Handling client escalations with limited resources means deciding in advance what you’ll do for whom, then being honest with everyone else. You can’t give every angry client your best people when you only have a few of them left. What you can do is give every client a clear answer, a named owner and a date.
When I cut the support team at AudienceView by about 60%, as I described in reducing a support team by 60%, the volume of routine work fell with it. The escalations didn’t fall at the same rate. Clients whose businesses had stopped overnight were frightened, short of money and, understandably, not in a patient mood. A smaller team was dealing with a higher share of difficult conversations.
This is the practical side of the same period: how I’d run an escalation process when the people available to run it have shrunk, and the clients on the other end are under more pressure than ever.
Why do escalations rise when the team shrinks?
Three things happen together. The clients are under stress, so small problems feel bigger. The team is stretched, so response times slow and small problems become real ones. And the people who used to catch trouble early, through relationships built over years, are no longer there. An escalation is often just a client discovering that nobody has been watching.
I’d also be honest that some escalations were fair. When you cut a team by more than half, service gets worse in places. Pretending it hasn’t is the quickest way to turn a legitimate complaint into a lasting grievance. Sometimes the right opening to an escalation call is to agree: yes, we’re slower than we were, and here’s what we’re doing about it.
How do you decide which escalations get your best people?
With a short, written set of criteria, agreed before you’re in the middle of it. In a normal quarter you can staff every escalation. In a reduced one you can’t, and deciding case by case, under pressure, leads to the loudest client winning. Rank by what’s at stake: is the client’s live service affected, is there a contractual or financial commitment, is there a risk to the relationship that matters beyond this one issue?
The severity and customer-tier thinking in my tiered SLA matrix applies directly. A severity-one issue for a client with a live event gets a senior person immediately. A general frustration gets a clear update and a time. Write it down, show it to your team, and share the principle with account managers so they can explain it to clients. It’s far easier to defend a rule than a judgement call.
Keep the structure of the escalation matrix I’ve described before: named roles, timed steps, a ladder that moves up when a deadline passes. With a reduced team, each rung may be a person who is doing three jobs. That makes the timing more important, not less, because it prevents an issue from sitting with someone who has no capacity to look at it.
What do you say to an angry client when you can’t do everything?
Start by listening, and let them finish. People in a crisis often need to say the whole thing before they can hear anything. Then restate the problem in your words, so they know you understood. Then tell them three things: what you’re going to do, who’s going to do it, and when they’ll hear from you next. That’s the shape of nearly every escalation call that went well.
Don’t promise what you can’t deliver. It’s tempting, because a promise ends an uncomfortable call. It also guarantees a worse one in a week. I’d rather tell a client “I can have an engineer look at this by Thursday, and I’ll update you Wednesday regardless” than promise a fix tomorrow that I know I can’t staff. Clients accept limits. They don’t accept being misled about them.
Be clear about what you can’t do. If a project or a custom report is on pause, say so, explain why in plain terms and say when you’ll revisit it. A “no, not now” with a date is better than a “yes” that never arrives. And acknowledge the client’s situation, because many of them were in genuine difficulty. Showing that you understand what they’re going through costs nothing and changes the tone of the whole relationship.
How do you protect the team handling the hard calls?
Take the heaviest calls yourself when you can, and share the rest. In a reduced team a few people end up carrying the escalations, because they’re good at them. That’s a fast route to burnout. Rotate the load, pair people on difficult calls, and make sure whoever has just had a bad one gets a few minutes before the next. I covered the people side in keeping a team motivated when everyone is scared, and this is where it’s tested.
Back your people in public. If a client demands to speak to someone more senior as a tactic, be willing to join the call without implying the agent did anything wrong. And when a client is abusive, say so and set a limit. People who are frightened for their business sometimes take it out on whoever picks up the phone, and your team doesn’t have to absorb that.
Keep a written log of every escalation: who, what, what was promised, what happened. It protects your team and gives you a record to learn from. It also stops the same story being retold three different ways in three different meetings.
Who owns the client relationship when the team is thin?
One named person for each client in escalation, and everyone on the team knows who. When resources are tight, the worst outcome is a client who gets three different answers from three different people because nobody was clearly in charge. Decide who owns the conversation, who owns the technical fix and who owns the update schedule, and write all three on the ticket. They can be the same person on a small account.
Hold a short daily review of open escalations, ten minutes at most. Go through each one: what’s the status, what’s blocking it, who needs help, and when does the client hear next? It sounds like overhead, and in a small team it’s the opposite. It stops things sitting in someone’s queue unnoticed, and it lets you move a person onto the issue that’s about to go wrong before it does.
Know when to ask the client for help. Many were more understanding than I expected when I told them plainly what was going on and what I needed from them. A good description of the problem, a named contact on their side and realistic expectations about timing all shorten the work. Treat the client as a partner in the fix, not an audience for it.
How do you stop escalations from happening in the first place?
By communicating before the client has to ask. Much of what looks like an escalation is a client who doesn’t know what’s happening. Tell your clients early about changes in response times, paused work and who to contact. I wrote about the principle in service levels during a spike, and it’s the cheapest fix there is: a message that answers the question before it becomes a complaint.
Look at the pattern behind the escalations. If three clients are escalating about the same problem, you have a problem to fix, not three incidents to handle. That’s the difference between incident and problem management in how a help desk works, and with a small team you can’t afford to keep answering the same complaint. Spend a few hours on the cause. It will pay back within a month.
Finally, build back trust deliberately. After a rough patch, go back to the clients who escalated and tell them what changed because of their feedback. Most never hear another word after the fix, and the ones who do remember it. The HDI community has good material on service recovery if you want an outside view.
Remember that escalations are information. Each one tells you something about where your service is weakest, which client segments are most exposed and which promises you’re making that you can’t keep. After the worst has passed, go through the log and look for themes. Did most of the escalations start with a slow first response? A missing handover? A feature that wasn’t working as the sales material said? The answers tell you where to put scarce resources first.
And look after your own judgement. When you’re stretched, it’s tempting to give in to whoever is pushing hardest, or to put off the hard calls until they become emergencies. Neither helps. Keep to your criteria, take the difficult conversation early, and ask someone you trust to sanity-check you when you’re tired. A decision made at the end of a long day is rarely your best one.
A smaller team can’t be everywhere. It can be honest everywhere, and that turns out to matter more.
Hutch Morzaria is a Director-level CX and Support Leadership professional with 19 years of experience building global support organizations across SaaS, Fintech, and enterprise technology. He has hired dozens of support and CX leaders across his career and holds ITIL Expert certification across V3 and V4.
