At kickoff, every open role can look covered. The recruiter will review inbound applications, run interviews, keep candidates moving, and still find time for outbound.

Then the week fills up. Interviews and debriefs are fixed on the calendar. Candidate replies and offer calls need an answer. Hiring managers add new requests. Net-new sourcing has no immediate deadline, so it is usually the first thing to slip.

An outbound capacity audit shows what each search actually received. It is not an industry benchmark or a recruiter productivity score. It is a way to compare the coverage agreed at kickoff with the work that happened, then decide which tasks need a recruiter and which can be delegated.

Outbound capacity is a triage problem

When outbound activity drops, start with the allocation before rewriting the message or blaming the market. Look at the recruiter’s week after inbound review, interviews, debriefs, candidate communication, offer work, and unplanned stakeholder requests. The time that remains is the time available for net-new pipeline work.

Now break that time down by requisition. A kickoff plan may say that every role will receive outbound support, but that promise is meaningless unless the team has decided how the available time will be divided. Without that decision, the gap usually appears later, when an aging role has no recent outreach and nobody knows whether that was intentional.

The audit should compare the planned coverage model with the work each search actually received. Some roles may be recruiter-led. Others may rely on a sourcer, an agency, an autonomous agent, or inbound applications. Some may be paused until the hiring manager resolves the brief. Any of those choices can be sensible. The problem is leaving the choice unstated.

This is not exact time-and-motion tracking. Calendar blocks, sent messages, and ATS activity will miss research and list-building work. Use those records as a starting point, then ask the recruiter to reconstruct the work the systems did not capture. Label the hours as an estimate rather than pretending every minute was observed.

Triage, done on purpose

Recruiting teams often avoid the word triage because it sounds like abandoning a role. In practice, triage is already happening. The newest requisition, the loudest stakeholder, the candidate waiting for a reply, or the interview happening tomorrow gets the available attention. The question is whether the trade-off is visible.

The cleanest way to make it visible is to name the coverage model at kickoff. That gives the hiring manager a chance to challenge the priority before the search starts, rather than discovering three weeks later that outbound never happened.

Siadhal Magos, Metaview’s CEO and co-founder, describes the part of recruiting that should remain human this way:

That is a point of view, and not an industry benchmark. It still gives teams a useful way to divide the work.

Research, list building, first-draft personalization, follow-ups, and scheduling are repeatable tasks. Defining the brief, correcting a weak search, handling edge cases, protecting confidential information, and building trust with candidates require judgment and context.

The dividing line runs through tasks, not job titles. A confidential executive search may still delegate research and list building. A high-volume role may need close recruiter involvement because the criteria are unsettled or the employer-brand risk is high. Decide which parts of each search need direct human attention instead of declaring the entire role manual or automated.

Three numbers worth measuring

A useful audit keeps three measures separate. Together they show where outbound time went, whether that work produced an active campaign, and whether the team is responding to the interest it already created.

Start with one recruiter and one recent week. That is enough to expose obvious gaps, but it is too small a sample for staffing targets or performance judgments. Repeat the exercise across several normal weeks before treating the pattern as stable.

1. Net-new outbound hours by search

Estimate the time spent building target lists, researching prospects, writing first messages, and following up with people who have not replied. Exclude inbound review and work with candidates who are already in process. Those activities matter, but they answer a different capacity question.

Calendar blocks are useful when recruiters protect sourcing time. Sent messages show that outreach happened, but not how long the research or personalization took. Use the calendar, activity records, and a short recruiter reconstruction together, then assign the estimated time to the relevant requisition.

The point is not to reward a high number of hours. It is to see which searches received focused outbound work and which did not.

2. New prospects entering outreach by search

Count the new prospects who entered active outreach for each requisition during the period. This is not a quality score or a quota. It shows whether the time recorded above resulted in a live campaign.

Keep the number beside the coverage model agreed at kickoff. Zero new prospects may be reasonable when the role is paused, inbound is sufficient, the pipeline is already full, or another owner is covering the search. The same zero is a problem when the team promised recruiter-led outbound and nobody noticed that it disappeared.

Do not compare raw volumes across very different searches without context. A narrow confidential search and a broad evergreen role should not be expected to produce the same number of new prospects.

3. Candidate replies awaiting action

Count active outbound conversations that need a response from the recruiter or hiring team. Then mark the replies that have waited beyond the threshold your team chose.

One working day can be a reasonable starting point, but it is not a universal service level. Seniority, geography, candidate expectations, weekends, and team coverage all affect what good response time looks like. Choose a threshold that fits the search and state it clearly.

This measure separates two problems that are often blended together: not enough new outreach and slow handling of the interest the team already generated. More messages will not fix a queue of unanswered candidate replies.

A simple pilot

1. Choose one recruiter and one recent, reasonably typical week.
2. List every open requisition and the coverage model agreed at kickoff.
3. Estimate net-new outbound hours for each search using calendar blocks, activity records, and the recruiter’s reconstruction.
4. Count the new prospects who entered outreach for each search.
5. Count active candidate replies that still need action, including those beyond the team’s chosen response threshold.
6. Mark every zero as intentional, unexplained, or caused by a change in the plan.
7. Review the findings with the recruiter and hiring manager, then repeat the audit across several representative weeks before making broader decisions.

Which searches deserve human outbound

Ask where direct recruiter involvement changes the work. Seniority alone is a poor rule.

Human involvement usually increases when the search is confidential, the talent pool is small, the criteria are still changing, the outreach carries reputational risk, or the candidate relationship needs careful handling. Even then, the recruiter does not have to perform every operational step.

A practical split can look like this:

• The recruiter and hiring manager define the brief, the evidence needed, and the acceptable trade-offs.
• A sourcer, agency, contractor, or agent researches the market and builds the target list.
• The recruiter reviews the list and corrects the search before outreach scales.
• Execution support drafts and sends outreach, manages follow-ups, and handles scheduling within agreed rules.
• The recruiter steps in for sensitive replies, exceptions, relationship-building, changes to the brief, and every hiring decision.

fillmore is Metaview’s autonomous recruiting agent for outbound pipeline work. Teams can describe a role, share a hiring brief, or explain the kind of person they need. fillmore can source and research candidates, build target lists, write personalized outreach, manage follow-ups, and book meetings with interested prospects.

It works through Slack, where recruiters and hiring managers can review progress, refine the search, adjust outreach, pause campaigns, and redirect priorities. That makes it relevant to searches that are receiving little or no sustained outbound work, but it does not remove the need for oversight.

A human still owns the brief, reviews whether the target list reflects the role, corrects drift, handles sensitive candidate conversations, and makes every hiring decision.

fillmore is currently available through a waitlist.

The operating change for talent leaders

The audit matters only if it changes the kickoff. Replace the general promise to run outbound with an explicit coverage model for every requisition.

• Recruiter-led outbound
• Recruiter-directed with sourcer or agent execution
• Agency or contractor-led
• Inbound-led for now
• Paused or awaiting a decision

For each role, record the owner, the next review point, and who takes over when a candidate replies. For recruiter-led work, add a rough allocation rather than promising unlimited coverage. For delegated work, decide how often the shortlist and messaging will be reviewed.

Revisit the allocation when the role or the market changes. A confidential search may close and release recruiter time. An outbound campaign may drift and need intervention. A role expected to fill through inbound may need outbound after the first week. The coverage model should change when the evidence changes.

Do not turn the first audit into a quota. A heavy interview week may leave little outbound time without indicating poor performance. A recruiter may spend fewer hours and produce a better shortlist because the brief was clearer. Use the numbers to ask what happened, and not to skip that conversation.

The goal is simple: every role should have an honest answer to how its pipeline will be built. A deliberate zero is manageable. A search that quietly receives no work for three weeks is not.