for all sourced roles<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Positive reply rate<\/p><\/td>\n<td><p>Measure last 20 cold sends<\/p><\/td>\n<td><p>Improve via signal-rich T1<\/p><\/td>\n<td><p>Segment by channel and fit band<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Interview show rate from sourced<\/p><\/td>\n<td><p>Current<\/p><\/td>\n<td><p>Track vs inbound<\/p><\/td>\n<td><p>Stable HM satisfaction<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Complaint \/ unsubscribe rate<\/p><\/td>\n<td><p>Near zero target<\/p><\/td>\n<td><p>Any spike pauses sequences<\/p><\/td>\n<td><p>Quarterly review with brand<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Time-to-first-shortlist from source<\/p><\/td>\n<td><p>Current<\/p><\/td>\n<td><p>Where gen AI time savings should appear (LinkedIn: ~20% workweek for TA using gen AI)<\/p><\/td>\n<td><p>Do not trade fairness for speed<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Time-to-fill context<\/p><\/td>\n<td><p>Your median<\/p><\/td>\n<td><p>Compare carefully to SHRM ~44 days (2025 cycle reports) \/ 39 days nonexec (2026 exec benchmarking)<\/p><\/td>\n<td><p>Lagging; never sole KPI<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Cost-per-hire context<\/p><\/td>\n<td><p>Your fully loaded cost<\/p><\/td>\n<td><p>SHRM 2025 averages ~$5,475 nonexec \/ ~$35,879 exec as external reference only<\/p><\/td>\n<td><p>Rework from bad sourced shortlists counts as cost<\/p><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p>Avoid vanity metrics (\u201cAI messages generated\u201d) without quality checks.<\/p>\n\n<hr>\n\n<h2 id=\"common-failure-modes\">Failure modes and anti-patterns<\/h2>\n<table class=\"i10x-table\">\n<tbody>\n<tr>\n<th><p>Failure<\/p><\/th>\n<th><p>What it looks like<\/p><\/th>\n<th><p>Fix<\/p><\/th>\n<\/tr>\n<tr>\n<td><p>Scorecard by JD paste<\/p><\/td>\n<td><p>Model ranks on buzzwords<\/p><\/td>\n<td><p>Rewrite must-haves as evidence tests<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Pedigree overweight<\/p><\/td>\n<td><p>Big logos rank above craft<\/p><\/td>\n<td><p>Ban employer prestige as a dimension<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Semantic over-inference<\/p><\/td>\n<td><p>Thin profiles get \u201cconfident\u201d scores<\/p><\/td>\n<td><p>Require quotes; score \u201cunknown\u201d as 1 max<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Sequence spam<\/p><\/td>\n<td><p>Five identical bumps for every role<\/p><\/td>\n<td><p>3-touch default; 5-touch only when justified<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>No human gate<\/p><\/td>\n<td><p>Auto-send to hundreds<\/p><\/td>\n<td><p>Human approve contact lists weekly<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Style bias bleed-in<\/p><\/td>\n<td><p>Polish beats substance in notes<\/p><\/td>\n<td><p>Evidence-only ranking; read\n<a href=\"https:\/\/i10x.ai\/blog\/ai-cv-bias\">CV bias research<\/a><\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Channel mess<\/p><\/td>\n<td><p>Same person messaged twice in two tools<\/p><\/td>\n<td><p>Single system of record for contact history<\/p><\/td>\n<\/tr>\n<tr>\n<td><p>Proxy \u201cdiversity\u201d scoring<\/p><\/td>\n<td><p>Model told to prefer names or schools<\/p><\/td>\n<td><p>Expand channels and remove false must-haves; no demographic proxies<\/p><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n<hr>\n\n<h2 id=\"faq\">Frequently asked questions<\/h2>\n<div class=\"i10x-faq\">\n<p><strong>What is AI candidate sourcing?<\/strong><br>\nIt is the use of AI to find and prioritize potential candidates (often passive), score them against a fit scorecard, and draft personalized outreach, with humans owning contact decisions and relationship risk.<\/p>\n<p><strong>Is AI sourcing better than Boolean search alone?<\/strong><br>\nNot automatically. Boolean is reproducible and strong for exact must-haves. Semantic search helps with title variation and ranking. The hybrid table in this guide is the practical default for most teams.<\/p>\n<p><strong>Boolean vs AI search: which should I start with?<\/strong><br>\nStart with a scorecard, then Boolean for must-have tokens, then AI ranking inside a capped export. Starting with pure AI lookalikes without a gate produces confident wrong matches.<\/p>\n<p><strong>What reply rates should I expect?<\/strong><br>\nNo universal benchmark in the allowed stats here. Measure positive reply rate by channel and fit band. Optimize signal and sequence length before volume. LinkedIn associates heavy AI-Assisted Messaging with about 9% higher likelihood of a quality hire (most vs least), which supports assistance, not spam.<\/p>\n<p><strong>How do I avoid spam risk with AI outreach?<\/strong><br>\nCap sequences (3-touch default), require real profile signals, human-edit T1, honor opt-outs and DNC, avoid purchased lists you cannot justify, and track complaint flags as a stop-the-line metric.<\/p>\n<p><strong>How do I source on GitHub with AI?<\/strong><br>\nMap must-haves to concrete code evidence, use AI to summarize repos against the scorecard, verify authorship, and write T1 that cites a real module or design choice. Do not rank on stars or follower counts as prestige.<\/p>\n<p><strong>How does CRM reactivation work with AI?<\/strong><br>\nFilter eligible past candidates, summarize last stage and outcome, draft a short permission-based re-open note, and stop if there is no reply. Always check do-not-contact and recontact rules first.<\/p>\n<p><strong>How do I avoid bias when sourcing with AI?<\/strong><br>\nUse the Sourcing Fit Scorecard with banned proxies, evidence-only scoring, no demographic inference, human review of contact lists, and periodic sampling of who was excluded. Screening-side style risk is documented in the\n<a href=\"https:\/\/i10x.ai\/blog\/ai-cv-bias\">i10X study<\/a>.<\/p>\n<p><strong>How many outreach touches should AI write?<\/strong><br>\nThree is enough for most cold sequences. Five is for hard-to-fill roles where each touch adds new value. More touches without new value usually hurt brand more than they help reply rate.<\/p>\n<p><strong>Does gen AI actually save recruiter time?<\/strong><br>\nLinkedIn\u2019s Future of Recruiting 2025 reports that TA professionals using gen AI save about 20% of their workweek on average. Savings depend on process design, not tool logos.<\/p>\n<p><strong>Can AI auto-send InMails or emails?<\/strong><br>\nTechnically some stacks allow it. Operationally, auto-send without human review is a brand and compliance risk. Prefer human approval, especially for executive and regulated roles.<\/p>\n<p><strong>Where does the EU AI Act fit?<\/strong><br>\nRecruitment and selection AI (targeted ads, filtering applications, evaluating candidates) can be high-risk under Annex III. Treat this as a signal to document systems and keep oversight. Get legal advice for your situation. This is not legal advice.<\/p>\n<p><strong>What should I measure first?<\/strong><br>\nPositive reply rate, interviews from sourced candidates, complaint rate, and quality feedback from hiring managers. Time-to-fill medians (SHRM reports 39 calendar days nonexecutive in the 2026 executives benchmarking release; ~44 days in widely reported 2025 cycle figures) are lagging; do not optimize only for speed.<\/p>\n<p><strong>Passive vs active: should sequences differ?<\/strong><br>\nYes. Passive cold defaults to shorter, higher-signal sequences. Active and CRM reactivation use different permission context. Do not run the same five-step blast on every segment.<\/p>\n<\/div>\n\n<hr>\n\n<div class=\"i10x-callout\">\n<strong>Key takeaways<\/strong>\n<p>Write fit as evidence. Use the Sourcing Fit Scorecard without demographic proxies, hybrid Boolean\/semantic search, 3-Touch default (5-Touch only when justified). Track positive replies and complaints. SHRM and LinkedIn show adoption and time savings; they do not excuse spam. Keep humans on the send button and align criteria with screening.<\/p>\n<\/div>\n\n<hr>\n\n<div class=\"i10x-cta\">\n<h3 id=\"run-ai-sourcing-on-i10x\">Run AI candidate sourcing on i10X<\/h3>\n<p>Turn scorecards, shortlists, and outreach drafts into a live workflow with human review checkpoints.<\/p>\n<p><a href=\"https:\/\/i10x.ai\/tools\/category\/business-management\/free-ai-recruiting\" rel=\"noopener\" target=\"_blank\">Launch Free AI Recruiting \u2192<\/a><\/p>\n<p>Related reading:\n<a href=\"https:\/\/i10x.ai\/blog\/ai-recruiting-workflow\">AI recruiting workflow<\/a>,\n<a href=\"https:\/\/i10x.ai\/blog\/ai-resume-screening\">AI resume screening<\/a>,\n<a href=\"https:\/\/i10x.ai\/blog\/multi-model-ai-screening\">multi-model AI screening<\/a>,\n<a href=\"https:\/\/i10x.ai\/blog\/ethical-ai-recruiting\">ethical AI recruiting<\/a>,\n<a href=\"https:\/\/i10x.ai\/blog\/ai-recruiting-guide\">AI recruiting guide<\/a>.<\/p>\n<\/div>\n\n<div class=\"i10x-sources\">\n<strong>Sources<\/strong>\n<ol>\n<li>SHRM 2025 Talent Trends: AI use in HR tasks 43% in 2025, up from 26% in 2024 (adoption context for AI in HR, including sourcing-adjacent work).<\/li>\n<li>LinkedIn Future of Recruiting 2025: 37% of orgs integrating or experimenting with gen AI in hiring (up from 27%); TA pros using gen AI report about 20% of workweek saved on average.<\/li>\n<li>LinkedIn: companies whose recruiters use AI-Assisted Messaging most are about 9% more likely to make a quality hire vs those who use it least (outreach quality signal).<\/li>\n<li>SHRM 2025 Recruiting Benchmarking (widely reported): median time-to-fill around 44 days for nonexecutive roles (speed pressure context for sourcing).<\/li>\n<li>SHRM 2026 Recruiting Executives Benchmarking: median time-to-fill 39 calendar days nonexecutive (updated external benchmark).<\/li>\n<li>SHRM 2025 Benchmarking Report averages (press): about $5,475 cost-per-hire nonexecutive, $35,879 executive (averages; medians can differ in other SHRM releases).<\/li>\n<li>EU AI Act Annex III: recruitment and selection AI as high-risk employment use cases, including targeted job ads and systems that analyse, filter, or evaluate candidates (high-level; not legal advice).<\/li>\n<li>i10X Research (June 2026), <a href=\"https:\/\/i10x.ai\/blog\/ai-cv-bias\">AI resume style and screening outcomes<\/a>: up to 42 pp hire-rate gap by writing style; 1,576 valid data points; 100 candidate profiles; largest single-evaluator score gap 29 points (style risk when sourcing notes rank polish).<\/li>\n<li>i10X silo used for process links:\n<a href=\"https:\/\/i10x.ai\/blog\/ai-recruiting-guide\">AI recruiting guide<\/a>,\n<a href=\"https:\/\/i10x.ai\/blog\/ai-recruiting-workflow\">AI recruiting workflow<\/a>,\n<a href=\"https:\/\/i10x.ai\/blog\/multi-model-ai-screening\">multi-model AI screening<\/a>,\n<a href=\"https:\/\/i10x.ai\/blog\/free-ai-recruiting-tools-2026\">free AI recruiting tools 2026<\/a>.<\/li>\n<\/ol>\n<\/div>\n\n<\/div>\n\n","protected":false},"excerpt":{"rendered":"<p>Learn AI candidate sourcing with a Sourcing Fit Scorecard, Boolean vs semantic table, and a 3-touch outreach sequence you can run this week.<\/p>\n","protected":false},"author":5,"featured_media":347,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[10,29],"tags":[],"class_list":["post-338","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai","category-ai-recruiting"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.8 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>AI Candidate Sourcing: Scorecards and Outreach That Work - i10X Blog<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/i10x.ai\/blog\/ai-candidate-sourcing\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"AI Candidate Sourcing: Scorecards and Outreach That Work - i10X Blog\" \/>\n<meta property=\"og:description\" content=\"Learn AI candidate sourcing with a Sourcing Fit Scorecard, Boolean vs semantic table, and a 3-touch outreach sequence you can run this week.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/i10x.ai\/blog\/ai-candidate-sourcing\" \/>\n<meta property=\"og:site_name\" content=\"i10X Blog\" \/>\n<meta property=\"article:published_time\" content=\"2026-08-17T12:16:41+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-08-18T07:22:53+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/i10x.ai\/blog\/wp-content\/uploads\/2026\/08\/ai-candidate-sourcing-abstract-network-search-featured.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1536\" \/>\n\t<meta property=\"og:image:height\" content=\"864\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Christopher Ort\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Christopher Ort\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"20 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/ai-candidate-sourcing#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/ai-candidate-sourcing\"},\"author\":{\"name\":\"Christopher Ort\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/#\\\/schema\\\/person\\\/c5af13ca4e2bbda197660fab76672060\"},\"headline\":\"AI Candidate Sourcing: Scorecards and Outreach That Work\",\"datePublished\":\"2026-08-17T12:16:41+00:00\",\"dateModified\":\"2026-08-18T07:22:53+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/ai-candidate-sourcing\"},\"wordCount\":4337,\"commentCount\":0,\"image\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/ai-candidate-sourcing#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/08\\\/ai-candidate-sourcing-abstract-network-search-featured.png\",\"articleSection\":[\"AI\",\"AI Recruiting\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/i10x.ai\\\/blog\\\/ai-candidate-sourcing#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/ai-candidate-sourcing\",\"url\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/ai-candidate-sourcing\",\"name\":\"AI Candidate Sourcing: Scorecards and Outreach That Work - i10X Blog\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/ai-candidate-sourcing#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/ai-candidate-sourcing#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/08\\\/ai-candidate-sourcing-abstract-network-search-featured.png\",\"datePublished\":\"2026-08-17T12:16:41+00:00\",\"dateModified\":\"2026-08-18T07:22:53+00:00\",\"author\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/#\\\/schema\\\/person\\\/c5af13ca4e2bbda197660fab76672060\"},\"breadcrumb\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/ai-candidate-sourcing#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/i10x.ai\\\/blog\\\/ai-candidate-sourcing\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/ai-candidate-sourcing#primaryimage\",\"url\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/08\\\/ai-candidate-sourcing-abstract-network-search-featured.png\",\"contentUrl\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/08\\\/ai-candidate-sourcing-abstract-network-search-featured.png\",\"width\":1536,\"height\":864,\"caption\":\"AI Candidate Sourcing: Scorecards and Outreach That Work\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/ai-candidate-sourcing#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/i10x.ai\\\/blog\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"AI Candidate Sourcing: Scorecards and Outreach That Work\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/#website\",\"url\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/\",\"name\":\"i10X Blog\",\"description\":\"Model comparisons, workspace guides, and practical ideas on AI productivity, agents, and multi-model work.\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/#\\\/schema\\\/person\\\/c5af13ca4e2bbda197660fab76672060\",\"name\":\"Christopher Ort\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/ea95f3291658df6863df50e0ba53ddde5c83538e2079f4b3b9b548cc92d90cca?s=96&d=mm&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/ea95f3291658df6863df50e0ba53ddde5c83538e2079f4b3b9b548cc92d90cca?s=96&d=mm&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/ea95f3291658df6863df50e0ba53ddde5c83538e2079f4b3b9b548cc92d90cca?s=96&d=mm&r=g\",\"caption\":\"Christopher Ort\"},\"url\":\"https:\\\/\\\/i10x.ai\\\/blog\\\/author\\\/christopher-ort\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"AI Candidate Sourcing: Scorecards and Outreach That Work - i10X Blog","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/i10x.ai\/blog\/ai-candidate-sourcing","og_locale":"en_US","og_type":"article","og_title":"AI Candidate Sourcing: Scorecards and Outreach That Work - i10X Blog","og_description":"Learn AI candidate sourcing with a Sourcing Fit Scorecard, Boolean vs semantic table, and a 3-touch outreach sequence you can run this week.","og_url":"https:\/\/i10x.ai\/blog\/ai-candidate-sourcing","og_site_name":"i10X Blog","article_published_time":"2026-08-17T12:16:41+00:00","article_modified_time":"2026-08-18T07:22:53+00:00","og_image":[{"width":1536,"height":864,"url":"https:\/\/i10x.ai\/blog\/wp-content\/uploads\/2026\/08\/ai-candidate-sourcing-abstract-network-search-featured.png","type":"image\/png"}],"author":"Christopher Ort","twitter_card":"summary_large_image","twitter_misc":{"Written by":"Christopher Ort","Est. reading time":"20 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/i10x.ai\/blog\/ai-candidate-sourcing#article","isPartOf":{"@id":"https:\/\/i10x.ai\/blog\/ai-candidate-sourcing"},"author":{"name":"Christopher Ort","@id":"https:\/\/i10x.ai\/blog\/#\/schema\/person\/c5af13ca4e2bbda197660fab76672060"},"headline":"AI Candidate Sourcing: Scorecards and Outreach That Work","datePublished":"2026-08-17T12:16:41+00:00","dateModified":"2026-08-18T07:22:53+00:00","mainEntityOfPage":{"@id":"https:\/\/i10x.ai\/blog\/ai-candidate-sourcing"},"wordCount":4337,"commentCount":0,"image":{"@id":"https:\/\/i10x.ai\/blog\/ai-candidate-sourcing#primaryimage"},"thumbnailUrl":"https:\/\/i10x.ai\/blog\/wp-content\/uploads\/2026\/08\/ai-candidate-sourcing-abstract-network-search-featured.png","articleSection":["AI","AI Recruiting"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/i10x.ai\/blog\/ai-candidate-sourcing#respond"]}]},{"@type":"WebPage","@id":"https:\/\/i10x.ai\/blog\/ai-candidate-sourcing","url":"https:\/\/i10x.ai\/blog\/ai-candidate-sourcing","name":"AI Candidate Sourcing: Scorecards and Outreach That Work - i10X Blog","isPartOf":{"@id":"https:\/\/i10x.ai\/blog\/#website"},"primaryImageOfPage":{"@id":"https:\/\/i10x.ai\/blog\/ai-candidate-sourcing#primaryimage"},"image":{"@id":"https:\/\/i10x.ai\/blog\/ai-candidate-sourcing#primaryimage"},"thumbnailUrl":"https:\/\/i10x.ai\/blog\/wp-content\/uploads\/2026\/08\/ai-candidate-sourcing-abstract-network-search-featured.png","datePublished":"2026-08-17T12:16:41+00:00","dateModified":"2026-08-18T07:22:53+00:00","author":{"@id":"https:\/\/i10x.ai\/blog\/#\/schema\/person\/c5af13ca4e2bbda197660fab76672060"},"breadcrumb":{"@id":"https:\/\/i10x.ai\/blog\/ai-candidate-sourcing#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/i10x.ai\/blog\/ai-candidate-sourcing"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/i10x.ai\/blog\/ai-candidate-sourcing#primaryimage","url":"https:\/\/i10x.ai\/blog\/wp-content\/uploads\/2026\/08\/ai-candidate-sourcing-abstract-network-search-featured.png","contentUrl":"https:\/\/i10x.ai\/blog\/wp-content\/uploads\/2026\/08\/ai-candidate-sourcing-abstract-network-search-featured.png","width":1536,"height":864,"caption":"AI Candidate Sourcing: Scorecards and Outreach That Work"},{"@type":"BreadcrumbList","@id":"https:\/\/i10x.ai\/blog\/ai-candidate-sourcing#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/i10x.ai\/blog"},{"@type":"ListItem","position":2,"name":"AI Candidate Sourcing: Scorecards and Outreach That Work"}]},{"@type":"WebSite","@id":"https:\/\/i10x.ai\/blog\/#website","url":"https:\/\/i10x.ai\/blog\/","name":"i10X Blog","description":"Model comparisons, workspace guides, and practical ideas on AI productivity, agents, and multi-model work.","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/i10x.ai\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Person","@id":"https:\/\/i10x.ai\/blog\/#\/schema\/person\/c5af13ca4e2bbda197660fab76672060","name":"Christopher Ort","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/secure.gravatar.com\/avatar\/ea95f3291658df6863df50e0ba53ddde5c83538e2079f4b3b9b548cc92d90cca?s=96&d=mm&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/ea95f3291658df6863df50e0ba53ddde5c83538e2079f4b3b9b548cc92d90cca?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/ea95f3291658df6863df50e0ba53ddde5c83538e2079f4b3b9b548cc92d90cca?s=96&d=mm&r=g","caption":"Christopher Ort"},"url":"https:\/\/i10x.ai\/blog\/author\/christopher-ort"}]}},"_links":{"self":[{"href":"https:\/\/i10x.ai\/blog\/wp-json\/wp\/v2\/posts\/338","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/i10x.ai\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/i10x.ai\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/i10x.ai\/blog\/wp-json\/wp\/v2\/users\/5"}],"replies":[{"embeddable":true,"href":"https:\/\/i10x.ai\/blog\/wp-json\/wp\/v2\/comments?post=338"}],"version-history":[{"count":3,"href":"https:\/\/i10x.ai\/blog\/wp-json\/wp\/v2\/posts\/338\/revisions"}],"predecessor-version":[{"id":382,"href":"https:\/\/i10x.ai\/blog\/wp-json\/wp\/v2\/posts\/338\/revisions\/382"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/i10x.ai\/blog\/wp-json\/wp\/v2\/media\/347"}],"wp:attachment":[{"href":"https:\/\/i10x.ai\/blog\/wp-json\/wp\/v2\/media?parent=338"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/i10x.ai\/blog\/wp-json\/wp\/v2\/categories?post=338"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/i10x.ai\/blog\/wp-json\/wp\/v2\/tags?post=338"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}