Most founders don't have a shortlisting process. They have a pile of resumes, a gut feeling, and a hiring decision they're not totally confident in. That's how you end up making expensive mistakes.
Shortlisting candidates isn't complicated, but it does require a system. This post walks you through a repeatable process you can use for every hire — whether you're bringing on your third employee or your fiftieth.
What shortlisting actually means
Shortlisting is the step between "applications close" and "interviews begin." You start with 40, 80, maybe 200 applicants and narrow it down to the 4–8 people worth having a real conversation with.
Done well, shortlisting saves you from scheduling 20 interviews for a role that only needs 5. Done poorly, it lets bias and noise drive out strong candidates before they ever get a shot.
The goal is simple: identify the people most likely to succeed in the role, as efficiently as possible.
Step 1: Define your criteria before you look at a single resume
This is where most small teams go wrong. They open the applicant pile and start reading, making up criteria as they go. By resume 30, they've completely shifted what they're looking for.
Before you open a single application, write down your criteria. Be specific.
Must-haves are non-negotiable. If a candidate doesn't have these, they're out — full stop. Examples:
- 3+ years of B2B SaaS sales experience
- Based in or willing to relocate to Austin, TX
- Can start within 6 weeks
Nice-to-haves are differentiators. If two candidates are otherwise equal, these break the tie. Examples:
- Experience selling into HR teams
- Has used HubSpot before
- Background in a startup environment
Aim for 3–5 must-haves and 3–5 nice-to-haves. More than that and you're describing a unicorn. Less than that and you're not screening for much.
Write these down in a shared doc or spreadsheet before anyone on your team touches the applications. This protects you from scope creep and keeps evaluation consistent if more than one person is reviewing.
Step 2: Do a fast pass — the 60-second filter
With your criteria in hand, do a rapid first pass through all applications. You're not evaluating deeply here. You're just sorting into three buckets:
- Yes — clearly meets must-haves, worth a closer look
- Maybe — unclear, missing some info, could go either way
- No — clearly missing a must-have
At this stage, spend no more than 60 seconds per application. You're scanning, not reading. Look for the must-haves. If they're obviously missing, move on.
A realistic distribution for a decent job post: 50–60% will be clear No's, 20–30% will be Maybe's, and 10–20% will be Yes's. If your Yes pile is enormous, your must-haves aren't strict enough. If it's empty, your job post might be scaring people off or your requirements are unrealistic.
Step 3: Score your Yes and Maybe pile
Now go back through the Yes and Maybe buckets with a simple scoring rubric.
Build a spreadsheet with one row per candidate and columns for each of your criteria. Score each criterion on a 1–3 scale:
- 1 = doesn't meet it
- 2 = partially meets it
- 3 = clearly meets it
For must-haves, any score of 1 should probably knock them out. For nice-to-haves, you're just adding up differentiation.
Here's a real example for a sales hire:
| Candidate | B2B SaaS exp (must) | Austin-based (must) | HubSpot (nice) | Startup exp (nice) | Total |
|---|---|---|---|---|---|
| Jordan M. | 3 | 3 | 2 | 3 | 11 |
| Sam T. | 3 | 3 | 1 | 2 | 9 |
| Alex R. | 2 | 3 | 3 | 1 | 9 |
This kind of structured scoring takes the "I don't know, I just liked their resume" out of the process. It also makes it easy to defend your decisions if someone asks why a candidate didn't make the cut.
You don't need fancy software. A Google Sheet works fine. But if you want this pre-built and connected to your job posts, Penroll handles the scoring setup as part of its hiring workflow.
Step 4: Check for red flags before inviting anyone
Once you have your top candidates ranked, spend five minutes doing a quick red flag check before sending interview invites. You're not running a background check — you're just catching obvious issues.
Things to look for:
- Unexplained 2+ year gaps (not always bad, but worth noting)
- Job-hopping pattern with no tenure over 12 months across multiple roles
- Mismatches between their resume and their application answers
- Generic cover letters that clearly weren't written for your role (if you required a cover letter)
None of these are automatic disqualifiers, but they're data points. If you're choosing between two equally scored candidates and one has three jobs in two years, that's relevant context.
Step 5: Narrow to your interview slate
For most roles at a small company, you want to interview 4–6 people at the first stage. More than that and you're burning time. Fewer than that and you don't have enough comparison.
Take your top-scored candidates who passed the red flag check and invite the top 5–6. If you only have 3 strong candidates, interview all 3 — don't pad the list just to hit a number.
If your top 10 candidates are tightly clustered in score, use your nice-to-haves as the tiebreaker. If they're still tied, look at the quality of their application answers — how they wrote, what they chose to emphasize, whether they seemed to actually read your job post.
What about work samples or pre-screening calls?
For some roles, it makes sense to add a lightweight step between shortlisting and a full interview. A 15-minute phone screen or a short take-home task can help you validate a candidate before committing to an hour-long interview.
This works well when:
- You have more qualified candidates than interview slots
- The role requires a specific skill that's hard to verify from a resume (writing, coding, design)
- You want to check basic communication or professionalism before investing more time
Keep any take-home task under 30–45 minutes of work. Anything longer is disrespectful of candidates' time and will cost you good people who are currently employed.
Step 6: Document your decisions
This one sounds boring but it matters. Keep a record of who made it through shortlisting, who didn't, and why.
You don't need a formal system. A column in your spreadsheet that says "Status" and "Reason" is enough. Something like:
- Jordan M. — Invited to interview — top scorer, strong sales background
- Sam T. — Hold — good fit but availability unclear, follow up if top candidates fall through
- Alex R. — Declined — missing required B2B experience
Why does this matter? A few reasons:
- If your top candidate falls through, you know exactly who to call next
- If you ever face a hiring discrimination claim, you have documentation that your decisions were criteria-based
- If you hire for the same role again in 12 months, you have a reference point for what worked
Common shortlisting mistakes to avoid
Reading resumes in order. The first 10 resumes you read set your mental benchmark. If they're all weak, you'll overrate mediocre candidates later. Shuffle the stack or do the fast pass first.
Screening for credential proxies instead of outcomes. "Must have a degree from a top 20 university" is a lazy filter. What you actually want is someone who can do the job. Look for evidence of that, not the packaging.
Letting one person make all the calls. If only the founder is shortlisting, you get only the founder's blind spots. Have a second person review your top 10, even casually.
Moving too slowly. Good candidates are interviewing at multiple companies. If it takes you three weeks to shortlist, you'll lose the best ones before you even schedule an interview. Aim to complete shortlisting within 5–7 business days of your application deadline.
Where Penroll fits
Penroll is built for exactly this kind of hiring — small teams making a handful of hires a year who need a process without the overhead of enterprise recruiting software. It helps you build structured criteria before applications come in, score candidates consistently, and move from application to interview slate without losing track of anyone. If you're tired of managing hiring in a jumble of email threads and spreadsheets, it's worth a look.