It’s April. You hired 12 people in January and February. Three are already gone. Two more look like they’re checking out. And you’re staring at your calendar wondering if you’re going to be in full re-hire mode before summer even starts.
This is Q2 for most small businesses. The Q1 hiring surge ends, the adrenaline fades, and you find out which hires actually stuck.
Employee retention metrics are the early-warning system that tells you trouble is coming before someone hands you a two-weeks notice. Not after. Before. Tracking the right employee retention metrics means you act on Monday morning instead of rebuilding your team in June.
Refered is the referral-first hiring system built for small and mid-sized businesses. It combines applicant tracking, referral automation, and retention analytics in one platform, so HR managers can see why people stay or leave from a single dashboard.
This post covers the five employee retention metrics that predict turnover before it compounds, including the formulas, the benchmarks for SMBs under 200 employees, and the weekly action trigger for each one. At the end, I’ll show you how to build a 30-minute weekly retention dashboard from what you already have.
Why Most SMBs Don’t Catch Retention Problems Until It’s Too Late
The “Hire and Hope” Trap
Most small businesses track hiring. They don’t track what happens after the hire.
You’ll see cost-per-hire in a spreadsheet somewhere. You’ll know your time-to-fill. But ask a typical SMB owner which employee retention metrics they check weekly, and the answer is usually a blank look.
That’s the “hire and hope” trap. You fill the role, cross your fingers, and find out 60 days later whether it worked. By then, the damage is already done: the person is disengaged, or they’ve already accepted an offer somewhere else, or they’ve started quietly training their replacement by doing half the job.
The businesses that stay out of this cycle don’t just hire better. They watch different numbers.
What a Weekly Retention Check Actually Looks Like
It’s not a full HR analytics session. It shouldn’t take more than 20 minutes.
Pull five employee retention metrics. Compare them to the week prior and to your benchmarks. Flag anything that’s moved into warning territory. That’s it.
The goal isn’t to generate a report. It’s to surface the problem early enough to have a conversation before someone decides to leave.
The 5 Employee Retention Metrics That Predict Turnover Before It Happens
| Metric | Formula | Healthy Benchmark (SMB) | Weekly Action Trigger |
| 90-Day Retention Rate | (Employees who stayed 90+ days / Total hired in period) x 100 | 80%+ | Below 75%, review onboarding for that cohort immediately |
| Voluntary Turnover Rate (Rolling 30-Day) | (Voluntary exits in last 30 days / Average headcount) x 100 | Under 1.5% monthly | Above 2%, identify which department or manager the exits cluster around |
| Retention by Source of Hire | (Employees still employed after 12 months / Total hired from source) x 100, segmented by channel | Referral hires: 85%+ / Job board hires: 65-70% | If job board retention drops below 60%, shift budget toward referral activation |
| Time-to-Productivity vs. Time-to-First-Resignation | Avg days to full productivity / Avg days to first voluntary exit | Gap of 60+ days between the two | If gap narrows to under 30 days, your onboarding or manager experience is the leak |
| Cost of Turnover (Rolling Quarter) | (Separation costs + Vacancy costs + Replacement costs + Training costs) per exit | 50-75% of departing employee’s annual salary | If quarterly turnover cost exceeds 15% of payroll, calculate ROI on referral program activation |
Metric #1: 90-Day Retention Rate
The first 90 days are the most dangerous period for any new hire. If someone’s going to leave early, they usually decide within the first three months. You just don’t always find out until month four. That’s why 90-day retention rate belongs at the top of any list of employee retention metrics worth tracking.
How to Calculate 90-Day Retention Rate
90-Day Retention Rate = (Employees who completed 90+ days / Total hired in the same period) x 100
Run this monthly for each hiring cohort. If you hired 10 people in January, check in April to see how many are still with you. That’s your January cohort’s 90-day retention rate.
Healthy, Warning, and Critical Benchmarks for SMBs
- Healthy: 80% or higher
- Warning: 70-79% — One or more root causes need investigation (onboarding, role clarity, manager match)
- Critical: Below 70% — Systematic problem. Run stay interviews with current employees in the same role immediately
Methodology note: These benchmarks are drawn from SHRM’s employee retention research and annual exit interview data on why employees leave, calibrated for businesses under 200 employees where individual exits carry more weight than at enterprise scale.
The Weekly Action Trigger
When I see this employee retention metric drop below 75% for a cohort, my Monday play is pulling the onboarding checklist for those hires. Were 30-day check-ins completed? Did they meet their direct manager weekly in the first month? Was the role they’re doing now the role that was described in the interview?
Three questions. Often, one of them explains the exit.
Metric #2: Voluntary Turnover Rate (Rolling 30-Day)
Involuntary turnover (terminations, layoffs) is noise. Voluntary turnover is the signal. When someone chooses to leave, they’re telling you something. This employee retention metric tracks it separately and fast.
How to Calculate Rolling Voluntary Turnover
Rolling 30-Day Voluntary Turnover = (Voluntary exits in last 30 days / Average headcount for the period) x 100
Rolling means you recalculate it weekly, always looking back 30 days. You want to catch a spike before it becomes a trend.
Healthy, Warning, and Critical Benchmarks for SMBs
- Healthy: Under 1.5% monthly (roughly 18% annualized, which aligns with national quit rate data across industries for service-sector employers)
- Warning: 1.5-2.5% monthly
- Critical: Above 2.5% — Annualizes to 30%+ turnover, which is a retention emergency for any team under 50 people
The Weekly Action Trigger
When this employee retention metric ticks above 2% in any rolling 30-day window, I’m looking at two things: which department is generating the exits, and whether any of them share a manager. Turnover isn’t random. It clusters. Find the cluster and you’ve found the problem.
Metric #3: Retention by Source of Hire
This is the most underused employee retention metric in most SMBs. And it’s the one that could actually change your entire hiring budget allocation.
Why Source of Hire Is the Most Underused Retention Metric
Most hiring managers track where candidates came from. Almost none of them track whether those candidates stayed.
That’s backwards. If your Indeed hires are leaving at 12 months and your referral hires are staying for three years, you have all the data you need to make a decision. You’re just not looking at it.
How to Segment Retention by Channel (Referral, Job Board, Agency, Direct)
12-Month Retention by Source = (Employees from Source X still employed at 12 months / Total hired from Source X in the period) x 100
Run this for each source: employee referrals, job boards (Indeed, ZipRecruiter), staffing agencies, direct applicants, and LinkedIn. Compare them side by side. The differences will surprise you.
What the Data Shows About Referred Hires
The research consistently backs this up. Across Refered’s SMB customer base, referred hires show approximately 30% higher 12-month retention than job-board hires. (Placeholder: Refered to confirm exact figure before publish.)
This tracks with broader industry data. Research confirming referred employees stay longer shows they are more likely to pass the one-year mark because they came in with realistic expectations, a social connection to the team, and existing cultural alignment. A referral isn’t just a candidate source. It’s a pre-screened cultural match. For more on why referred hires stay longer, see Refered’s analysis of referral retention impact.
If you’re currently spending 80% of your recruiting budget on job boards and 20% on referrals, that ratio might be costing you more in turnover than you’re saving in job posting fees.
Metric #4: Time-to-Productivity vs. Time-to-First-Resignation
This employee retention metric is a ratio. A gap. And the size of that gap tells you whether your onboarding is working.
How to Measure Both Numbers
Time-to-Productivity: The average number of days between a new hire’s start date and the date they’re fully independent in their role. You’ll need to define what “fully productive” means for each role, but even a rough definition works.
Time-to-First-Resignation: The average number of days between a new hire’s start date and the date of their first voluntary exit, calculated across all voluntary exits.
Gap = Avg Time-to-First-Resignation – Avg Time-to-Productivity
You want this gap to be wide. If people are leaving before they ever get productive, you have a problem that no job posting is going to fix.
Healthy, Warning, and Critical Gap Benchmarks
- Healthy: Gap of 60+ days (people are getting productive well before they’re deciding to leave)
- Warning: Gap of 30-60 days
- Critical: Gap under 30 days, or negative (people are leaving before reaching full productivity)
The Weekly Action Trigger
A narrowing gap almost always points to one of two things: the onboarding process isn’t setting people up for success fast enough, or the manager experience in the first 30 days is driving early exits. When I see this gap shrink week over week, my first call is to the managers whose teams are seeing the exits, not to the recruiting team.
Metric #5: Cost of Turnover (Rolling Quarter)
Most articles on employee retention metrics give you a number and stop. They tell you what each employee departure actually costs is around 50% of a salary and call it a day. That’s not a formula. That’s a talking point.
The Real Formula (Most Articles Get This Wrong)
Real turnover cost has four components:
Total Turnover Cost = Separation Costs + Vacancy Costs + Replacement Costs + Training Costs
- Separation costs: Exit interviews, HR admin time, severance if applicable
- Vacancy costs: Lost productivity during the open role (estimate daily output value x days vacant)
- Replacement costs: Job posting fees, recruiter time, interviewing hours across the team
- Training costs: Manager time, onboarding materials, reduced new hire output during ramp
Add those up per exit. Track the rolling quarter total. Compare it to what you’re spending on retention programs. For most SMBs, the math will make referral program investment look very cheap.
Benchmarks for SMBs Under 200 Employees
- Healthy: Cost of turnover per exit at 50-75% of departing employee’s annual salary
- Warning: 75-100%
- Critical: Above 100% of annual salary, which happens more often than people expect in skilled trades, healthcare, and hospitality roles
The Weekly Action Trigger
If your rolling quarterly turnover cost is climbing above 15% of total payroll, it’s time to run the math on referral program activation. Most SMBs find that even a modest referral bonus program, properly automated and tracked, pays back within one prevented exit.
How to Build Your Weekly Retention Dashboard in 30 Minutes
You don’t need a dedicated HR analytics for small business platform to run these employee retention metrics. You need three things: your ATS, your payroll system, and a spreadsheet (or a purpose-built dashboard).
What to Pull From Your ATS
From your applicant tracking system built for SMBs, pull: hire date by source, cohort start dates, and exit dates with exit type (voluntary vs. involuntary). If your ATS doesn’t track source of hire at the candidate level, start doing it now. This is the data that makes Metric #3 possible.
What to Pull From Payroll
From payroll: headcount by period (for rolling turnover calculations), separation dates, and compensation data (for cost-of-turnover calculations). Most payroll platforms export this in a few clicks. You don’t need an API.
How Refered Pulls It All Into One View
The manual spreadsheet approach works, but it breaks down fast when you’re managing 50 or more employees and tracking multiple hiring cohorts at once.
Refered’s retention analytics dashboard connects your referral source data, your ATS records, and your retention outcomes in one place. Instead of pulling three exports and doing cross-tab math on a Sunday night, you see all your employee retention metrics in a single view,retention rate by source, 90-day cohort performance, and voluntary turnover trend, updated in real time.
If your current setup has you spending more than 30 minutes a week on these numbers, that’s the case for looking at a platform that combines tracking and reporting. See Refered analytics to understand what that dashboard looks like in practice.
Frequently Asked Questions
How often should HR managers check retention metrics?
Weekly is the right cadence for the most volatile employee retention metrics, like rolling voluntary turnover and 90-day cohort retention. A weekly check takes 15-20 minutes and catches problems early enough to act on them. Monthly reviews are fine for slower-moving numbers like cost of turnover and source-of-hire retention.
What’s a healthy employee turnover rate for small businesses?
For SMBs under 200 employees, target a voluntary turnover rate below 18% annually, which works out to roughly 1.5% monthly. Industries like hospitality, retail, and healthcare run higher by default. If you’re above 25% annually and not in a high-churn industry, something in your culture, compensation, or management is driving it.
Which retention metric is the most important to track first?
Start with 90-day retention rate. It’s the fastest employee retention metric to calculate, it covers the highest-risk period for new hires, and a problem here almost always points to a fixable root cause: onboarding process, manager relationship, or role misalignment. Fix the 90-day number and everything else tends to improve.
Can I track retention metrics without an HR analytics tool?
Yes. A well-structured spreadsheet with hire dates, exit dates, exit types, and source of hire is enough to run all five employee retention metrics in this post. The limitation is time: pulling and calculating manually works at 20-30 employees but becomes unsustainable at 50 and above. At that point, an integrated platform starts paying for itself in hours saved.
How do referral hires affect retention metrics?
Referred employees consistently outperform job-board hires on 12-month retention across nearly every industry. They arrive with realistic expectations, a social connection to the team, and a built-in accountability loop. For SMBs running an employee referral program, tracking retention by source of hire is how you measure and prove the ROI of that program over time.
You can’t fix what you’re not watching. And you can’t watch everything, so you need to watch the right things.
These five employee retention metrics (90-day retention rate, rolling voluntary turnover, retention by source of hire, time-to-productivity vs. time-to-first-resignation, and rolling cost of turnover) give you a complete picture of where your retention is breaking down, which hiring channels are actually working, and what each departure is really costing you.
The businesses that slow Q2 attrition aren’t doing anything magical. They’re checking their employee retention metrics every Monday and acting on the ones that are moving in the wrong direction.
Refered is the referral-first hiring system built for small and mid-sized businesses. It combines applicant tracking, referral automation, and retention analytics in one platform, so HR managers can see why people stay or leave from a single dashboard.
If you want to see what a weekly retention dashboard looks like when the data is already connected for you, Contact Refered today, see their analytics and take a look at how it works in your stack.
