Employee turnover is often the final stage of a problem leaders could have identified, diagnosed, and addressed much earlier.
Engineering firms would never wait for a slope failure, foundation problem, or quality breakdown before deciding whether warning signs deserved attention. We investigate conditions, evaluate risk, implement mitigation, and monitor performance. Yet many firms manage employee retention in exactly the opposite way. Concerns are acknowledged, reassuring conversations are held, and meaningful action is deferred until a resignation letter arrives.
By then, the retention problem is usually no longer developing. It has matured.
A resignation letter may feel sudden to leadership, but it’s often the final documentation of a condition that has been visible for months: compensation concerns, unclear advancement, sustained overload, diminished flexibility, lack of recognition, reduced access to meaningful work, or repeated feedback that produced no visible change. The letter is not the first indicator. It’s a lagging indicator. Gallup reports that departing employees believe 42% of voluntary turnover could have been prevented.
The warning signs usually come first
Employees rarely move directly from satisfied to resigned. More often, they test whether the organization is willing and able to respond. They raise a concern with a manager. They ask about compensation, advancement, staffing, workload, or the future of their role. They become less willing to volunteer for extra assignments. A highly engaged employee becomes unusually quiet. A strong performer begins asking questions about policies that once seemed unimportant.
No single behavior proves that someone intends to leave. The risk appears in patterns: the same issue raised more than once, an important promise that remains unresolved, or a visible decline in participation after a concern was supposedly addressed.
Many managers believe they have acted because they held the conversation. But listening is only the beginning. If the employee hears empathy while experiencing no change, the conversation can make the problem worse. It confirms that leadership understands the condition but has chosen not to mitigate it.
Compensation matters but it’s not the whole diagnosis
A holistic retention strategy should not minimize compensation. Pay is often the clearest signal of how an organization values responsibility, performance, licensure, specialized knowledge, and market demand. Gallup reports that pay and benefits were the most common single reason employees cited for leaving a job in 2024, but they accounted for only 16% of primary reasons. That finding matters in both directions: compensation cannot be ignored, and compensation alone does not explain most departures.
An employee can be paid competitively and still leave because the work has become less meaningful, the promotion path is opaque, the manager is ineffective, or the workload is unsustainable. Conversely, leaders should not use culture, purpose, flexibility, or appreciation as substitutes for market-aligned compensation. Goodwill does not correct a persistent pay gap.
The correct diagnosis may involve several conditions at once. A midcareer engineer may be carrying senior-level responsibility without the title or compensation. A technically strong employee may see management as the only path upward. A young professional may receive plenty of production work but too little context, mentoring, or client exposure to see a future at the firm. Treating all three as generic engagement problems would be the management equivalent of applying the same foundation solution to every subsurface profile.
The talent market leaves less room for preventable loss
Retention becomes more important when proven talent is difficult to replace. ACEC workforce research reported that approximately 184,000 engineers retired or otherwise exited the workforce in 2022 while approximately 166,000 entered, leaving an estimated annual gap of 18,000 engineers. The same research found a gap of more than 8,400 people in core civil, mechanical, and electrical disciplines.
The Bureau of Labor Statistics projects about 22,700 civil engineering openings each year from 2025 through 2035, many resulting from people transferring occupations or leaving the workforce, including retirement. At the same time, firms face greater cost and uncertainty in international recruiting. Current federal policy requires a $100,000 payment for certain H-1B petitions involving workers outside the United States and extends that requirement through September 21, 2027, subject to limited exceptions. That raises the cost of one pathway firms have used to reach specialized technical talent.
These pressures do not mean every engineering discipline or geography faces the same shortage. They do mean that firms should stop assuming a departing employee can be replaced quickly without operational consequences. Experienced engineers carry client trust, project history, technical judgment, mentoring capacity, and institutional knowledge. Losing one person can reduce far more capacity than a staffing chart shows.
A retention risk mitigation plan
Engineering firms already possess the mindset needed to manage this risk. The process can be organized into four familiar stages:
- Monitor the conditions. Use regular stay conversations, compensation benchmarking, workload reviews, development plans, and patterns in employee feedback to identify changes before they become exits. Annual engagement surveys are useful only if leaders also pay attention between surveys.
- Diagnose the cause. Determine whether the primary issue is compensation, role design, manager quality, workload, flexibility, recognition, advancement, or meaningful work. Ask what has changed, what remains unresolved, and what another employer could offer that would make the employee willing to listen.
- Implement mitigation. Translate the conversation into a specific response. Identify what will change, who owns the action, and when it will occur. If the organization cannot meet the request, explain that honestly. A clear "no" is more useful than an indefinite promise.
- Verify performance. Return to the issue after 30, 60, or 90 days. Confirm what was completed, what remains open, and whether the employee's experience actually improved. A mitigation plan is incomplete until its effectiveness has been evaluated.
Conversation without action is not mitigation
Every material employee concern should lead to one of three honest outcomes:
- The firm can address it and establishes an action and deadline.
- The firm cannot address it yet and establishes a review date.
- The firm cannot address it and explains why.
The most damaging response is reassurance without a decision.
This doesn’t mean that every complaint must produce the employee's preferred outcome. Organizations have financial, operational, and equity constraints. It does mean that leaders should close the loop. Employees can often accept an answer they don’t like more readily than they can accept being repeatedly heard and quietly ignored.
Recognition should also be treated as operating feedback rather than an occasional morale exercise. Gallup found that employees receiving high-quality recognition were 45% less likely to have left their organizations two years later. Recognition is most credible when it is timely, specific, and connected to work the firm claims to value: technical quality, mentoring, client service, sound judgment, collaboration, and risk reduction.
Retention belongs to operations
Human resources can provide market data, systems, policies, and coaching, but retention cannot be delegated to HR. Project managers determine workload. Practice leaders allocate meaningful assignments. Technical leaders create development opportunities. Executives control compensation, promotion structures, and organizational priorities. Direct managers determine whether employee concerns become actions or disappear into conversation.
That is why retention should be reviewed alongside backlog, utilization, succession, project risk, and client concentration. For critical roles, firms should know where they have single points of failure, which employees hold difficult-to-transfer knowledge, and whether identified successors are receiving the experience required to step forward.
Counteroffers may occasionally retain someone, but they are not a strategy. When money, flexibility, staffing support, or advancement becomes available only after an employee resigns, the organization sends an unintended message: the concern was affordable to solve, but it was not urgent until departure created a business problem.
Employees rarely resign because of one difficult day. They resign after a pattern of conditions, conversations, and commitments that produced too little change. Engineers would not call a project failure unforeseeable simply because warning signs were documented but never acted upon. We should bring the same discipline to retaining people that we bring to protecting projects: monitor conditions, diagnose risk, implement mitigation, and verify performance before the failure occurs.
![]() |
Donald J. Smith, P.E., MBA, is Group Leader of Power and Energy at ECS. Connect with him on LinkedIn. |
