How we helped a legal staffing company add $1.8M in annual EBITDA.

They needed their growing services to make up for a shrinking one.

The legal staffing company wanted to grow its earnings and strengthen the business ahead of a future sale. Its hourly document-review work was shrinking, and leadership wanted to launch new AI-related services while growing the lines that were already doing well.

Their earnings goal was to move from $5.4M to at least $7M in EBITDA, the measure of earnings before interest, taxes, depreciation, and amortization. But most of the services they wanted to grow depended on the same recruiters finding and placing the right attorneys.

Before choosing what to build, we looked at how those recruiters handled the requests already coming in. The company had demand for interim counsel, but it was losing work when another provider presented suitable candidates sooner.

Launching more services would put more requests through that same process, which meant we first needed to understand what was making it slow.

Their best recruiters knew who would fit, but the rest of the team couldn’t easily use that judgment.

We followed requests through the recruiting process and compared the work with past placements. Candidate information was scattered across spreadsheets and inboxes, so recruiters kept rebuilding searches from records they had to find and check by hand.

But putting those records in one place would only solve part of the problem. The most experienced recruiters knew which backgrounds suited particular clients, which past placements had worked, and when a candidate who looked suitable on paper would be a poor match.

That reasoning hadn’t been written down, and so the rest of the team couldn’t consistently use it. Hiring more recruiters would still leave new staff learning those decisions from the same senior colleagues.

We needed to make both the records and the matching criteria available in the work itself, so a recruiter could prepare a shortlist without starting from memory each time.

We built the assistant around the decisions that had already worked.

The company bought an applicant tracking system, and its vendor partner moved the candidate records into it. That gave the recruiting team one place to find and maintain the information, while we focused on the matching judgment that a database alone wouldn’t provide.

We worked with their recruiting leads to capture the criteria they used, then built an assistant that applied those criteria to the company’s placement history. It brought together previous assignments, extensions, client feedback, availability, and rate information, with the supporting evidence beside each suggested candidate.

The recruiter could then review why someone appeared on the list, check the fit, and use a drafted note to prepare the submission. The recruiter still decided whom to recommend, but no longer had to reconstruct all of the supporting information by hand.

We tested the rankings against past placements and then used the assistant on live requests, so the team could assess both the suggestions and how they worked in an actual search.

A missing status showed us where the safeguard needed to change.

During the pilot, the assistant ranked a candidate whose conflict-clearance status was blank. An administrator caught it before the shortlist went to the client.

We’d written the rule to stop a candidate with a missing status, but the imported record contained an empty text field that the rule hadn’t treated as missing. We told the company’s leaders that day and changed the rule to require an affirmative clearance before a candidate could appear in a shortlist.

The administrators completed the missing records, and the submission checklist remained in place. That gave the process both a stricter rule in the software and a person checking the shortlist before it left the company.

Their project managers already knew how to deliver the new service.

The company also wanted to offer validation of AI-assisted document review. Its project managers had already done this work on client matters, but each assignment required them to assemble the method and documentation again.

We helped them turn that experience into a written method and built a generator that prepared the documentation from the matter’s sampling results. The matching assistant also used recorded reviewer performance to help the team find attorneys suited to the validation work.

That gave the company a repeatable package it could staff and sell for a fixed fee. The client’s team sold and delivered the service, while the method, matching support, and documentation we helped put in place made it easier to repeat.

We taught their team to keep improving the system.

The company needed to adapt the matching criteria as new requests arrived, so we trained its administrators, recruiting leads, and systems owner to maintain the records, change approved criteria, and check the results.

After handover, an administrator added a new specialist skill and its matching checks with a lead’s approval, without our involvement. A project-management leader also extended the validation template when a client needed an additional field.

The team could now make those changes itself, with its systems owner and vendor partner responsible for ongoing technical support.

The same recruiting team completed 85 more placements.

At the follow-up twelve months after delivery, annual interim placements had risen from 250 to 335. That was 34% more placements with the same recruiting team, turning more of the demand their commercial team brought in into work the company could bill.

The validation service had also billed $2.2M across twelve matters, giving the company another source of earnings as hourly document-review work continued to decline.

The company reported $7.2M in trailing-twelve-month EBITDA, up from $5.4M. A partner confirmed those results in our follow-up, including the placement growth and validation billing. The company had added $1.8M in annual EBITDA and surpassed its $7M earnings goal.

The gain reflected the wider business: more placements and validation work, growth in contracts work, the continued decline in hourly review, and changes in costs. The validation fees contributed to the reported earnings.

Growth no longer depended on rebuilding each search from memory.

The company came to us with new services to launch and an earnings goal to reach. We found that the growing work depended on matching knowledge that the existing team couldn’t easily share, then built the tools and standards that let more recruiters use it.

Their experienced staff kept making the decisions that needed their judgment, while the team gained a repeatable way to prepare shortlists, deliver validation work, and maintain what we’d built.

If you’re trying to grow your own company, start with your goal, look at what it'll take to achieve it, and then identify what's keeping you from taking action. Just remember that every company is different and that matching knowledge held by a few senior recruiters was just their bottleneck and it might not be yours.