since the last time I optimized for talent ceiling on an ops hire
The pattern: rockstars leave in 8-14 months. The operation regresses. The B+ player who stayed 4 years compounded better.
For the first 6 years of building operations teams, I optimized for talent ceiling. I hired the most impressive resumes I could find. I paid above-market for “A-players.” I sold myself on the idea that a small team of stars would outperform a larger team of solid players.
In year 7, I started noticing the pattern across my hiring history. The rockstars left, on average, in 8-14 months. The B+ players who stayed 4+ years quietly compounded into something more valuable than the rockstars ever produced in their tenure.
By year 8, I stopped hiring rockstars for operations roles entirely. Here is what I learned in the transition and what I optimize for now.
What the rockstar pattern actually looked like
Five hires across different operations roles, all between years 4-7 of my career. All technically high performers. All gone by month 14.
Hire 1: senior support engineer, joined from a recognized SaaS company. Resume was elite. First 90 days: outperformed every metric I'd set. Month 4: started asking about expanded scope. Month 7: started doing skip-level networking inside our org. Month 11: left for a higher-title role at a larger company. We were back to baseline + knowledge gap within a month.
Hire 2: ops manager, ex-McKinsey + ex-startup ops. Brought in to redesign the operating model (irony noted). Months 1-3: produced a brilliant operating model that the team couldn't actually execute. Months 4-8: tried to drive execution; team revolted. Month 9: left for a strategy consulting role.
Hire 3: CX lead, ex-Zappos. Knew customer service operations as well as anyone I'd worked with. Months 1-6: dramatically improved CSAT. Month 7-12: bored. Month 13: gave notice for a head-of-CX role at a Series B.
Hire 4: community manager, prior experience running a 50K-member community. Excellent at the work. Months 1-4: community engagement metrics moved 30%. Month 8: started building her own community on the side. Month 11: left to do it full-time.
Hire 5: data analyst, top-of-class CS degree. Built our first real metrics dashboard. Took 6 weeks; was masterful. Months 4-10: ran the dashboard. Month 11: said the role wasn't analytically challenging enough. Month 12: left for a data science role at a tech company.
Pattern: every one of them was a great fit for what I hired them for and a terrible fit for the role they'd grown into by month 12. The mismatch wasn't their fault; it was inherent to the rockstar profile in an operations role.
Why rockstars don't fit operations roles long-term
Three structural reasons.
Reason 1: operations is about consistency, not innovation. The job is to make sure the same workflow runs the same way week after week. Rockstars get bored doing the same work twice. By month 4, they're trying to redesign their own role. By month 8, they've redesigned it. By month 12, they want a different role entirely (one they can redesign again).
Reason 2: rockstars have options. The same resume that made them attractive to me made them attractive to 50 other companies. They get pinged on LinkedIn weekly. They take meetings. They leave when the right offer comes.
Reason 3: knowledge walks with them. When a rockstar leaves, they take 6-12 months of context with them. The replacement (who is usually a non-rockstar because rockstars are rare) takes 3-6 months to rebuild that context. The operation runs at 60-70% capacity during the transition.
The net effect of hiring rockstars in operations: a 60% utilization rate over the average 14-month tenure, plus a 3-6 month gap to rebuild, plus the recruiting cost. Compare to a B+ player at 90% utilization for 4 years.
The math isn't even close.
What I optimize for now
After 8 years of the alternative, here's what I hire for in operations roles.
1. Consistency over ceiling. The candidate's last 3 roles averaged 2+ years each. Steady. Boring trajectory. No big leaps. This is the single most predictive signal of how long they'll stay.
2. Genuine interest in the work itself. Not "this is a stepping stone to something bigger." Actually interested in the operational work. Asks detailed questions about the workflow during the interview. Enjoys the craft of doing the same thing well repeatedly.
3. Limited employer competition for their skill set. Lives in a market where there aren't 50 companies competing for them. Geographic factor matters here (Tier-2 US cities, LatAm, Eastern Europe). Compensation expectations are reasonable because alternatives are limited.
4. Comfortable being a "best-fit B+" rather than an "ambitious A." This is the hardest signal to evaluate. Usually shows up in how they describe their own career trajectory. The honest version: "I want to be excellent at the thing I do, not be promoted away from it."
When all four signals are present, the hire usually stays 3+ years and compounds. When fewer than three are present, the hire usually leaves in under 18 months and the pattern repeats.
Operations is a craft, not a stepping stone. The people who treat it as a craft are the ones you want. The people who treat it as a stepping stone are the ones to avoid, regardless of how impressive their resume is.
What this changed about how I build operations companies
The shift from rockstar-hiring to consistency-hiring changed three things about how I structure operations companies.
Change 1: I stopped paying above-market for ops roles. Above-market compensation signals "we expect you to be ambitious and chase the next level." Market-rate compensation signals "we expect you to do this role well for a while." The signal matters more than the dollar amount.
Change 2: I stopped recruiting from elite-brand companies for ops roles. The elite-brand alumni are exactly the candidates who'll leave for the next elite-brand opportunity. I recruit from companies that are competent but not elite. The candidates stay.
Change 3: I built the Pod model around consistency hires. Every Pod specialist is hired against the 4-signal framework above. The replacement bench is also hired against the framework. The result: low variance in output across the team and high tenure across the workforce.
This is the structural reason PodFleet's replacement guarantee is feasible. Low turnover means the bench rarely fires; when it does, the replacement is hired in days, not months.
Where rockstars still belong
To be precise: the framework above is about operations roles. Rockstars are still the right hire for some roles.
Right for rockstars:
- Founder/CEO (obviously)
- Head of design (the output IS the brand)
- Lead engineer at an engineering-led company
- Sales leadership at a competitive go-to-market
- Any role where the output is brand-defining creative judgment
Wrong for rockstars:
- Customer support, community management, content operations, data ops, admin, AI configuration
- Most of what falls under “operations” in a 7-figure business
- Most of what a Pod covers
The distinction is: roles where the output is creative judgment can be A-player roles. Roles where the output is consistent execution should be B+-player roles. Hiring A-players for B+ roles produces churn and dissatisfaction on both sides.
The honest test for your own ops hires
Pull the resumes of your last 5 operations hires (over the last 3-4 years). For each one, note:
- Average tenure in their previous roles
- Whether they came from elite-brand companies
- Whether they've already left or are likely to leave within a year
If 3+ of the 5 fit the rockstar profile, you're optimizing for the wrong thing for ops. The replacement cost is invisible because you're so used to it. The math compounds against you over years.
The fix takes 6-12 months: change the hiring criteria for new ops roles. The current rockstars will leave on their own. The replacements (B+ consistency hires) will compound.
This is the most useful operations-hiring shift I made in 14 years. I wish I'd made it 6 years earlier.