Most early hiring falls apart not because a founder doesn't care about getting it right, but because there's no process at all. Job posts go out, resumes pile up in an inbox, interviews happen on gut feel, and three weeks later nobody remembers why a candidate got rejected. You don't need an HR department to fix this. You need five or six repeatable steps you can run yourself in a few hours a week.
This guide walks through building that process from scratch, assuming you have no recruiter, no applicant tracking system, and no budget for either.
Why "no HR team" doesn't mean "no process"
A process isn't paperwork. It's just a set of decisions you make once so you don't have to remake them for every candidate: what the role actually needs, where you'll look, how you'll filter, how you'll interview, and how fast you'll decide. Skip these decisions and every hire becomes a one-off improvisation, which is exactly how bad hires happen. If you want the fuller cost picture of getting this wrong, the real cost of a bad hire at an early-stage startup is worth reading before you post a single job.
The good news: a lean process is often better than a heavy one. You're not managing headcount across departments, you're filling one or two roles well. That's a solvable problem in an afternoon of setup.
Step 1: Define the role before you write the job post
Write down, in plain language, the three or four outcomes this person needs to deliver in their first six months. Not a list of skills, outcomes. "Ship our billing integration" is specific. "Strong backend skills" is not, and it's why you'll end up with 200 resumes that all look similar.
Also decide, before you post anything:
- Must-have vs nice-to-have skills. Keep the must-have list to three items max. Everything else is trainable or negotiable.
- Seniority band and comp range. Vague comp is the single biggest cause of wasted interview time.
- Remote, hybrid, or onsite, stated plainly, because this filters out mismatched applicants before you ever open a resume.
Step 2: Write a job post that filters, not just attracts
A job post's real job is to reduce your inbound, not maximize it. Include the outcomes from Step 1, the comp range, and one or two lines about your actual stage and constraints (small team, fast decisions, ambiguity expected). This self-selects out candidates who need a bigger, more structured environment, which saves you screening time later.
Step 3: Pick one or two places to post, not five
Posting everywhere feels productive but usually just means five inboxes of unqualified applicants instead of one. Pick the channel that matches the role: generalist boards work for high-volume, lower-specificity roles, while platforms built specifically for product and tech hiring tend to produce a sharper-fit shortlist for engineering, product, and design roles because the pool and the matching are narrower to begin with.
This is also where the "no HR team" constraint actually matters most. Legacy job boards were built assuming someone on the other end has time to sift a few hundred applicants manually. WellHired is built the other way: verified employer listings, AI-assisted matching that narrows the pool before you ever open a resume, and a simple enough product that you can run the whole posting-to-shortlist flow yourself without a recruiter or an ATS. You can post a role and see how the shortlist works without needing anyone else on your team to manage it.
If you want a broader menu of where to post before narrowing down, best places to post a startup job for free or nearly free covers the tradeoffs across channels.
Step 4: Build a simple screening filter
You don't need software to screen consistently, you need a rubric. Before resumes arrive, write three or four screening questions tied directly to your must-haves from Step 1, and decide what a pass or fail answer looks like for each. Then apply the same rubric to every resume in the order they arrive, not in whatever order feels most interesting to open first.
If resume volume is more than you can reasonably read in an evening, that's the point where automated or AI-assisted screening earns its keep, not to replace your judgment but to apply your rubric consistently at volume. For a deeper comparison of where manual review breaks down, see AI-assisted screening vs manual resume review for a time-strapped founder.
Step 5: Structure the interview, even if it's just you
Unstructured interviews are where most no-HR processes fall apart, because every conversation ends up testing something different. Fix this with a short, fixed interview plan:
- One question per outcome from Step 1. If the role needs to ship a billing integration, ask them to walk through how they'd approach it.
- One real work sample or scenario, not a puzzle. A short, paid take-home or a live problem close to actual job tasks tells you more than any behavioral question.
- A fixed scorecard. Rate each answer 1 to 5 against a written standard, before you discuss the candidate with anyone else, so first impressions don't quietly override the evidence.
If you're hiring for a role outside your own expertise, structure matters even more. How to interview engineers when you're not an engineer yourself and how non-technical founders can vet engineering candidates both walk through how to build a scorecard you can trust even without technical depth.
Step 6: Decide fast, and tell people either way
Slow decisions are the most common self-inflicted wound in founder-led hiring. Good candidates take other offers while you're "still thinking about it." Set a rule: decide within 48 hours of the final interview, and respond to every candidate, yes or no, within that window. This isn't just good manners, it protects your reputation as an employer in a small talent pool where candidates talk to each other.
Step 7: Keep one lightweight record instead of a full ATS
You don't need an applicant tracking system for two open roles. A single shared spreadsheet with columns for candidate name, stage, scorecard rating, and next step is enough to keep you honest and stop candidates from silently falling through the cracks. Update it after every interview, not at the end of the week from memory.
For a fuller list of no-cost tools that cover posting, screening, and scheduling without a recruiter, see 10 free or low-cost tools for founders hiring their first team and best tools to screen candidates without a dedicated recruiter.
Putting it together
None of these seven steps require a headcount, a subscription to enterprise HR software, or specialized training. They require writing a few things down before you start, applying them consistently, and moving faster than your instinct to "wait and see more resumes." That's the entire difference between a process and chaos with a job post attached.
FAQ
Do I really need a formal process for just one or two hires?
Yes, arguably more than a company hiring at volume. With only one or two hires, you can't average out a bad decision across a larger cohort, so the cost of an unstructured process falls entirely on those one or two roles.
What's the minimum viable version of this if I have almost no time?
At minimum: write the three must-have outcomes for the role, use a fixed set of interview questions tied to them, and commit to responding to every candidate within 48 hours. That alone fixes most of what goes wrong in ad hoc hiring.
Can I run this process without any paid tools at all?
Mostly yes. A spreadsheet, a shared document for scorecards, and a free or low-cost job posting channel cover the basics. Paid tools start to earn their cost once resume volume gets high enough that manual screening eats hours you don't have.
How is this different from just using a recruiting agency?
An agency effectively outsources these same steps to someone else, at a cost. Running the process yourself with a platform built for self-serve hiring is generally cheaper and keeps you closer to the candidate, though it takes more of your own time upfront. See WellHired vs a recruiting agency for a direct cost comparison.
At what point should I bring in an HR hire or recruiter?
There's no fixed headcount trigger, but a common signal is when you're running more than two or three simultaneous open roles, or hiring across multiple departments at once. Below that, a lean self-run process is usually faster and cheaper than adding a dedicated hire to manage hiring itself.