There is a version of a support manager that leadership loves and that I have learned to find worrying. They handle escalations directly. They jump into the queue when volume spikes. Their name appears in resolved tickets. They are praised for staying close to customers and not losing touch with the front line. What is actually happening, in most cases, is that the team is quietly telling you something has broken.
I am not talking about a manager who reviews a couple of tickets a week to stay calibrated. I am talking about the manager who carries a regular share of the queue, who is effectively a senior agent with an org chart title attached. That pattern tends to look like dedication from the outside and feel like dedication from the inside. In practice, it means the team has a gap the manager is filling with their own hours instead of fixing.
What it creates that nobody talks about
The first side effect is a shadow escalation path. When users discover that contacting the manager directly gets faster results, they start doing it. This is rational behavior on their part and an entirely predictable outcome. The tickets that move through that path never enter the queue, never get measured, and never get fixed at the source. The manager's resolution rate looks good. The team's visibility into actual demand looks artificially low.
The second effect is subtler. When agents see their manager handling tickets alongside them, the implicit message is that the manager is a faster or more capable resolver. This is not necessarily the message the manager intends to send, but it is the one that lands. Agents stop bringing hard cases to the manager for coaching and start waiting for the manager to handle hard cases directly. The transfer of skill stops moving in the right direction.
The third effect is the most expensive: every hour a support manager spends resolving individual tickets is an hour not spent on the work that only a manager can do. Agents can close tickets. Only a manager can spot the pattern across two months of tickets that tells you a product team needs to change something. Only a manager can advocate for a process change that removes a whole category of tickets from the queue. Only a manager can sit in a 1:1 with a struggling agent and work through what is actually going wrong. These are not glamorous tasks. They are not visible in a resolved-ticket count. And they compound over time in ways that individual ticket resolution never does.
The honest steelman
There are real reasons a manager might handle tickets regularly, and not all of them are warning signs.
Very small teams are the clearest exception. If you manage two agents and the queue surges, you get in the queue. That is not a management failure; that is arithmetic. The team does not have enough bodies to cover the volume without the manager, and trying to maintain a strict manager-only-does-manager-things rule in a three-person operation is a good way to drop SLAs and burn out the people you are supposed to be developing.
Crisis handling is another legitimate exception. A large enterprise customer is about to churn and their account is on fire. The manager takes the ticket. That is good judgment, not a symptom of a broken system. The question is whether it is a one-off response to a genuinely unusual situation or a pattern that has normalized.
Some managers also need periodic hands-on work to stay honest about what the job actually asks of agents. A manager who has not handled a live customer issue in eighteen months is making decisions about agent workload and training with increasingly stale information. Occasional ticket handling to maintain calibration is different from carrying a regular share of the queue.
The word that matters in all of these is regular. If a manager is in the queue most days, something deserves examination.
What is actually going wrong
When I have seen this pattern, the root cause usually falls into one of three buckets.
The team is understaffed. The queue is genuinely too large for the agents available, and the manager is papering over a headcount gap. This is worth naming directly because filling the gap with manager time delays the evidence that surfaces the headcount argument. If the manager absorbs the overflow, the queue metrics stay reasonable, leadership sees no crisis, and the team stays understaffed.
The manager is not comfortable in the management layer. Some support managers were outstanding agents who were promoted because they were outstanding agents. The agent role had clear feedback loops and clear measures of success. The management role does not. Coaching is slower to show results than resolution. Pattern analysis does not feel as satisfying as a closed ticket. For managers who are still orienting to what the job actually asks of them, the ticket queue is a retreat to the familiar. This is not a character flaw. It is a transition that needs support, and naming it honestly is usually more useful than pretending it is not happening.
The manager is being measured the same way agents are. If a manager's performance review includes ticket volume, resolution time, or anything that creates direct incentive to handle individual tickets, the organization has built the wrong measurement system. A manager optimizing for personal ticket metrics is making a rational response to bad incentives. The fix is upstream, not in the manager's behavior.
What manager time should actually look like
The most valuable thing a support manager can do with their time is work at a level of abstraction above the individual ticket. This means reading tickets for patterns rather than resolving them. It means identifying which categories of tickets should not exist and doing the cross-team work to eliminate them. It means sitting with an agent after a hard ticket and walking through what made it hard and what the agent might do differently. It means reviewing the knowledge base for articles that are outdated or missing. It means being the person who translates what the support queue is saying into language that influences product and engineering decisions.
None of this shows up in a resolution count. All of it compounds. A manager who removes one category of ticket from the queue by fixing the upstream cause has done more for the team than a year of handling individual tickets in that category.
Every hour a support manager spends closing tickets is an hour not spent on the work that could reduce how many tickets exist. The team can close tickets. Only the manager can decide to build fewer of them.
If you manage a support team and you are regularly in the queue, the right question is not whether you should be there. It is what the fact that you are there is telling you about the team, the volume, the headcount, or the measurement system. The queue is a signal. What is it saying?
I hold this view fairly strongly, but I know it is not universal. If you manage a small team and think regular ticket handling is the right call, I would genuinely like to hear how you think about the tradeoff. Drop me a note.
Comments