AI with Michal

Talent sourcing playbook

A documented, repeatable system for finding and engaging passive candidates, covering search strategies, outreach sequences, qualification criteria, and handoff points, built so a team can run consistent sourcing on any role without starting from scratch each time.

Michal Juhas · Last reviewed May 4, 2026

What is a talent sourcing playbook?

A talent sourcing playbook is a documented, repeatable system for finding and engaging passive candidates. It covers where to look, how to qualify a profile before outreach, how to write and sequence messages, and where the hand-off happens to a recruiter or hiring manager. The goal is for anyone on the team to run a consistent sourcing campaign on any role without rebuilding the process from zero each time.

A playbook is not a job description template or a list of platforms. It is the decision logic behind how your team sources: which sources produce which types of candidates, what criteria qualify a profile before it enters the outreach queue, and what a good hand-off looks like.

Illustration: talent sourcing playbook as an open document feeding three connected stage chips for search, qualification, and outreach, with a team receiving the output and a cycle arrow looping improvements back to the playbook

In practice

  • When a new sourcer joins a team and can run their first search campaign without a two-hour knowledge transfer call because the criteria, Boolean strings, and outreach templates are documented, that is the practical payoff of a sourcing playbook.
  • A senior sourcer who opens a Notion page and follows a checklist for a machine learning engineering search (sources, screening criteria, outreach message variant, follow-up timing) is using a playbook as operational infrastructure rather than institutional memory.
  • A team where every sourcer explains the qualification criteria differently in debrief, and reply rates vary by 40 percent across sourcers on identical roles, is a team that needs a playbook.

Quick read, then how hiring teams use it

This section is for sourcers, recruiters, TA managers, and team leads who want a shared vocabulary for discussing sourcing consistency, AI integration, and process documentation. Skim the first part for the core concept. Read the second when you are building or auditing your team's sourcing system.

Plain-language summary

  • What it means for you: A sourcing playbook is how your team stops reinventing the search every quarter. It captures the search logic, qualification criteria, and outreach approach so good sourcing practice becomes institutional rather than individual.
  • How you would use it: You document one role family per sprint: where to find candidates, what qualifies them, how to engage them, and when to hand them off. You update it after each hiring cycle.
  • How to get started: Pick your highest-volume role family. Document how your best sourcer runs a search for it, including what they look for, what they skip, and what their first outreach says. That is your first playbook entry.
  • When it is a good time: Whenever a new team member joins, whenever reply rates drop, or whenever two sourcers explain the same role differently in a debrief.

When you are running live reqs and tools

  • What it means for you: At scale, a playbook is the interface between your process and your tools. AI-assisted sourcing (semantic search, candidate data enrichment, personalised outreach via prompt chains) compounds in value when the human logic behind it is documented and reviewed regularly.
  • When it is a good time: Before you connect any workflow automation to outreach. Automating an undocumented process just multiplies the inconsistency.
  • How to use it: Keep the playbook in the format the team actually opens (Notion, Confluence, a shared drive folder). Pair narrative guidance with ATS templates so the logic and the execution live near each other.
  • How to get started: Audit your current sourcing process by role family. For each one, document: primary source, qualification criteria, outreach sequence, and last time reply rates were reviewed. Gaps in that list are the playbook entries you write first.
  • What to watch for: A playbook that reflects market conditions from 18 months ago. Assign a quarterly review owner per role family and treat the review as seriously as you treat a pipeline meeting.

Where we talk about this

On AI with Michal workshops, building a sourcing playbook is a running thread across both the AI in recruiting and sourcing automation tracks. We walk through live role briefs, search strategies, and outreach sequence logic together, and participants leave with draft documentation for their own role families rather than a generic template. If you want the room conversation and peer review on your specific playbook, start at Workshops and bring a role brief you are actively sourcing.

Around the web (opinions and rabbit holes)

These are starting points, not endorsements. Playbook content, especially sourcing strategy, is highly market-specific. Read critically and test before adopting.

YouTube

Reddit

Quora

Related on this site

Frequently asked questions

What does a talent sourcing playbook actually contain?
A sourcing playbook documents the repeatable system your team uses to find and engage passive candidates. At minimum it covers: where to look (platforms, communities, databases, Boolean strings or semantic search approaches by role family), how to qualify (a light scorecard or criteria checklist before a profile enters the outreach queue), how to engage (message sequence, timing, personalisation rules, who reviews drafts), and how to hand off (what state a candidate record should be in before it moves to a recruiter or hiring manager). Many teams also include failure cases: which sources produce low-quality leads and why, and which job families have historically required a different approach. The document does not need to be long; it needs to be accurate and used.
How does AI change the way a sourcing playbook works?
AI accelerates each playbook step without replacing the playbook's logic. Semantic search finds candidates who match a brief without requiring a Boolean string per search. A prompt chain drafts personalised outreach from a profile and brief in seconds rather than minutes. Candidate data enrichment adds verified contact details to sourced profiles without manual lookups. What AI does not change: the decision criteria for who qualifies, the human review gate before outreach goes out, and the handoff protocol. Teams that update their playbook to name where AI runs each step, and who reviews the output, get repeatable gains. Teams that let AI run unchecked tend to see outreach tone drift and candidate experience complaints within a few weeks.
How do I build a sourcing playbook for a role family my team has not hired before?
Start with the intake call. Ask the hiring manager which companies produce people who succeed in this role (not just which schools or titles), what the usual career path looks like three years in, and what the role is not, so you do not source the wrong profile confidently. Use that to draft search criteria and a short qualification checklist before you source a single profile. Run a small batch of 15 to 20 candidates through your outreach with different message variants, note what reply rates look like, and calibrate the criteria against who actually progresses. After one hiring cycle you have a draft playbook entry. After two or three cycles you have a reliable one. Document what did not work as explicitly as what did.
What are the signs a sourcing playbook is outdated?
Reply rates drop without an obvious campaign change. Sourcers re-explain the same criteria in every intake. Candidates who progress consistently miss the hiring manager's standard even though the scorecard says they should not. The platform that used to produce 60 percent of shortlists now produces 20 percent. AI outreach drafts that worked in Q1 feel templated and generic by Q3. These are all signals that the playbook was written for a market or role definition that has since shifted. Assign a named owner and a review cadence (quarterly for high-volume roles, annually for specialist searches) so the playbook gets updated before sourcing quality degrades visibly.
How does a talent sourcing playbook connect to compliance and GDPR?
Sourcing passive candidates triggers GDPR obligations in the EU: you must have a lawful basis for processing their personal data, notify them of how their data is used, and honour deletion requests. A sourcing playbook should document the lawful basis for each sourcing channel (legitimate interest for most outbound sourcing, with a documented balancing test), the retention period for unsolicited profiles, and the process for handling a data subject access request. Teams that source at scale without this documentation face audit exposure. Including a data handling section in the playbook, linked to your legal or compliance team's guidance, means sourcers can answer candidate data questions without escalating to legal on every reply. See candidate data enrichment for additional data sourcing obligations.
Should the sourcing playbook be a shared document or built into the ATS?
Both formats work; the failure mode is picking one and then not maintaining it. A shared document (Notion, Confluence, Google Docs) is easier to update and review collaboratively, good for narrative guidance, and visible to new team members during onboarding. ATS-based playbooks (template messages, saved Boolean strings, stage-level checklists) are closer to where sourcers actually work and easier to enforce consistently. Most mature teams keep a short narrative document for the why and a set of ATS templates for the what. The critical rule: one person owns updates and there is a published review cadence. A sourcing playbook that reflects how the team worked 18 months ago is worse than no playbook because it creates false confidence.
Where can sourcing teams build and iterate a playbook alongside peers?
The sourcing automation workshop on AI with Michal builds sourcing sequences live, covering search strategies, outreach templates, and the AI steps that speed each stage without removing the human review gates. Attendees leave with a working draft they have tested on a real role brief, not a slide deck of best practices. Membership office hours are useful for playbook reviews: you share your current approach and get peer feedback on what to tighten. The Starting with AI: foundations in recruiting course covers the prompt and output review habits that need to be in the playbook before you wire automation on top. Read AI sourcing tools for recruiters for a practitioner breakdown of tools used in production sourcing playbooks.

← Back to AI glossary in practice