The standard hiring process, resume review, 60-minute interview, reference check, is optimized for large organizations with HR infrastructure. For a solo consultant making a first or second hire, it’s the most expensive and least predictive approach available.
Interviews measure interview performance: articulateness, confidence, how well someone presents themselves under a social evaluation dynamic. That’s a real skill, but it’s not the skill you’re hiring for. You’re hiring for output quality, communication style, and how someone behaves when the work is hard and the feedback is direct.
A paid test project measures the actual thing. One week, a real task, $200-500 in their pocket. You see their output. You see how they communicate during uncertainty. You see how they handle your feedback. That’s 80% of what you need to know.
Why Interview-Only Hiring Has a 40% Success Rate
Research on hiring consistently shows that unstructured interviews have low predictive validity, roughly 0.18 correlation with job performance in meta-analyses. That means interviews explain about 18% of the variance in how someone performs in the role. The rest is unexplained.
The reasons are well-established. Good interviewers assess things like confidence, narrative coherence, and social intelligence. These are real qualities, but they don’t reliably predict whether someone will produce excellent work, communicate well under pressure, or take feedback gracefully. They predict how someone performs under the specific conditions of an interview.
For a solo consultant, a bad hire is costly in ways beyond the monetary. You train them for 30 days, they underperform, you spend 30 more days working around them before you acknowledge the fit is wrong, and then you go through the hiring process again, having burned 60 days of management time plus the cost.
The test project approach changes the information set. You’re not adding a test project to the interview process, you’re replacing most of the interview process with a test project. The interview becomes a 20-minute screening conversation to confirm basic fit, followed by the test project, followed by a debrief call. That’s the full process.
The Test Project Structure
A good test project has four characteristics:
1. Real work, not a hypothetical. Give them an actual task that produces something useful, a piece of research, a first-draft document, an analysis, a design mockup. “What would you do if a client asked for X?” is not a test project. “Here’s a client situation, write the first draft of the recommendation section” is.
2. Clear scope with a defined deliverable. Write a brief. What is the task? What does done look like? What’s the format? What are the standards? A vague brief is a test of scope-clarification skills, which is sometimes useful to test, but only if that’s explicitly part of the evaluation. For most roles, give a clear brief and evaluate the output.
3. Fair pay. $200-500 for a week of real work is the range. Calculate it roughly at their expected hourly rate: a 5-hour task for someone at $60/hour is $300. A 10-hour task for someone at $50/hour is $500. Don’t do unpaid test projects, they filter out the confident candidates who have other options, leaving you with people who are desperate or unclear on the value of their own time.
4. Mutual evaluation built in. Frame the project explicitly as a two-way evaluation: “This is a chance for me to see your work and how you operate, and for you to see how I manage and brief work. The debrief call at the end is as much for you to evaluate whether this is the right fit as it is for me.” This framing is honest and produces better debrief conversations.
An unpaid test project filters for the wrong population: people who can’t afford to say no to unpaid work, or people who don’t value their own time enough to push back. Pay fairly and you attract the confident professionals who have options, which is exactly who you want to hire.
Running the Test Project Week
Day 1, Brief delivery: Send the written brief by 9am. Include: background context, the specific task, the deliverable format, the evaluation criteria, and the due date (end of day Friday or similar). Ask them to confirm receipt and flag any clarifying questions before proceeding.
The questions they ask before starting are evaluation data. A candidate who asks 2-3 well-targeted clarifying questions is showing good professional judgment. A candidate who asks no questions and dives in may produce a good output or may produce something that missed the point entirely. A candidate who asks 10 questions about every detail may struggle with ambiguity in real work.
Day 3, Optional check-in: Send a brief message: “Just checking in, any blockers or questions as you work through the project?” This creates a low-stakes opportunity to see their mid-project communication style. Do they give a useful update? Do they surface a question they should have asked earlier? Do they ignore the check-in?
Day 5, Delivery: They deliver the output. Review it against your brief and your quality standard. Note specifically:
- Does it address the brief?
- Is the depth appropriate?
- Is the format clean?
- What’s missing?
Then give them one round of feedback: 2-3 specific notes on what to adjust. This is the most important evaluation moment, their response to feedback is more predictive of long-term fit than their initial output.
Day 6-7, Revision and debrief scheduling: They revise based on your feedback. The quality of the revision tells you how they internalize notes. Then schedule the debrief call.
The Debrief Call
The debrief call is 20-30 minutes and serves three purposes: close the evaluation, gather their impressions, and make a decision.
Your side of the conversation:
“Here’s what I thought was strong about your work on this: [specific]. Here’s where I saw room for growth: [specific]. How did you find the project, what was harder or easier than you expected?”
Their answer to the last question is critical. Strong candidates:
- Identify the parts of the brief that were genuinely ambiguous
- Acknowledge the feedback you gave and explain how their thinking evolved
- Ask questions about how this type of work typically runs in your practice
Weak signals:
- “I thought it went great, no issues” (no self-reflection)
- Defensive responses to your critical feedback
- Questions that are entirely about pay and logistics rather than the work
Handling a Borderline Result
Most test projects fall into one of three categories:
Clear yes: Output is strong, process was clean, feedback was handled well, debrief was thoughtful. Hire.
Clear no: Output missed the brief significantly, communication was poor, feedback was met with defensiveness. Decline respectfully and note specifically what wasn’t a fit, this feedback is valuable for them and protects your reputation.
Borderline: Output was adequate but not impressive, or strong output with concerns about process or communication. This is the most common result for good candidates.
For borderline cases, ask the follow-up question at the end of the debrief: “Looking back at the project, is there anything you’d do differently with more context or time?”
A borderline candidate who identifies exactly the gaps you noticed, and explains why they made the choices they made, is demonstrating self-awareness and growth orientation. That’s a candidate worth hiring on probation: “I’d like to try a second engagement with more context on my standards. Here’s how I’d structure it.”
A borderline candidate who sees no room for improvement is showing you their ceiling. The initial output was their best judgment. That’s the information you needed.
What 80% Success Rate Actually Means
The 80% figure (vs. ~40% for interview-only processes) comes from practitioners who’ve tracked hire outcomes systematically. The mechanism is simple: you’re evaluating actual work product, actual communication under real conditions, and actual feedback responsiveness, not interview performance.
The failures in test-project hiring come from two sources:
-
The test project wasn’t representative. It was too easy, too hypothetical, or too different from real work. This produces false positives, candidates who ace the test and underperform in the role.
-
Ignoring process signals. A great output delivered with poor communication, missed check-ins, and defensive feedback responses is telling you something. If you hire for the output and ignore the process signals, you’ve partially undercut the value of the test.
Evaluate all four dimensions, output, process, communication, feedback handling, and your success rate will run well above interview-only processes. One week and $200-500 is the most efficient hiring investment you can make.
Ready to send stronger proposals?
Build, send, and track proposals in one place so follow-up is easier.
Start your free trial →
