Week 22 Day 2: The Resume Is the Least Useful Part of Hiring

A resume is a marketing document. It is designed to present the candidate in the best possible light. Every bullet point is an achievement claim with no context about the surrounding reality. 'Led a team of eight engineers to deliver a platform migration ahead of schedule' could mean the candidate was a transformative leader, or it could mean the schedule was padded by three months and the team succeeded despite the candidate's leadership. The resume cannot tell you which one it was.

Here is what resumes can and cannot tell you. Resumes can tell you: where someone worked, what their title was, what technologies they listed, and how long they stayed. Resumes cannot tell you: how they treated their teammates, whether they finished what they started, how they responded to failure, whether they sought feedback or avoided it, or whether they elevated the people around them. Every attribute that matters for long-term team health is invisible on the resume. I use resumes for exactly one purpose: generating behavioral questions. If the resume says 'migrated legacy system to microservices architecture,' my question is not 'tell me about the migration.' My question is: 'What was the hardest decision you had to make during that migration, and who disagreed with you?' The first question invites a rehearsed narrative. The second requires real recall and honest reflection. I also look for what is missing from resumes. Gaps in employment are not red flags -- they are conversation starters. Short tenures are not disqualifying -- they are prompts to understand what happened. The most revealing resume feature is consistency of pattern, not impressiveness of individual entries.

The limited predictive validity of resume data is well-established in selection research. Cole, Rubin, Feild, and Giles (2007) found that resume evaluations had a validity coefficient of only 0.18 for predicting job performance, compared to 0.51 for structured interviews and 0.54 for work samples. Research by Brown and Campion (1994) specifically demonstrated that resume evaluations are susceptible to the same cognitive biases that affect other judgment tasks: the halo effect (one impressive entry influences evaluation of all other entries), anchoring (early information disproportionately affects overall impression), and prototype matching (evaluators compare candidates to a mental prototype of the 'ideal candidate' rather than to job-relevant criteria). The behavioral question technique described in level_2 is an application of the Behavioral Description Interview (BDI) methodology developed by Janz (1982), based on the principle that the best predictor of future behavior is past behavior in similar situations. Janz's research demonstrated that behavioral questions produce more valid and less fakeable responses than situational or hypothetical questions. The 'what is missing' approach reflects what Holmes (1892) famously illustrated as 'the curious incident of the dog in the night-time' -- the information value of absence, which in selection contexts often reveals more than the information that is present.