Structure and the min-mid-max framework
Compensation bands define the range of salary for each job level in each market. A senior engineer in San Francisco has a different band than a senior engineer in Austin. A product manager has a different band than an engineer. Each band typically has a min (entry level), mid (competent performer), and max (expert performer at that level). The band might span 50-100k for a junior role and 150-250k for a senior role.
The purpose is consistency and fairness. Two engineers at the same level in the same market earn within the same band. It prevents arbitrary pay, salary compression (where new hires earn more than veterans), and favoritism. It also signals to employees the path to higher compensation: move to the next level, and you move to a higher band. The band is not a cage; it is a framework that individuals can navigate through skill and tenure.
Market dynamics and the merit curve
Compensation bands must track market rates or the company bleeds talent. When tech salaries jump 20 percent across the industry, out-of-sync bands cause resignations. HR reviews market data annually (usually via salary surveys) and adjusts band minimums upward. But individual raises within the band are discretionary, based on performance and contribution. A high performer might reach the top of their band; an average performer stays in the middle.
Bands also encourage retention and reduce entitlement. A person at the max of their band faces an implicit message: your next raise is a promotion to the next level. This motivates leveling up. If there is no leveling opportunity, the person often leaves. Transparent bands create clarity: everyone knows the ceiling at their level and the pathway above it. Mystery compensation creates resentment; transparent bands create alignment.