You’re staring at the same problem every partner in a growing consulting firm faces. You’ve got more work than your team can handle, but the math on hiring another body doesn’t quite close. A mid-level consultant costs you $120K all-in, needs six months to get productive, and even then spends 30% of their time on work that doesn’t bill.

The alternative you’ve been hearing about is automation. AI agents that handle the repetitive stuff. But nobody’s given you the real numbers on what that trade-off actually looks like in a consulting context.

Let’s break it down with the specifics that matter for firms doing $1M to $25M in revenue.

The True Cost of Hiring Another Consultant

When you hire, you’re not just paying salary. You’re paying for ramp time, for the work that doesn’t convert to revenue, and for the opportunity cost of senior people training instead of billing.

A mid-level consultant or project coordinator runs you $90K to $140K in salary, depending on your market. Add benefits, payroll tax, software seats, and overhead allocation, and you’re at $120K to $180K all-in. That’s the easy part.

The harder cost is utilization. In most consulting firms, a new hire takes four to six months to reach 60% billable utilization. During that ramp, they’re consuming senior time for training, shadowing client calls, and getting feedback on deliverables. One partner told me his last hire cost him 40 hours of his own time in the first quarter, time he would’ve billed at $350 an hour.

Even after ramp, your new consultant spends significant hours on non-billable work. Proposal writing. Research at the start of engagements. Internal knowledge management. Depending on your mix, that’s 20% to 40% of their week. You’re paying full freight for work that generates no immediate revenue.

Then there’s the project coordination layer. If you’re hiring someone to manage timelines, chase deliverables, and keep clients updated, you’re paying $70K to $100K for work that’s essential but hard to bill separately. Clients expect it as part of the engagement, so it comes out of your margin.

Add it up and a new hire costs you $150K to $200K in year one when you account for ramp, training drag, and non-billable hours. In year two, if they hit 65% utilization, you’re still carrying $80K to $120K in cost that doesn’t show up on an invoice.

What Automation Actually Costs

An AI agent doesn’t have a salary. It has a build cost, a run cost, and a maintenance cost. The numbers are different enough that most partners I talk to haven’t built the mental model yet.

Build cost for a production-grade agent ranges from $8K to $25K, depending on complexity and how much existing infrastructure you have. If you’re starting from scratch, you’re paying for the workflow design, the integration with your CRM and document systems, and the testing to make sure it doesn’t hallucinate in front of a client.

Run cost is the ongoing expense of the AI model, the data pipeline, and the hosting. For most consulting use cases, this runs $200 to $800 per month. A Proposal Generation Agent that gets used 15 times a month might cost you $400 in API calls and infrastructure. A Research Agent running daily might hit $600.

Maintenance is the part people underestimate. Agents need tuning. Clients change, your service offerings shift, and the agent’s outputs need to stay aligned. Budget 3 to 6 hours per month of technical time, or about $2K to $4K annually if you’re outsourcing it.

Total first-year cost for a well-built agent: $18K to $35K. Year two and beyond: $4K to $12K annually.

Compare that to the $150K+ you’re spending on a human doing similar work, and the math starts to make sense.

Where Agents Actually Fit in a Consulting Firm

Not every task is a good fit for automation. You’re not replacing the senior consultant who walks into a boardroom and reads the room. You’re replacing the 30 hours they spent the week before building the deck they’re presenting.

Proposal writing is the clearest win. A partner or senior consultant spends 20 to 40 hours on a major proposal. They pull case studies from old decks, rewrite capability statements, and build a pricing model from scratch. A Proposal Generation Agent pulls all of that from your past work, drafts the narrative, and hands you a 70% complete document in 20 minutes. You spend 3 hours refining it instead of 30 building it.

One firm I work with cut their proposal cycle from three weeks to five days. Same win rate, 75% less senior time per opportunity.

Research and synthesis is the other big one. Every engagement starts with secondary research. Industry trends, competitor moves, market sizing. It’s necessary work, but it’s repeated across clients and it doesn’t require judgment until the synthesis stage. A Research Agent runs structured searches, pulls sources, summarizes findings, and generates a one-page brief. The consultant reviews it, adds context, and moves to analysis. What used to take two weeks of junior time now takes two hours of senior review.

Knowledge management is the long-tail win. Your firm has produced thousands of pages of insight. Decks, reports, white papers, meeting notes. Almost none of it is searchable or reusable. A Knowledge Agent reads your entire corpus and answers questions across it. A partner prepping for a pitch can ask, “What did we recommend to manufacturing clients on supply chain risk in 2024?” and get a synthesized answer with citations in 30 seconds.

That capability doesn’t replace a hire, but it makes your existing team 20% more effective. If you’re a 10-person firm, that’s two full-time equivalents of capacity without adding headcount.

The Ramp Time Advantage

Here’s the part that doesn’t show up on a spreadsheet but changes the economics in a real way. A human hire takes six months to get productive. An agent is productive the day it goes live.

You don’t train an agent on your methodology. You encode your methodology into the agent. It doesn’t forget. It doesn’t leave for a competitor. It doesn’t need performance reviews.

When you win a new client, your team can spin up a Research Agent for that engagement in a day. When you lose a client, you don’t have underutilized headcount. The agent scales down to zero cost when it’s not running.

That flexibility matters more as you grow. A $5M firm that’s scaling to $8M has lumpy demand. Some quarters you’re slammed. Others you’re scrambling to keep people billable. Agents let you handle the peaks without carrying fixed cost through the valleys.

The Hybrid Model

The firms getting this right aren’t choosing between hiring and automating. They’re doing both, but in a different sequence.

Instead of hiring two consultants to handle growth, they hire one and build three agents. The consultant focuses on client-facing work and strategic analysis. The agents handle proposals, research, and knowledge synthesis. The consultant’s utilization stays above 70% because they’re not bogged down in pre-work.

One advisory firm I know hired a senior consultant at $160K and built a Proposal Generation Agent and a Research Agent for a combined $40K. The consultant billed 1,200 hours in year one because they weren’t spending 400 hours on non-billable work. The firm’s revenue per head went from $320K to $480K.

That’s the model. Agents handle the repeatable work. Humans handle the judgment calls and the client relationships. You get the capacity of two hires for the cost of one hire plus a modest technology investment.

If you’re trying to map this to your own firm, we built a worksheet that walks through the decision framework. Deploy Your First Business Agent gives you the cost model, the use case prioritization, and the build-versus-buy checklist. It’s a 20-minute exercise that’ll tell you whether automation makes sense for your next capacity decision.

What This Looks Like in Practice

Let’s walk through a specific example. A 12-person strategy consulting firm doing $4M in revenue. They’re turning down work because they don’t have the capacity to take on more than six active engagements at a time. The bottleneck is proposal development and engagement kickoff.

They have three options. Hire a junior consultant for $120K to handle research and proposal support. Hire a project coordinator for $85K to manage timelines and client communication. Or build two agents: a Proposal Generation Agent and a Research Agent.

They choose the agents. Build cost is $32K total. Run cost is $600 per month. They deploy both agents in eight weeks.

The Proposal Generation Agent cuts proposal time from 35 hours to 8 hours per opportunity. The firm goes from submitting one major proposal per month to three. Win rate stays the same, but they’re seeing three times the pipeline velocity.

The Research Agent handles engagement kickoff. What used to take a junior consultant two weeks now takes the agent two hours and a senior consultant four hours to review and contextualize. Engagements start faster, and the senior consultant bills those two weeks instead of supervising research.

Six months in, the firm has added $800K in revenue without adding headcount. They’re now hiring, but they’re hiring a senior consultant who can close and deliver, not a junior who would’ve spent half their time on work the agents now handle.

That’s a real example. The numbers are specific to one firm, but the pattern holds across the 30+ consulting firms we’ve worked with.

The Leakage You’re Not Measuring

Most consulting firms don’t track the cost of repeated work. You’re not measuring how many hours your team spends recreating the same analysis for different clients. You’re not counting how much senior time goes into proposals that don’t close.

We see this leakage run between $80K and $300K annually for firms in the $1M to $25M range. That’s not revenue you’re losing. It’s margin you’re paying for work that could be automated or reused. You can put a first number on it with our AI cost savings estimator before you book anything.

A Research Agent doesn’t just save time on one engagement. It builds a knowledge base that makes every subsequent engagement faster. A Proposal Generation Agent doesn’t just speed up one proposal. It captures your best case studies and pricing models so every proposal is better than the last.

The ROI isn’t just the hours saved. It’s the compounding effect of turning one-time work into reusable assets. If you want to see where that leakage is happening in your firm, the AI audit for consulting firms is built to surface it. It’s a 60-minute conversation that maps your workflows, identifies the repetitive work, and shows you what an agent would look like for your top three use cases.

When Hiring Still Makes Sense

Agents don’t replace judgment. They don’t replace client relationships. They don’t replace the senior consultant who can walk into a room and shift the conversation.

If you need more client-facing capacity, hire. If you need deep domain expertise that takes years to build, hire. If you need someone who can sell and close, hire.

But if you need capacity for proposals, research, project coordination, or knowledge synthesis, build an agent first. The cost is lower, the ramp is faster, and the flexibility is higher.

The firms that are winning this transition are the ones that see agents as a capacity layer, not a replacement strategy. They’re hiring fewer people, but the people they hire are more senior, more billable, and more effective because the agents handle the grunt work.

How to Make the Call

Here’s the framework I’d use if I were running a consulting firm today and trying to decide between a hire and an agent.

If the work requires judgment, context, or client interaction, hire. If the work is repeatable, structured, or based on past examples, automate.

If you need the capacity in the next 90 days, hire. If you can wait 60 days for an agent to be built and tested, automate.

If the work is lumpy and you’re not sure you’ll need it in 12 months, automate. If the work is steady and you’re confident in the pipeline, hire.

If the cost of getting it wrong is high, hire. If the cost of getting it wrong is a few weeks of iteration, automate.

Most firms I talk to end up doing both, but they start with agents for the high-volume, low-judgment work. Proposals. Research. Knowledge management. That frees up budget and capacity to hire for the high-judgment, client-facing work.

If you’re building with Claude or Codex right now, grab the free Working With Claude field guide. Thirty-two pages on the full ecosystem, Claude Code in depth, and how to roll agents out properly. Get the free guide.

The Next 12 Months

The consulting firms that figure this out in the next year are going to have a structural advantage. They’ll have higher revenue per head, better margins, and more flexibility to scale without adding fixed cost.

The firms that wait are going to keep hiring into the same capacity constraints, paying $150K per year for work that could be automated for $20K.

You don’t need to automate everything. You need to automate the right things. The repetitive work that’s burning senior time. The pre-engagement research that’s getting repeated across clients. The proposals that take 30 hours when they should take 3.

Start with one agent. Pick the use case that’s costing you the most senior time. Build it, test it, and measure the impact. If it works, build the next one. If it doesn’t, you’re out $15K and you learned something.

The cost of trying is low. The cost of not trying is a competitor who’s running leaner, pricing more aggressively, and closing faster because they figured this out before you did.

If you want to see where to start, see Omni for consulting firms. It’s a 60-minute audit that’ll show you the three highest-value automation opportunities in your firm, with cost models and build timelines. No deck, no sales pitch. Just the map you need to make the call.

The hire-versus-automate question isn’t binary. It’s a portfolio decision. The firms that treat it that way are the ones that’ll be running $10M operations with $5M cost structures. That’s the edge.