Guides

The Book-Size Ceiling: How Many Companies Can One Recruitment Consultant Really Cover in 2026?

A capacity-math breakdown of recruitment consultant book size: how many client companies one BD consultant can realistically research, monitor and engage manually, and how an AI layer moves the ceiling.

TB Team Boilr

· July 7, 2026 · 14 min read

Abstract dark liquid-metal texture with pooling emerald light, representing the capacity ceiling of a recruitment consultant's book of business

TL;DR

Every recruitment agency owner is being asked the same question in 2026: grow billings without growing headcount [4]. The honest answer starts with a number nobody quotes in board meetings: how many client companies can one consultant actually research, monitor for signals, and engage well at the same time? The classic "book of business" benchmark - 10 key accounts plus 20 back-up accounts, 30 total - has held for decades [2], and it maps almost exactly onto how many hours a person has to manually watch a company for hiring intent. B2B sales research puts a similar hard ceiling on account ownership, warning that reps rarely succeed past 100-150 accounts even with software help [1]. What actually moves the ceiling isn't asking consultants to work harder. It's removing manual monitoring from the equation entirely, which is what an AI layer like boilr's Signals and Company Brain is built to do.

Why book size is suddenly the metric agency owners can't ignore

"Book size" - the number of client companies a consultant actively owns, researches and sells into - used to be a back-office planning number. In 2026 it has become the central capacity question for agency owners, for a simple reason: the traditional lever for growing billings, hiring more fee-earners, has stopped working the way it used to.

Headcount growth no longer pencils out. Agencies are being told directly that "growing internal headcount the traditional way no longer stacks up financially" against leaner, nationally-scaled competitors [4].
Fixed costs are already high per seat. Core monthly running costs for a recruiting agency sit at roughly $25,000-$30,000 before a single placement commission is paid [7] - and that's before the fully-loaded cost of the consultant's own salary, benefits, and desk.
Billings per consultant already have a wide, visible spread. A competent UK recruitment consultant typically bills £120,000-£200,000 a year; a first-year consultant bills £60,000-£100,000; a top biller in a specialist niche can clear £300,000+ [5]. That spread is mostly explained by how many good-fit accounts each person is actually covering well, not by effort.
Cost-per-hire keeps climbing on the client side too. Average cost-per-hire for a non-executive role sits around $5,475, and executive hires now average $35,879, a 21% jump since 2022 [3]. Clients are more scrutinous about who they let into their hiring process, which raises the bar for how well-researched a consultant's approach has to be before they'll take the call.
"Do more with the same team" is now explicit strategy, not a slogan. Agencies are being told to "deliver speed, quality, and flexibility with smaller teams and tighter budgets" [4], which is a direct instruction to increase book size per head, whether or not anyone has worked out what that means in practice.

None of this works if you simply tell a 360 consultant to "take on more accounts." Somewhere there is a real ceiling on how many companies one person can actually watch, understand, and engage with credibility. The rest of this article is about finding that ceiling, and what changes it.

What "book size" actually means, and the benchmark that has held for decades

A consultant's book of business is not just a client list. It's the set of companies where the consultant has (or is building) enough context to spot a hiring need early, know who to speak to, and have a credible reason to reach out this week rather than "just checking in." Long before AI was part of the conversation, veteran recruitment trainer Barbara Bruno set out what a sustainable book actually looks like:

10 key accounts - clients who call the consultant first, get a monthly touchpoint whether or not there's an open role, and generate the bulk of predictable revenue [2].
20 back-up accounts - companies who know the consultant, hire in the niche every year, but also work with two or three other agencies. These get promoted to key-account status if a key account goes quiet [2].
30 accounts total is what Bruno calls a "recession-proof" territory - large enough to survive a couple of key accounts pausing hiring, small enough that the consultant can still track what's happening at each one [2].

The same research flags the failure mode most desks actually live in: the majority of recruiters have five clients or fewer that represent 75%+ of their billings [2]. That's not a healthy book, it's concentration risk with a job title. It happens because building context on a new account, understanding its hiring patterns, its org chart, its objections, its buying signals, is genuinely slow manual work, so consultants default to deepening their existing three or four relationships rather than spreading themselves across thirty accounts they can't keep up with.

Borrowing the math from B2B sales: there's a hard ceiling on accounts per person

Recruitment BD is account-based B2B sales with a candidate-side twist, so it's worth checking what account-based sales teams have already worked out about coverage limits. The pattern holds up well:

An activity-based ceiling. Model it from available time: roughly 1,500 productive selling hours a year, or 125 hours a month. At around 3 hours of account-specific work a month (research, monitoring, touches), that caps out at roughly 40-50 accounts a person can genuinely stay on top of [1].
A cognitive ceiling, not just a time one. Sales-capacity research explicitly points to Dunbar's number as the reason reps "rarely succeed" once they're covering more than 100-150 accounts at a time, even with software support [1]. Past that point, a human genuinely cannot hold enough context on each account to sell well.
An activity threshold that quietly shrinks the "real" book. Reps typically have to actively touch at least 75% of an account within a 30-day window to justify keeping it, which is why AE books usually settle at 100-115 live accounts and BDR/SDR books at 125-150, even when the nominal target is higher [1].

Line that up against the 30-account recruitment benchmark and something clicks: 30 accounts is roughly what a single person can manually monitor for hiring signals, funding news, and org changes, on top of actually selling and delivering. It has held for decades not because recruiters lack ambition, but because the manual research workload per account has never gone down. Until now, the constraint on book size was never selling capacity. It was watching capacity.

The manual coverage math: what actually caps a book today

Here's a simple way to model where the ceiling comes from for a manually-run desk. It isn't a universal constant, every desk's numbers differ, but the shape of the maths is the same everywhere:

Variable	Typical manual value	What it represents
BD hours available per week	10-15 hrs	Time left for prospecting after delivery, calls, and admin on a 360 desk
Manual research + monitoring time per account/week	25-40 min	Checking job boards, LinkedIn, Companies House filings, news, funding databases per company
Extra time when a signal actually fires	1-2 hrs	Deep-dive research, finding the right contact, drafting a genuinely relevant approach
Resulting book-size ceiling	~20-30 accounts	Matches the classic 10 key + 20 back-up benchmark almost exactly [2]

Push past that ceiling manually and something has to give. Either the consultant stops truly monitoring the back-up accounts (so they only look "active" on paper), or the quality of research per account drops (so outreach turns generic, which is exactly the trap that keeps cold approaches converting at a fraction of signal-based ones), or the consultant simply burns out trying to keep 50 spinning plates in the air. None of those outcomes grow billings. They just spread the same billings more thinly and add risk.

Where the ceiling actually breaks first as a book grows

Agency owners who push book size up without changing the underlying research workflow tend to hit the same handful of failure points, usually in this order:

Signal blindness on back-up accounts. The consultant is technically "assigned" 40 companies but is only really watching the top 10. A funding round or a new VP of Engineering at company #35 goes unnoticed for weeks.
Personalisation decay. As accounts pile up, outreach reverts to generic templates because there's no time to research each one properly, which is the fastest way to turn a warm signal-based approach into something that reads like cold outreach.
Stale ICP fit. Accounts get added to "grow the book" without being checked against the agency's actual ideal client profile, so headcount grows on paper while genuinely winnable accounts shrink as a share of the total.
No shared memory across the desk. Everything the consultant has learned about an account, who the real decision-maker is, what objection killed the last conversation, sits in that one person's head and inbox, not anywhere the agency can retrieve it.
The book collapses on exit. When that consultant leaves, the accounts don't just need a new owner, they need someone to rebuild months of context from zero, which is exactly when competitors who were already watching those companies get the meeting instead.
Burnout masquerading as a headcount problem. The agency reads the strain as "we need to hire another consultant" when the real issue is that manual monitoring time per account was never designed to scale past 30 in the first place.

What actually changes when an AI layer takes over the monitoring

The lever that raises the ceiling isn't asking a consultant to research faster. It's removing manual research and monitoring from the loop entirely, and only handing the consultant a company once there's a genuine, verified reason to act. That's the specific job an AI sales employee like boilr is built to do, and it's worth being precise about which parts of the workflow actually change.

What gets automated

Continuous monitoring across the whole book, not just the top accounts. boilr's Signals layer scans 10,000+ verified sources around the clock, tracking funding rounds, executive moves, hiring velocity, tech-stack changes and job-posting activity, and can surface a hiring need 48-72 hours before it is posted publicly.
ICP scoring at book-wide scale. The Companies module scores every account against the agency's own ideal client profile across industry, headcount, region, funding stage, hiring velocity and tech stack, so a book of 200 companies stays filtered to the ones actually worth a consultant's time, not just the ones that happen to be top of mind.
Decision-maker enrichment. Once a company clears the ICP bar, the right contact and two or three specific reasons to reach out are surfaced automatically, removing the 20-30 minutes per account per week that used to go into manually tracking "who's who" at each client.
Shared memory that survives the book handing over. Company Brain stores winning messages, ICP patterns, objection handling and signal history at agency level, not in one consultant's head, so a book can actually be reassigned without starting from zero.

What stays firmly human

Verifying and sending outreach. Every drafted approach goes through the consultant before it goes out, typically 5-20 minutes a day of review rather than hours of research.
The actual relationship-building call. Discovery conversations, objection handling in real time, and negotiation stay entirely human.
Judgement on which signals matter. An AI layer surfaces the signal; the consultant still decides whether now is the right moment for this particular client.
Closing. Proposals, fee negotiation and contract signing remain relationship-driven and human-led.

Manual book vs AI-assisted book: the comparison

Dimension	Manual desk	AI-assisted desk (boilr)
Monitoring time per account/week	25-40 minutes	Near zero (continuous, automated across 10,000+ sources)
Signal-to-awareness lag	Days to weeks, if noticed at all	48-72 hours ahead of public job postings
Realistic book-size ceiling	~20-30 accounts	150-300+ accounts under active monitoring
ICP consistency across the book	Degrades as book grows, no time to re-check fit	Every account scored continuously against the ICP
What survives a consultant leaving	Whatever they wrote down, often little	Full account context in Company Brain, reassignable same day
Daily consultant time on top-of-funnel	2-5 hours	5-20 minutes of review and approval

The number that jumps out is the book-size ceiling itself. Removing manual monitoring doesn't just save time, it changes the constraint entirely. The bottleneck stops being "how many companies can I watch" and becomes "how many verified, well-researched conversations can I actually have and close", which is a much more favourable problem for an agency to solve, and one that scales billings without adding another desk.

The KPIs that tell you if your book is already over capacity

Book size on its own is a vanity number. These are the metrics that actually reveal whether a consultant's book is healthy, stretched, or already collapsing under its own weight:

Metric	What it reveals	Warning sign
% of book actively touched in 30 days	Whether accounts are truly owned or just assigned on paper	Below 75% suggests accounts are effectively unmanaged
Revenue concentration (top 5 accounts as % of billings)	Whether the book is diversified or one client loss away from a bad quarter	Above 75% mirrors the risky "5-client" pattern [2]
Average research time per new outreach	Whether personalisation is holding up as the book grows	Rising sharply as book size increases
Time from signal to first contact	Whether the consultant is still first to engage after a trigger event	More than a few days, competitors are likely already in
Handover time when a consultant leaves	How much account context actually lives outside one person's head	Weeks of ramp-up for a replacement suggests no shared memory
ICP-fit % of the current book	Whether "growing the book" is adding winnable accounts or just noise	Falling as headcount of assigned accounts rises

5 mistakes agencies make when trying to grow book size in 2026

Mistake #1: Treating book size as a headcount metric instead of a coverage metric

Why it fails: Assigning a consultant 60 accounts on a spreadsheet doesn't mean 60 accounts are being watched. Without a monitoring layer, most of that "growth" is invisible to the client and invisible to the consultant.

Fix: Measure the percentage of the book actively touched in the last 30 days, not the raw headcount of assigned accounts.

Mistake #2: Scaling book size before fixing shared memory

Why it fails: A bigger book with no Company Brain just means more context locked in one person's head, which is a bigger liability the moment they resign.

Fix: Put winning messages, ICP patterns and account history into a shared system before, not after, expanding anyone's book.

Mistake #3: Adding accounts without checking ICP fit

Why it fails: A bigger book of poor-fit companies converts worse than a smaller book of well-matched ones, and burns consultant time on accounts that were never going to buy.

Fix: Score every candidate account against the agency's ICP (industry, headcount, funding stage, hiring velocity) before it's added to anyone's book.

Mistake #4: Assuming AI removes the need for consultant judgement

Why it fails: Clients can tell when outreach was sent without a human actually deciding it was relevant. Automation without review reads as spam, however well-timed the signal was.

Fix: Keep verification and sending human. The AI layer's job is to remove research time, not the final judgement call.

Mistake #5: Growing the book faster than the desk's actual selling capacity

Why it fails: Removing the monitoring bottleneck doesn't remove the calls, meetings and proposals bottleneck. A consultant freed from 3 hours of daily research still only has so many discovery calls in a day.

Fix: Grow the book in stages, measure meeting-booked rate and proposal-to-close rate at each stage, and only expand further once the selling side is keeping pace.

A 30-day plan to audit and safely expand your book size

Rather than guessing at a new target book size, run a short structured audit before expanding anyone's list:

Week 1: Audit the current book

List every account currently assigned to each consultant. Mark which ones were genuinely touched in the last 30 days versus which exist only on paper. Calculate revenue concentration in the top 5 accounts per desk.

Week 2: Score every account against the ICP

Run the full book through the agency's ideal client profile (industry, headcount, region, funding stage, hiring velocity). Flag poor-fit accounts for removal or reassignment rather than continuing to service them out of habit.

Week 3: Pilot AI-assisted monitoring on a subset

Take one consultant's book and route signal detection, ICP scoring and contact enrichment through an automated layer for two to three weeks. Track how research time per account changes and how quickly signals reach the consultant compared to the manual baseline.

Week 4: Measure and expand deliberately

Compare meeting-booked rate and time-to-first-contact before and after. If the automated layer is holding quality steady while cutting monitoring time, expand that consultant's book in controlled increments (for example 10 accounts at a time), watching proposal-to-close rate at each step rather than jumping straight to a much larger number.

Want to see what your own book-size ceiling actually looks like? boilr's Company Brain and Signals modules are built to raise it without adding headcount.

Frequently Asked Questions

What is a "book of business" or book size for a recruitment consultant?

A book of business is the set of client companies a recruitment consultant actively owns, researches and sells into. It's not just a client list. It includes the context the consultant holds on each account, hiring patterns, decision-makers, past objections, and signals worth acting on. Book size refers to how many of these accounts one person is realistically covering well at a given time.

How many client companies can one recruitment consultant realistically manage?

The long-standing benchmark for a manually-run desk is around 30 accounts: 10 key accounts that generate the bulk of billings, plus 20 back-up accounts held in reserve [2]. That figure aligns closely with how much manual research and monitoring time a person actually has available each week. With an AI layer handling monitoring and ICP scoring, that realistic ceiling can move well past 100 accounts, similar to the coverage ranges seen in software-assisted B2B account management [1].

Why can't agencies just tell consultants to take on more accounts in 2026?

Because the bottleneck was never willingness, it was manual research time. Watching a company for hiring signals, funding news and org changes takes roughly 25-40 minutes a week per account by hand, on top of selling and delivery work. Past a certain book size, something has to give: monitoring quality, outreach personalisation, or the consultant's own capacity. Simply expanding the list without changing the workflow tends to spread the same billings more thinly rather than growing them.

What actually changes about book size when you add an AI research layer?

It removes manual monitoring as the constraint. Instead of a consultant spending 25-40 minutes a week per account checking for signals, an automated layer like boilr's Signals module scans thousands of sources continuously and only surfaces a company once there's a verified, specific reason to act. The consultant's time shifts from research to review, typically 5-20 minutes a day, freeing up capacity to cover a much larger book without sacrificing personalisation.

Does a bigger book size automatically mean higher billings?

No, and this is the most common mistake agencies make when scaling. A bigger book of poorly-fit or unmonitored accounts converts worse than a smaller, well-covered one. Book size only translates into billings when every account added is scored against the agency's ICP and genuinely monitored, not just assigned on paper.

How does a shared Company Brain affect book size when a consultant leaves?

Normally, when a consultant leaves, their book collapses with them: account history, objection-handling patterns and relationship context typically leave in their head, and a replacement has to rebuild that from scratch. A shared knowledge layer stores winning messages, ICP patterns and signal history at agency level, so a departing consultant's book can be reassigned with its context intact rather than starting from zero.

Is the ideal book size the same for a 360 desk as for a BD-only consultant?

No. A 360 consultant splits time between delivery (interviewing, managing candidates through process) and business development, which leaves fewer hours for account research and naturally caps book size lower, often towards the 20-30 range. A BD-only or account-manager role with no delivery responsibilities can typically sustain a larger book, and benefits even more from automated monitoring since research time is the entire job.

How do I know if my agency's current books are already over capacity?

Check the percentage of each book genuinely touched in the last 30 days, and how concentrated billings are in the top five accounts. If touch rates are low and more than 75% of billings come from a handful of clients, the book is likely already over capacity, whatever the nominal account count says on paper [2].

Sources

Information sourced from public industry reports, benchmarks, and research publications as of July 2026.

Back to all articles