The Problem
A head of talent acquisition at a global IT services company had what seemed like a reasonable workflow:
Source 100 candidates
Send coding assessment
Wait for submissions
Interview the top performers
Hire
Simple. Clean. Efficient on paper.
In reality? 30% completion rate.
For every 100 assessments sent, only 30 came back. Not 30 good ones. Just 30 total submissions. The other 70 candidates vanished into the void.
"We'd spend weeks building a pipeline, send out assessments on Friday, and by Monday we'd have a graveyard. Crickets. Radio silence."
The numbers were brutal. To hire 5 engineers, they needed to source 300+ candidates just to account for the assessment dropout. The entire top of their funnel was being burned by a single step.
The Expensive Band-Aid
What do you do when your filter eliminates 70% of your pipeline through pure apathy?
You hire more recruiters.
The company doubled their sourcing team. More LinkedIn licenses. More Naukri subscriptions. More recruiters making more calls to build bigger pipelines to feed the same broken assessment process that would waste most of it.
Six-figure investment. Same 30% completion rate.
The head of TA described the math to me with the kind of exhaustion that comes from explaining obvious problems to people who don't want to hear them:
"We were spending ₹15 lakhs annually on sourcing tools and recruiter salaries to generate candidates who wouldn't complete a 2-hour coding test. The assessment was costing us more than the salaries of the people we were trying to hire."
The Lightbulb Moment
Then someone asked the obvious question nobody had asked: What if we flip the order?
Instead of assessment-first, what if they tried interview-first?
The experiment was simple. Take 100 candidates. Interview 50 of them first, then send assessments. Send assessments first to the other 50, then interview.
The results weren't subtle:
Assessment-first group: 30% completion
Interview-first group: 70% completion
Same candidates. Same assessment. Same company. The only variable was sequence.
"Suddenly candidates who wouldn't touch our coding test were completing 6-hour take-homes because someone had actually talked to them first. It was like discovering fire."
The Psychology
Why does this work?
An interview creates reciprocal investment. When a real human spends 30 minutes talking to a candidate about their background, asking about their projects, explaining the role and the company, something changes. The candidate stops seeing the assessment as spam from a random company and starts seeing it as the next step in a conversation they're already having.
The assessment transforms from "Why should I do free work for strangers?" to "This person seems genuinely interested in hiring me. I should show them what I can do."
A recruiter at a unicorn fintech who discovered the same pattern put it more bluntly: "Nobody completes assessments for companies they don't care about. But after a good screening call, they care."
The Hidden Cost
Here's the part that makes this story a horror story rather than just a process optimization:
They knew the 30% completion rate was a problem for eighteen months before they tested the obvious solution.
Eighteen months of budget meetings about hiring more sourcers. Eighteen months of frustrated engineers waiting for candidates. Eighteen months of recruiters burning through LinkedIn connections to feed a broken funnel.
"We optimized everything except the thing that was actually broken. We bought bigger nets instead of fixing the hole in the boat."
The interview-first approach didn't just improve completion rates. It improved candidate quality. Candidates who completed assessments after interviews were more likely to accept offers, more likely to show up on start dates, more likely to succeed in the role.
Why? Because the interview had already done the two-way filtering that the assessment couldn't do. Mutual interest was established. Context was shared. Expectations were aligned.
The Broader Pattern
This isn't an isolated story. The "assessment graveyard" problem is everywhere.
A CTO at a fast-growing payments company told me they abandoned take-home assignments entirely after tracking completion rates. "We'd send coding challenges on Monday and spend the rest of the week chasing candidates to submit them. The amount of recruiter time spent on follow-ups was insane."
A senior recruiter at a major enterprise search company described the same dynamic: "High-volume, junior roles where multiple interviewers need coordination? The assessment becomes a black hole. Candidates see it as extra homework from a company that hasn't earned their time yet."
The pattern is always the same: companies treat assessments as filters when they should be treating them as conversations.
The Real Learning
The learning isn't "do interviews before assessments." Some companies need assessments first for volume reasons. Some roles require technical screening before human time investment.
The learning is simpler and more fundamental: every step in your hiring process either builds momentum or kills it.
The assessment-first approach killed momentum. Candidates felt evaluated before they felt engaged. They were being tested by strangers for an opportunity they didn't understand at a company they'd never spoken to.
The interview-first approach built momentum. By the time candidates reached the assessment, they were already bought into the process. They had context. They had motivation. They had a relationship, even a small one.
Your hiring process is a sales process. Every interaction either moves the candidate toward "yes" or pushes them toward "no." The companies that understand this design processes that compound interest rather than create friction.
The ones that don't wonder why they need to source 300 candidates to hire 5.
This is [2/n] in a series about the strange, broken, and occasionally enlightening things that happen inside hiring processes. Based on 60+ conversations with recruiters, hiring managers, and engineering leaders across FAANG, hypergrowth startups, and Fortune 500 companies.