Sourcing Passive Candidates on LinkedIn: A Framework That Survives Contact With Reality

Most guidance on passive sourcing consists of a list of search operators and a message template. Both are the easy parts. The parts that actually determine whether the approach works are judging how open a person is likely to be, and writing something specific enough that a busy person who is not job hunting decides to reply.

Openness is a spectrum, not a switch

The active/passive binary is a useful piece of vocabulary and a poor model of behaviour. In practice people distribute across a range, and the appropriate approach differs at each point.

Position	Characteristic	What works
Actively looking	Applying, visible availability signals	Speed. They are talking to several companies already.
Quietly dissatisfied	Not applying, but would take a call	A concrete alternative. They need a reason to act, not persuasion that action is possible.
Content but curious	Happy, would listen to something genuinely better	Specificity about why this role is a step up for them in particular.
Committed	Recently moved, mid-project, equity cliff ahead	Almost nothing now. Worth a relationship for later.

The largest group — and the reason outbound recruiting exists — is the middle two. They will not appear in any applicant pool, and they will not respond to a message that reads as if it were sent to a thousand people, because such a message gives them no information about whether this particular opportunity is better than what they have.

Reading openness signals honestly

A number of profile signals are widely treated as indicators of openness. They vary considerably in how much they actually tell you, and some commonly-repeated ones are close to worthless.

Signals with genuine information content

• Tenure relative to their own pattern. Someone whose history shows moves roughly every three years, currently at three and a half, is at a different point than someone with a ten-year tenure at one company. The comparison that matters is against their own history, not an industry average.
• A recent structural change at their employer. Acquisition, layoffs, a leadership change, a strategy reversal. These change the value of staying, and they are observable.
• Recent lateral movement or a title that has not changed in an unusually long time. Stalled progression is one of the more reliable drivers of openness, though it must be read carefully — some senior roles simply do not have annual title changes.
• Public work with no obvious connection to their day job. Side projects, talks, writing in an adjacent domain. Often a signal of interest that their current role is not satisfying.

Signals that are weaker than they appear

• Profile updates. Frequently cited as an intent signal. It is heavily confounded — people update profiles after promotions, for personal branding, or because a tool prompted them. Weak on its own.
• Availability badges. Real information about availability, but visible to every recruiter at once. These candidates are the most contacted segment of the market. Include them, do not build a strategy on them.
• Engagement with recruiting content. Extremely noisy. People engage with content for professional-visibility reasons entirely unrelated to their own job search.

No single signal justifies an inference. Two or three in combination — a person at the far end of their own tenure pattern, at a company that just went through a leadership change, whose title has not moved in four years — is a reasonable basis for prioritising outreach. One signal alone is a coin flip.

Search: wide on judgement, narrow only on constraints

The instinct when a search returns too many results is to add filters. This is usually the wrong move, and it is worth being precise about why.

Divide your criteria into two categories before you touch a search interface:

Hard constraints	Judgement criteria
Location and work authorisation	Has operated at the relevant scale
Genuinely non-negotiable domain experience	Has done this without extensive support structures
Language requirements where the role demands it	Depth versus breadth of ownership
Regulatory or clearance requirements	Trajectory and rate of progression

Only the left column belongs in a filter. Every judgement criterion you convert into a keyword filter removes qualified people whose profile happens to use different words — and you never find out who they were. This is the recall problem covered in detail in our piece on precision and recall in shortlists.

The title problem, specifically

Job titles are the most commonly used filter and among the least reliable. The same responsibility carries wildly different titles across company sizes and cultures; the same title covers wildly different responsibilities. Filtering to a list of exact titles is a large sacrifice of recall for a small gain in precision.

If you must use titles, use them as one input among several rather than as a gate — and always include the adjacent titles that describe the same work. The person who did the job you are hiring for is often not titled as if they did.

When the title problem is severe enough that a search returns almost nobody, the fix is usually vocabulary rather than persistence. Headhunting for niche roles covers how to take your search terms from the market instead of from the job description.

A note on scale

Wide search creates a review problem: more results means more profiles to read. This is a real cost, and it is the reason teams narrow in the first place. The resolution is not to narrow but to make evaluation cheaper — which is the entire argument of our unit economics model, and the problem Hiris is built to solve by attaching the evidence for each criterion to each profile so verification takes seconds.

Outreach: specificity beats personalisation

These are routinely conflated and they are not the same thing.

Personalisation is inserting facts about the person into an otherwise generic message. Mentioning their recent post, congratulating them on an anniversary, referencing their alma mater. It is easy to automate, and because it is easy to automate, recipients have learned to discount it entirely.

Specificity is demonstrating that you understood what this person does and can articulate why it maps to this particular role. It is difficult to fake, which is exactly why it works — the effort is legible in the message itself.

Structure that holds up

1. Why you, specifically. One sentence identifying the thing about their experience that made you reach out. Not their title — the substance.
2. What the thing is. Concretely. The problem, the stage, the team shape. Enough that they can evaluate it without a call.
3. Why it might be a step up for them. This requires an actual hypothesis about their situation, and it is the sentence most messages omit.
4. A small next step. A question they can answer in one line, not a calendar link.

The same message, two ways

Generic, with personalisation bolted on:

Hi Anna — saw your recent post on distributed systems, great insights! I'm recruiting for a Senior Backend Engineer role at a fast-growing startup. Your background looks like a great fit. Would you be open to a 30-minute call this week to discuss?

Specific:

Hi Anna — you spent two years migrating a monolith to event-driven services at a company with no dedicated platform team, which is unusual and is close to the problem here. We are pre-platform-team, about to do the same migration, and the first hire owns the design rather than inheriting it. Is that the kind of problem you would want again, or have you had enough of it?

The second is barely longer. It contains a real hypothesis about her experience, states the role honestly including its constraints, and asks a question that is genuinely interesting to answer — including the option of saying no, which makes replying feel low-risk. It also cannot be sent to anyone else, which is precisely the point.

What to leave out

• Company boilerplate. If they want it they will look.
• Superlatives about growth or culture. They carry no information because every message contains them.
• A thirty-minute call as the opening ask. That is a large commitment to someone who has not yet decided the role is interesting.
• Compensation as the lead, unless it is genuinely exceptional. It signals that nothing else about the role is compelling.

Follow-up: fewer, with new information

A reasonable default is two follow-ups after the initial message, spaced several days apart. The rule that matters more than the cadence is that each follow-up must contain something new — a detail about the team, a change in the role, a specific answer to an objection you anticipate. A follow-up that only repeats the original ask adds nothing and costs goodwill.

Consider also the arithmetic of your own funnel. If your reply rate on the first message is low, the fix is almost never more follow-ups; it is either a targeting problem, in which case the messages are going to the wrong people, or a specificity problem, in which case the messages are not earning attention. Follow-ups amplify whatever the first message was. They do not repair it.

What to measure

Four numbers, tracked by search rather than in aggregate, because aggregates hide the variation that matters:

• Precision — of the profiles surfaced, how many survived review. The leading indicator for everything downstream.
• Reply rate — substantive responses, including declines. Declines are information; silence is not.
• Positive reply rate — the subset expressing genuine interest. Diverging trends between this and overall reply rate usually indicate a targeting problem rather than a messaging one.
• Reply-to-screen conversion — whether interest survives the first conversation. A high reply rate with poor conversion generally means the message oversold the role.

That last pattern is worth watching for specifically. It is the signature of outreach that is working for the wrong reasons, and it costs more than a low reply rate because it consumes screening time and damages your reputation with exactly the people you most want to reach later.