Article Summary
Recruitment firm owners ask whether their database is paying off and run cost-per-placement math. Wrong test. The real question: what percentage of revenue is your delivery cost — recruiters plus the tools they need to source, screen, and place — and does revenue per recruiter rise after the tool? Reclassify your tech stack into COGS, run it through 60-15-15, and the recruitment firm database ROI answer gets sharp.
The ROI question most recruitment firm owners are asking is the wrong one
Most owners measure database ROI like this: cost of the tool divided by placements influenced. That math hides the only number that matters — whether revenue per recruiter went up after the tool was added.
According to SourcrLab, recruiters spend roughly 30 to 40 percent of their time on manual tasks — data entry, tab switching, re-typing notes. Owners buy a database, an enrichment tool, or an automation layer to claw that time back. Then twelve months later, revenue per recruiter looks identical to where it started. The time savings got absorbed somewhere — usually into more sourcing, more outreach, more activity that doesn’t translate into placements.
I run a fractional CFO practice for service founders doing $1M to $20M, and the recruitment firm owners I work with all run into the same trap. They evaluate tools individually instead of asking the structural question: is the entire delivery layer — humans plus tools — generating enough revenue per dollar spent?
Bennett Financials is a fractional CFO and tax planning firm that helps service business founders doing $1M to $20M diagnose growth bottlenecks, fix margins, and build businesses worth selling. The framework I use to answer the database ROI question isn’t a recruiting framework. It’s a margin framework called 60-15-15. And the first thing it does is force you to classify costs correctly — which is where most recruitment firms have a quiet, expensive problem.
Why your tech stack belongs in COGS, not G&A
Walk into a recruitment firm’s books and you’ll usually find the ATS, LinkedIn Recruiter seats, sourcing tools, parsers, and enrichment under “Software” or “Subscriptions” — sitting in G&A.
That’s wrong. Here’s the rule: anything required to deliver the service belongs in COGS.
Think of it like this. A recruitment firm without LinkedIn Recruiter and an ATS isn’t running lean — it’s not running. Those tools aren’t back-office overhead. They’re delivery infrastructure. Your recruiters can’t make placements without them. By the same logic an HVAC company puts diagnostic equipment in COGS, your firm puts the database in COGS.
What happens when you reclassify? Three things shift, and they all matter.
First, gross margin drops on paper. The truth shows up. If your “real” GM is 48% instead of the 62% your books showed, you have new information about how scalable the business actually is.
Second, G&A drops, which is what it should look like. The 60-15-15 framework targets ≤15% G&A. If you’ve been hiding delivery costs in G&A, you’ve also been hiding the fact that your real G&A is probably already at target.
Third — and this is the one that fixes the ROI question — your COGS line now contains everything required to make a placement. Now you can run the labor efficiency test that actually answers whether the database is paying off.
Picture a $4M recruitment firm owner running an ATS, five LinkedIn Recruiter seats, a sourcing tool, an enrichment subscription, and a parser. Financial Models Lab estimates around $1,200 monthly for a recruitment agency’s core tech stack covering an ATS and LinkedIn Recruiter licenses — but that’s per recruiter, not the firm total. Multiply by five recruiters and you’re at $72K to $120K a year in tools alone, before you’ve added enrichment, parsing, or any of the AI sourcing layers most firms have stacked on since 2023.
That number doesn’t belong in G&A. It belongs sitting next to the recruiter salaries it enables.
The 60-15-15 test for recruitment firm tech ROI
The framework is simple. 60% gross margin. 15% S&M. 15% G&A. 30% operating margin. Below 55% GM and scaling makes you busier, not wealthier.
The metric that does the work for tech ROI is labor efficiency:
Revenue ÷ all delivery labor (including delivery tools) ≥ 3.5x
Run the $4M founder through it.
- Revenue: $4M
- 5 recruiters fully loaded: ~$600K. Paraform’s cost analysis puts fully loaded annual cost per recruiter at $175,000 to $190,000 for tech recruiters; agency recruiters tend to land lower at the $100K to $130K range with benefits and overhead.
- Delivery tools (5 seats of LinkedIn Recruiter alone runs ~$50K, plus ATS and enrichment): ~$120K
- Total delivery cost: ~$720K
- Labor efficiency: $4M ÷ $720K = 5.5x
That firm is fine. The tools are paying off. But change one variable — say recruiter compensation runs $700K instead of $600K because the team is overpaid relative to placements — and labor efficiency drops to 4.9x. Add another sourcing tool and another enrichment seat and you’re at 4.5x. Add a second ATS because nobody migrated off the old one and you’re at 4.0x. Layer one more “AI sourcing” tool that nobody actively uses and you’re at 3.5x — the floor.
Three points to notice. The tools didn’t kill the firm — stack creep did. Each individual tool ROI calculation looked fine. The cumulative number is what the cost-per-placement math will never catch. And below 3.5x, you’re not running a recruitment firm. You’re running a tools subscription that happens to employ recruiters.
Why 3.5x as the floor? Below that, the math stops working — there isn’t enough margin left after delivery to fund sales, leadership, and a 30% operating margin. At 5x and above you have room to invest in growth. The 3.5x band is where the framework holds; below it, every other line item gets squeezed.
The three signals your database ROI is fake
Across the recruitment firms I work with, three patterns show up when the tech stack stops paying off and nobody notices.
Signal 1: Revenue per recruiter is flat twelve months after the tool. You bought the database to claw back the 30-40% manual time. Revenue per recruiter should be up. If it’s flat, the time savings didn’t compound into more placements — they got absorbed into more activity that didn’t close.
Signal 2: You added the tool. You didn’t remove anything. Stack creep is the silent margin killer for recruitment firms at $3M-$7M. The new sourcing tool doesn’t replace LinkedIn Recruiter. The new ATS migration stalls and you pay for both for fourteen months. The enrichment tool adds data but the old enrichment subscription has another year on the contract. Every individual decision was rational. The cumulative result is a 5-point drag on gross margin.
Signal 3: Your close rate didn’t change. This is the one most owners miss. The tool surfaces more candidates, the recruiters present more shortlists, and close rate stays at 32%. The tool didn’t fix the bottleneck. The bottleneck wasn’t sourcing — it was pricing or positioning upstream of sourcing.
Tools don’t fix pricing problems. If your close rate is above 60%, you have room to triple your fees, and no database in the world will surface that for you. If your close rate is below 30%, the problem is what you’re selling and how — not who you’re calling.
That’s the part the cost-per-placement math will never see, because it isolates the tool from the upstream economics that actually drive whether a placement happens.
If your firm is also overpaying tax — and most service firms at $3M-$10M are — that’s another silent margin issue. The tax planning side of what I do typically uncovers $50K to $300K annually for recruitment firms in this revenue band, which can fund a margin transformation without cutting a single tool.
Want to know where your business sits against the 60-15-15 standard? The Scale-Ready Assessment runs your actual numbers, builds a custom tax strategy, and produces a full enterprise value report. Free for US-based service businesses doing $1M–$20M. Book your free Assessment — 15 spots per month.
How to actually measure recruitment firm database ROI
Five steps. Run them in order.
Step 1: Reclassify. Pull every line item required to make a placement out of G&A and into COGS. ATS, LinkedIn Recruiter, sourcing tools, enrichment, parsers, automation layers, AI sourcing tools, candidate assessment fees. If a recruiter needs it to do the work, it’s COGS.
Step 2: Calculate labor efficiency including tools. Revenue ÷ (delivery labor + delivery tools). If you’re at 3.5x or below, the tools aren’t paying off as a category — they’re absorbing capacity.
Step 3: Pull trailing 12 months of revenue per recruiter, before and after each major tool added. This is the only honest test of whether a tool drove output. If revenue per recruiter is flat or down after the tool went live, the time savings got absorbed.
Step 4: Run the close rate diagnostic. If close rate is 60% or higher, your bottleneck isn’t sourcing — it’s pricing. Adding a database won’t fix it. Per the BF 60-15-15 framework, close rate at 60%+ means you can double or triple fees; close rate at 30-40% means pricing is right and the issue is somewhere else.
Step 5: Decide what to consolidate. With a clean COGS line and labor efficiency calculated, the consolidation calls get obvious. Two ATS subscriptions running in parallel for ten months. An enrichment tool nobody logs into. A parser that’s bundled into the ATS you already have.
The output of those five steps is a number you can actually defend — to yourself, to a prospective buyer, and to your team.
Case study — Eden Data: clean classification changed the decisions
A cybersecurity consulting firm launched in early 2021 with zero revenue. The founder needed finance leadership from day one — not just bookkeeping.
The pain: the founder initially expected spreadsheets and year-end taxes only. Standard bookkeeping. The kind of finance setup most service firms have when they’re under $1M.
What I did: embedded fractional CFO from the startup phase. Taxes, forecasting, equity and compensation guidance, ongoing decision support. Got the P&L classified correctly from the beginning — delivery costs in COGS, infrastructure in G&A, no muddled middle. Available via text when fast decisions needed to be made.
Results: scaled from $0 to ~$300K MRR with finance operating as always-on decision support. Pricing decisions, cash planning, hiring timing, and strategic tradeoffs all backed by a P&L that told the truth.
The friction: the founder had to recalibrate expectations of what strategic finance actually looks like. The shift from “reporting” to “embedded decision support” took deliberate effort on both sides. There were several months of recalibration before the founder stopped sending questions to the bookkeeper that should have come to me.
The key insight for recruitment firms: when delivery costs (people plus tools) are classified correctly from the start, every hiring and tooling decision gets sharper. You’re not arguing about whether a tool is worth it in the abstract — you’re looking at a labor efficiency number and deciding whether it gets better or worse with the tool added.
The recruitment firms I work with — I serve recruitment firms specifically as part of the practice — usually arrive after they’ve stacked tools for two or three years and the GM is sub-50%. The reclassification work is the unlock that makes every subsequent decision easier. It’s also the foundation for building a business worth selling at a defensible multiple.
What this means for your enterprise value
A recruitment firm that can show a buyer a clean P&L — delivery costs in COGS, real gross margin, labor efficiency above 3.5x — sells at a different multiple than a recruitment firm with $200K of “software subscriptions” sitting in G&A that nobody can defend in diligence.
Across 5,000 benchmarked companies, the score-to-multiple band runs from 2.76x EBITDA at the bottom to 6.27x at the top. Same revenue, same EBITDA, different multiple. A buyer paying 6.27x is paying for predictability and clarity. A buyer paying 2.76x is buying a job and discounting accordingly.
If you’re the one personally deciding which tools to renew every year, that’s also an owner dependence problem — and owner dependence is the single biggest lever in the multiple. The buyer is asking: can this business run without you? “I personally manage the tech stack” is the answer that drops the multiple by two points.
Book a free Scale-Ready Assessment — three deliverables: full 60-15-15 financial diagnostic, a tax plan, and an enterprise value report showing your current multiple and the gap. 15 spots per month.
