Sourcing Strategy
When To Use
Inbound is unusable, InMail replies have collapsed, and you need a plan rather than more of the same. Use it at the start of a search, or two weeks into one that is not moving. It answers: which channels, how much effort each week, and what do we stop?
When Not To Use
If you only need to design one channel in depth, such as asking staff for introductions, run Employee Referral Program. If you have not mapped where the pool sits, run Talent Mapping first; a channel plan without a map is a guess.
Inputs
- The job scorecard and the Talent Map for the role
- Your own past results per channel for similar roles: source of hire, response rates, time spent, or "no data"
- The tools and seats you have, and your weekly hours for sourcing
- Your retention period and candidate notice for past applicants, as confirmed with an adviser
If you have none of this, I start from the scorecard and your weekly hours, mark every channel "no data", and label the plan a first draft.
Approach
Channel mix and candidate rediscovery are practitioner methods, described here generically: spread effort across the routes where the pool actually sits, and reuse people who already applied, within the rules on keeping applicant data (ICO recruitment and selection guidance; check with a qualified adviser on retention). The judgment is to plan by evidence, not habit. The failure it prevents is the recruiter sending a fourth week of InMails to the same exhausted pool because that is what last year's plan said.
Workflow
- Ask at most three questions: which channels you ran last time and what came of them, what seats and hours you have, and how long past applicants may be kept and contacted under your notice.
- List every candidate channel for this role: outbound search, referrals, communities and groups, past applicants and silver medallists, job boards, events, alumni networks. Beside each, your own past result or "no data".
- Choose the mix from the Talent Map: weight channels that reach the teams and feeder fields where the pool is, and cut channels that only reach the manager's narrow list if the map shows the pool is elsewhere.
- Past applicants: include them only within the retention period and notice you confirmed with an adviser, and only people who applied in that window. Contact goes to a person, never an automated sweep.
- Add free routes if you have no paid seat: Google X-ray from the Boolean strings, open communities, meetups and conference speaker lists at team level, alumni groups.
- Set weekly targets per channel as [placeholders] you choose (approaches, conversations), a review date after [n] weeks, and a stop rule: any channel below [threshold] at review is paused and its hours move.
- Write the "what to stop" list now: channels, habits or tools that cost time with no result in your own data.
Output Format
# Sourcing Strategy: [Role title]
## Channel mix
| Channel | Why it fits (Talent Map row or feeder field) | Your past result | Weekly target | Owner |
|---|---|---|---|---|
| [channel] | [reason] | [result or no data] | [placeholder] | [name] |
## Free routes
| Route | What you run | Hours per week |
|---|---|---|
| [route] | [action] | [placeholder] |
## Past applicants
| Window | Notice confirmed with adviser | Who contacts them |
|---|---|---|
| [dates] | [yes or pending] | [name] |
## What to stop
- [channel or habit, and the evidence]
## Decision
[Recruiter or TA lead name] confirms the mix and targets by [date]; review on [date] against the stop rule of [threshold].
Done When
- Every channel carries your own evidence or an honest "no data"
- The mix points back to rows of the Talent Map
- Past applicants are included only with a confirmed retention window
- Each channel has a target, an owner, a review date and a stop rule
Quality Bar
- Targets and thresholds are the recruiter's numbers, never Claude's.
- No scraped or enriched personal data in any channel.
- Retention and notice for past applicants: check with a qualified adviser.
- A channel that cannot show results at review is paused, not defended.
- Name what stops as clearly as what starts.
Next
Run recruit-recruiting-email-templates (Recruiting Email Templates) to write the outreach for the channels you chose.
About the makers
This pack is made by 90n (90n.ai): AI-native ways of working and tools for Product, Marketing and HR teams, built by ex-McKinsey founders. We map where time and money leak, redesign the work around AI and stay until your people run it themselves, and we run our own company on Claude. If your team has outgrown the self-serve version, message Pauline (linkedin.com/in/paulinebertry).
1---2name: recruit-sourcing-strategy3description: Plans the sourcing channel mix for one role across outbound, referrals, communities, past applicants, job boards and events, with free routes when you have no paid seat, weekly targets per channel and a list of what to stop. Use for "run recruit-sourcing-strategy", "sourcing strategy", "sourcing plan for this role", "inbound is unusable", "InMail replies have collapsed", "sourcing with no budget", "candidate rediscovery", "where should I source", part of the AI for Recruiting Pack by 90n.4metadata:5 source: https://skillmd.com/skills/polar-bear-org/recruit-sourcing-strategy6---78# Sourcing Strategy910## When To Use11Inbound is unusable, InMail replies have collapsed, and you need a plan rather than more of the same. Use it at the start of a search, or two weeks into one that is not moving. It answers: which channels, how much effort each week, and what do we stop?1213## When Not To Use14If you only need to design one channel in depth, such as asking staff for introductions, run Employee Referral Program. If you have not mapped where the pool sits, run Talent Mapping first; a channel plan without a map is a guess.1516## Inputs17- The job scorecard and the Talent Map for the role18- Your own past results per channel for similar roles: source of hire, response rates, time spent, or "no data"19- The tools and seats you have, and your weekly hours for sourcing20- Your retention period and candidate notice for past applicants, as confirmed with an adviser21If you have none of this, I start from the scorecard and your weekly hours, mark every channel "no data", and label the plan a first draft.2223## Approach24Channel mix and candidate rediscovery are practitioner methods, described here generically: spread effort across the routes where the pool actually sits, and reuse people who already applied, within the rules on keeping applicant data (ICO recruitment and selection guidance; check with a qualified adviser on retention). The judgment is to plan by evidence, not habit. The failure it prevents is the recruiter sending a fourth week of InMails to the same exhausted pool because that is what last year's plan said.2526## Workflow271. Ask at most three questions: which channels you ran last time and what came of them, what seats and hours you have, and how long past applicants may be kept and contacted under your notice.282. List every candidate channel for this role: outbound search, referrals, communities and groups, past applicants and silver medallists, job boards, events, alumni networks. Beside each, your own past result or "no data".293. Choose the mix from the Talent Map: weight channels that reach the teams and feeder fields where the pool is, and cut channels that only reach the manager's narrow list if the map shows the pool is elsewhere.304. Past applicants: include them only within the retention period and notice you confirmed with an adviser, and only people who applied in that window. Contact goes to a person, never an automated sweep.315. Add free routes if you have no paid seat: Google X-ray from the Boolean strings, open communities, meetups and conference speaker lists at team level, alumni groups.326. Set weekly targets per channel as [placeholders] you choose (approaches, conversations), a review date after [n] weeks, and a stop rule: any channel below [threshold] at review is paused and its hours move.337. Write the "what to stop" list now: channels, habits or tools that cost time with no result in your own data.3435## Output Format36```markdown37# Sourcing Strategy: [Role title]38## Channel mix39| Channel | Why it fits (Talent Map row or feeder field) | Your past result | Weekly target | Owner |40|---|---|---|---|---|41| [channel] | [reason] | [result or no data] | [placeholder] | [name] |42## Free routes43| Route | What you run | Hours per week |44|---|---|---|45| [route] | [action] | [placeholder] |46## Past applicants47| Window | Notice confirmed with adviser | Who contacts them |48|---|---|---|49| [dates] | [yes or pending] | [name] |50## What to stop51- [channel or habit, and the evidence]52## Decision53[Recruiter or TA lead name] confirms the mix and targets by [date]; review on [date] against the stop rule of [threshold].54```5556## Done When57- Every channel carries your own evidence or an honest "no data"58- The mix points back to rows of the Talent Map59- Past applicants are included only with a confirmed retention window60- Each channel has a target, an owner, a review date and a stop rule6162## Quality Bar63- Targets and thresholds are the recruiter's numbers, never Claude's.64- No scraped or enriched personal data in any channel.65- Retention and notice for past applicants: check with a qualified adviser.66- A channel that cannot show results at review is paused, not defended.67- Name what stops as clearly as what starts.6869## Next70Run recruit-recruiting-email-templates (Recruiting Email Templates) to write the outreach for the channels you chose.7172## About the makers7374This pack is made by 90n ([90n.ai](https://90n.ai)): AI-native ways of working and tools for Product, Marketing and HR teams, built by ex-McKinsey founders. We map where time and money leak, redesign the work around AI and stay until your people run it themselves, and we run our own company on Claude. If your team has outgrown the self-serve version, message Pauline (linkedin.com/in/paulinebertry).