You hire a candidate with a stellar resume. They went to a good school. They spent three years at a top tech company. They interviewed well. Six months later, they’re underperforming. Twelve months later, they’re gone.

You’ve just spent $100,000 on salary, benefits, onboarding, and training. They produced maybe six months of real value. The team had to compensate for their underperformance. Two colleagues got distracted managing them. The replacement hiring process costs another $15,000. When you do the math, this one bad hire cost you $50,000 to $150,000 in direct and indirect costs.

This isn’t a rare edge case. Across all companies, roughly 10% of new hires don’t work out. For companies that don’t invest heavily in hiring quality, the number climbs to 20-30%. That means if you’re hiring ten people a year, one or two of them are costing you $100K+ each.

The root cause isn’t mysterious. Let’s discuss why your hiring process keeps failing.

The Modern Hiring Stack: How to Actually Build It

Most companies know their hiring process is broken. Few know how to fix it in a way that sticks. The answer isn’t a new platform or a bigger recruiting budget — it’s a structured sequence of steps, executed consistently, that replaces gut feel with ground truth. And the companies that figure this out don’t just reduce bad hires. They build teams that compound over time — where each good hire makes the next one easier, where institutional knowledge accumulates instead of walking out the door every 14 months.

The problem is that most hiring advice is abstract. “Be more structured.” “Check references thoroughly.” “Assess for culture fit.” None of that tells you what to actually do differently on Monday morning. This is an attempt to fix that.

Start With the Work Sample — Before the Interview

Most companies put interviews first. That’s backwards.

By the time a candidate sits down with your team, you’ve already formed an impression from their resume. That impression creates bias that’s almost impossible to overcome, even in structured settings. You’re not assessing the candidate anymore — you’re assessing whether they confirm the story you’ve already told yourself about them.

The fix is simple in concept and underused in practice: ask the candidate to do the job before you spend three hours talking about whether they can do the job.

A work sample is a time-limited, role-relevant exercise that mirrors actual job tasks. For an engineer, that might be a code review, a debugging problem, or a system design question. For a marketer, it might be analyzing a real campaign dataset and presenting a recommendation. For a sales hire, it’s a mock discovery call with an internal stakeholder playing the prospect. For a manager, it’s reviewing a team conflict scenario and proposing how they’d handle it.

The goal isn’t to trick the candidate or put them through an endurance test. It’s to observe them doing the actual work, which tells you things that no resume or interview ever will. How do they approach ambiguity? Do they ask clarifying questions or make assumptions? Do they communicate their thinking or just deliver an answer? How do they handle feedback mid-exercise?

Research consistently shows that work samples are 50–60% predictive of job performance. Resume credentials sit at 10–15%. That’s not a marginal difference — it’s a different category of signal entirely. Platforms like TestGorilla and Codility have made it significantly easier to implement pre-built assessments across roles without building everything from scratch, though the most effective work samples are still custom-designed to mirror the actual role.

One practical note: compensate candidates for meaningful work sample time. If you’re asking someone to spend three hours building something for you, pay them for it. It signals that you value their time, it attracts candidates who are serious, and it removes a barrier that otherwise filters out strong candidates who have less flexibility.

Run Structured Interviews — But Know What You’re Assessing

Once you’ve seen someone do the work, the interview shifts from “can they do this job?” to “will they thrive here?” Those are different questions, and they require different conversations.

Structured interviews mean the same core questions, asked in the same order, scored against the same rubric, for every candidate in the same role. This sounds mechanical. It isn’t. Structure doesn’t mean rigid or robotic — it means disciplined. You can ask nuanced behavioral questions and still score them consistently. The discipline is in the scoring, not the tone.

The behavioral question format — “Tell me about a time when…” — exists for a reason. Past behavior is a better predictor of future behavior than hypothetical responses to hypothetical scenarios. “What would you do if a teammate missed a deadline?” tells you almost nothing. “Tell me about a time a teammate missed a deadline that affected your work, and walk me through what you did” tells you a great deal.

What you’re listening for is specificity. Candidates who are strong in a given area tend to have immediate, concrete examples. They remember the details because they actually lived through the situation. Candidates who are weaker in an area tend toward generality — they talk about what they “typically” do or what they “believe in” rather than what they actually did. That pattern is consistent enough to be a reliable signal once you know to look for it.

Involve multiple team members in the interview process, and debrief before anyone hears each other’s scores. Group debriefs where the hiring manager speaks first are useless — everyone anchors to the senior person’s opinion and the independent signal is lost. Score independently, share scores simultaneously, then discuss where there’s disagreement and why. Disagreement is often where the most useful insight lives.

Reference Checks Are Where Most Companies Quietly Fail

Reference checks are treated as a formality at most companies. Candidates submit two or three names they’ve pre-selected, the recruiter makes a couple of ten-minute calls, hears nothing alarming, and the box gets checked. That’s not a reference check. That’s confirmation bias with a phone call attached.

A real reference check looks very different. You’re calling three to five people — ideally including direct managers, not just peers the candidate chose — and you’re asking behavioral questions that are hard to spin. Not “would you recommend this person?” but “in what situations would this person struggle?” and “what would you tell their next manager that would help them get the best out of this person?” and “if you could change one thing about how they work, what would it be?”

The difference in what you learn is dramatic. Most references are happy to give a positive endorsement. Far fewer are willing to fabricate specific examples. So when you ask someone to describe a time the candidate received critical feedback and how they handled it, you get real information — or you get a very revealing pause.

Research from the Harvard Business Review shows that references are far more forthcoming when interviewers frame questions as “helping the candidate succeed in their next role” rather than evaluating them. That framing is also simply more accurate — you’re trying to understand how to manage and develop this person, not just whether to hire them.

What you’re listening for across multiple references is consistency. If three different people, independently, describe someone as excellent at execution but prone to conflict when given ambiguous direction, that’s not an outlier — that’s a pattern. A single reference saying something slightly negative might be a personal dynamic. Three references describing the same limitation in different words is the truth.

Also worth noting: the references people don’t offer are sometimes more informative than the ones they do. If a candidate has had five managers in eight years and can only offer one as a reference, it’s worth asking why.

Define 90-Day Milestones Before the Offer Letter Goes Out

Most new hires walk into their first week without a clear picture of what success looks like at 30, 60, or 90 days. Their manager has a vague sense of it. The new hire has an even vaguer one. And so the first quarter becomes a period of alignment that should have happened before the offer was signed.

The 90-day milestone framework is straightforward: before you make an offer, define what the hire should be able to do or have accomplished by day 30, day 60, and day 90. Be specific. Not “getting up to speed on the product” but “able to independently run a customer onboarding call without manager oversight.” Not “building relationships with the team” but “has completed one-on-ones with all eight direct stakeholders and has a documented understanding of each team’s current priorities.”

This exercise does something valuable independent of the hiring process: it forces the hiring manager to get clear on what the role actually requires. Most managers have never written down what a successful first 90 days looks like. The act of writing it down often surfaces misalignment within the leadership team about what they’re actually hiring for — which is better to discover before the hire starts than after.

Once the hire begins, the 90-day milestones become the backbone of weekly one-on-ones. Not in a punitive way — in an aligned way. “Here’s where you are against the milestones we agreed on. Here’s what I’m seeing that’s going well. Here’s where I think you need support.” That conversation is far more useful than a generic check-in, and it catches mismatches early, when they’re still fixable, rather than at the 12-month performance review when they’ve calcified.

If someone is consistently behind their 90-day milestones by week six, you know. You don’t have to wait until they’ve been underperforming for six months while their manager quietly hopes things will improve. Early intervention is almost always more humane and less expensive than late intervention.\

The Retention Payoff Nobody Puts in the Spreadsheet

Here’s the part that gets left out of most ROI analyses on hiring quality: structured hiring doesn’t just reduce bad hires. It increases retention across all hires, including the good ones.

When you hire someone through a process that involved a real work sample, structured interviews with their future teammates, deep reference calls, and a clear 90-day plan, that person walks in differently than someone who cleared two unstructured interviews and a quick resume review. They’ve been genuinely assessed. They know what they’re walking into. Their teammates have already invested in their success. Their manager has articulated what a good first quarter looks like.

That’s a fundamentally different starting point, and it produces fundamentally different outcomes. According to SHRM, employees who go through a structured hiring and onboarding process are significantly more likely to still be with the company at the 18-month mark — a milestone that matters because the first 18 months is when the majority of early-tenure departures happen.

Team involvement in hiring also creates accountability that persists after day one. When three colleagues participated in the interview process and gave positive feedback, they have a stake in that person succeeding. That’s not a small cultural shift. It moves the posture from “let’s see how the new person does” to “we picked this person together, and we’re going to help them win.” That kind of implicit commitment shows up in how teams onboard new members, how quickly they share context, and how willing they are to invest time in someone who’s still finding their footing.

The compounding math is worth sitting with. A company that moves from a 10% bad hire rate to a 5% bad hire rate while simultaneously improving 18-month retention by 30% across all hires isn’t just saving on replacement costs. It’s building something that resume-based hiring quietly destroys — institutional knowledge, team cohesion, and the kind of trust between colleagues that makes organizations actually function well under pressure.

Structured hiring takes longer to execute. It requires more coordination. It asks hiring managers to do real work in a part of their job they usually treat as secondary. And it pays for itself — in reduced turnover, stronger teams, and the compounding advantage of consistently bringing in people who were genuinely assessed and genuinely fit — many times over.

The process isn’t the obstacle. The willingness to treat hiring as a core operational discipline, rather than something that happens between real work, is.

Ready For Change?

The cost of bad hiring is concrete and measurable. The path to better hiring is clear. The ROI is immediate.

Start small: implement structured interviews and deep reference calls for your next five hires. Track their 90-day performance and 18-month retention against your baseline. Once you see the improvement, you’ll have the data to justify rolling out the process company-wide.

Your next hire doesn’t have to be another bad hire. Structured, behavior-based hiring can change that.

Learn More