Two senior roles that had been open all year

Template — replace with a real engagement. The structure below is what converts; substitute your own client, numbers and timeline before publishing.

A Bulgarian company with its own payments product, roughly 120 people, most of them in Sofia. They needed two senior backend engineers with real production experience of payment flows and the regulatory surface around them. Both roles had been open for most of a year. Their internal recruiter had been through two agencies before us, and had a folder of CVs from people who were not close to the level required.

The pool for that profile in Sofia is small and mostly known to itself. Everyone qualified was employed, most had already been approached about these exact roles, and a few had formed an opinion about the company from a previous unsuccessful interview. Approaching the same people again with the same pitch was not going to produce a different answer.

When we mapped what had happened to the candidates who had come close, the pattern was not sourcing. It was the process. Four interview rounds spread across five to six weeks, a take-home exercise that ran to a full weekend, and a compensation band set against a salary survey that was two years old. Candidates were not rejecting the company. They were accepting something else while still in the pipeline.

The first deliverable was not a shortlist. It was a written argument that the roles were not fillable as specified, with the evidence, and three options: raise the band, cut a level of seniority and hire someone who grows into it, or keep both and accept a much longer search. They chose to raise the band and compress the process. That conversation is the part of this engagement that actually mattered.

What changed

• Two interview stages instead of four, with the hiring manager and the CTO in the same session, and a decision committed within TBC working days of the final round.
• The take-home replaced with a paid, time-boxed review of a real pull request from their codebase — closer to the job, far cheaper for a candidate with a family.
• A rewritten market brief, priced against what these engineers were actually being paid in Sofia at the time, not against last year’s survey.
• A named-approach list agreed with the client, so we were not re-contacting people who had already been burned by the old process without acknowledging it.

Sourcing itself was unglamorous: direct outreach, referrals from engineers we had placed elsewhere, and a small number of people who had left the company years ago on good terms. We presented TBC candidates in total. Presenting more would have meant presenting people we did not believe in.

Both roles were filled within TBC weeks of the brief being rewritten, after roughly TBC months of no progress. The time from first contact to signed offer averaged TBC days across the two hires.

The more useful outcome is that the client kept the shortened process for every senior role afterwards and filled the next one without us. We are comfortable saying that out loud. An agency that only ever solves the sourcing symptom gets to run the same search again next year, which is good for the agency and bad for the client.

Both hires were still in place at TBC months. Cost was TBC, against TBC months of an unshipped roadmap item that the roles existed to deliver.

Before publishing: replace this paragraph with what actually went wrong on the engagement and how you handled it. A case study in which nothing went wrong reads as marketing and buyers discount it.