Why are we so divided?

Mainstream media profits from "us vs. them" narratives, turning every story into a battle to keep you watching. 1440's unbiased newsletter delivers what actually happened, so you can stay informed without being manipulated.

Then I notice the account that hasn't raised a ticket in six weeks.

Eight users, a file server, a line of business app, and a printer that's been dying since the day I met them. Six weeks of silence.

My first thought is that we've finally got on top of it.

My second thought, the one that turns out to be right, is that they've given up.

That's support fatigue. And I've caused it more than once, so this isn't me pointing at you from a stage. It's what it looked like from my side of the desk.

What support fatigue actually is

I want to be careful with the word, because fatigue is doing something specific here.

It is not the same as unhappy. An unhappy client tells you. They ring up, they say the thing is broken, they say they're not paying for it. Unhappy is loud, and loud is workable. You can have a conversation with loud.

Fatigue is quiet. Fatigue is a client who has stopped expecting the relationship to improve, so they've stopped putting anything into it.

The signs are small. They stop logging tickets and start saving problems up until something falls over. They nominate one long suffering person to "deal with the IT stuff", and that person becomes the only human in the building who talks to you. They sign off your monthly report with "all good thanks" on a report they haven't read. They reply to your proposed improvement with "leave it with me" and never bring it up again.

Then one of two things happens. Either they leave, or they stay, and staying is worse. A client who renews because switching is exhausting isn't a client. They're a hostage with a signed agreement, and hostages eventually get brave.

Your dashboard will not tell you

Here's the uncomfortable part. Support fatigue looks like success in a service desk tool.

If the client has stopped asking for things, your numbers get better. Tickets closed, average response time, first contact resolution, SLA percentage. Every one of those improves when the client quietly decides the cost of asking is higher than the cost of living with it.

A quiet account is not a healthy account. It's an account that has done the maths and decided against you.

CSAT has the same problem. You survey at the point of a closed ticket, when the person has just been helped, by a human they like, about a thing that is now fixed. That's the best possible moment, and it's the wrong one. The fatigue is in the week between the tickets, and it's sitting with the person who signs the invoice, who hasn't spoken to your helpdesk in a year.

You're surveying the wrong person at the wrong time, then wondering why the score is 4.6 out of 5 while the renewal goes to a competitor.

In partnership with

Sales Owns Retention, Not Just Closed Deals

Pebl’s Chief Business Officer, Dutta Satadip, joins The GTM Garage to show how Pebl merged Operations and GTM under one leader, and how it aligned goals and incentives across the customer lifecycle.

You’ll also see how Pebl uses AI across that lifecycle and the stack behind prospecting, pipeline, and handoff to onboarding.

15 min conversation. 15 min open Q&A. For Seed to Series A teams.

Where it actually comes from

Most of the fatigue I've seen comes from six things. If you run a managed IT consultancy, I'd bet you've got at least three of them in the building right now.

1. Every ask costs the client something.

A ticket isn't a free action for a client. They have to notice the problem, decide it's worth reporting, remember how they're supposed to report it, log into something they've used twice, and describe it a second time to a technician who wasn't listening.

Any one of those is nothing. Twenty of them in a month is a tax. The client starts to feel like the service they're paying for is a second job they didn't apply for.

2. You've made them their own project manager.

"Can you get the user to replicate it?" "Can you chase the ISP?" "Can you get approval from the director?"

Every one of those moves the work from your desk onto theirs. It's efficient for you, and the client often has the relationship or the authority. But they hired you to take the IT problem away, and instead they've been handed a project plan and a to do list. They won't complain. They'll just quietly conclude that doing it themselves would be faster.

3. The same problem keeps coming back.

The first time is an incident. The second time it's a pattern. The third time it's a decision you're avoiding.

"It's a known issue" means two different things depending on who hears it. Inside your business it means you've seen it, you know the workaround and you're tracking it. Outside your business it means this will keep happening and nobody is going to do anything about it.

4. Every technician is a stranger.

Repeating your environment to a new person is one of the most quietly infuriating things in professional life. So every handover is another round of "so how does the file share work", another explanation of why the finance team can't use the new thing.

Ticket notes are not context. A ticket note says what happened in the last 40 minutes. Context is knowing the director hates being copied on anything, that the line of business vendor only answers emails before 10am, and that the backup job is noisy but fine.

5. They can't see the work.

When IT goes well it's invisible. But invisible work is indistinguishable from no work, and nobody pays for what they can't see.

So the only evidence of your effort becomes the invoice and the list of things that broke. You've accidentally built a relationship made entirely of bad news and a monthly bill.

6. They only hear from you when something's wrong.

The phone rings when a thing has failed, the report arrives when it's time to justify the invoice, and the survey arrives when something went badly enough to warrant one. There's no moment in the month where the client gets to feel like the service is on their side.

That's not a support problem. That's a relationship designed, by accident, to only transmit complaints.

How to design it out

Right. Now the useful bit.

You can't fix this by being nicer on the phone. Nice is table stakes. You fix it by changing how often the client has to think about IT at all. Eight moves, roughly in order of how much difference they make.

1. Cut the number of asks, not the number of tickets.

These are different numbers and most practices only measure one. Ticket count tells you how much work you did. Asks per user per month tells you how often your client had to interrupt their day because of you.

You want the second one falling. Self service for the boring stuff. A knowledge base written in plain English rather than technician. Proactive monitoring so you find it before they do. Patching that actually runs. The top ten fixes at each client documented and handed over.

One warning: if you bill by the hour and then reduce the number of asks, you've cut your own revenue. Which is why this works with a retainer and badly without one. More on that shortly.

2. Make the ask cheap.

One door, not four. Whatever the door is, a phone number, an email, a chat box, a form, it should take about thirty seconds and shouldn't require remembering a password, choosing a category, or deciding how urgent something is. You decide urgency. You're the professional.

If you must have a form, make it four lines. Who's affected. What were you doing when it broke. What does it stop you doing. How do we reach you. Then a human takes over.

3. Own the third parties.

The vendor. The ISP. The telco. The line of business app support team who only respond to a ticket from a licensed contact.

You chase them. Not the client. "We'll chase the vendor and come back to you by Thursday" is worth more than a faster close time on the tickets you control, because the chasing is what actually ruins their week.

4. Close the loop properly.

Every closed ticket gets the same three lines.

What was wrong. What we changed so it doesn't happen again. What to do if it does.

That paragraph is the difference between "fixed" and "fobbed off". It also forces you to actually do the second line, which is where most of the fatigue is created.

5. Fix the pattern, not the ticket.

Once a month, pull anything that has come in more than once at the same client in the last 90 days. If it's happened three times, it goes on the problem list with a price and a date.

Not everything is worth fixing, and the client is allowed to say no. But the decision has to be visible and theirs. "We've seen this four times, here's why, here's what it costs to end it, your call" is a completely different relationship to "that's a known issue".

6. Report what didn't happen.

Prevention has no ticket attached, so you have to write it down or it didn't happen.

One page a month. Fourteen thousand blocked connection attempts. Sixty-two endpoints patched. The backup restore tested on the fourteenth and it worked. The phishing email three people reported because you trained them.

No jargon, no dashboard screenshot with fourteen pie charts. One page for the person who signs the cheque, in the language a business owner uses. And end it with one decision you want them to make, because a report that asks for nothing gets read like a newsletter, and one that asks for one small thing gets read like a message.

7. Give them one name.

An account owner, a named engineer, a photo, a direct number, and a handover note so the relationship survives a holiday.

The moment you have two people, it's the most useful thing you can do. A client who knows who to ask will ask. A client who has to work out who to ask waits until it's on fire.

8. Ask about friction before they tell you.

Once a quarter, twenty minutes, with the person who signs the invoice. Not a review. Not a sales call. Three questions.

Where are we slow?

What are you avoiding telling me?

Where do we make you do work you don't think you should have to do?

That third one is the whole article in a sentence. Ask it, write down the answer, pick one thing off the list, fix it before the next quarter.

The bit nobody wants to hear: this is often a pricing problem

Support fatigue is frequently not a delivery problem at all. It's a pricing design problem, and you built it.

If support is a flat fee and the client suspects every ticket is being silently counted against them, they'll ration it. That's rational. If support is billed per incident, they'll hide things from you, because every problem now has a price and nobody wants to be the person who cost the company four hundred dollars by asking about a printer. Also rational.

Both end in the same place. The small stuff doesn't get raised, the small stuff becomes big stuff, and the client leaves with a story about how you were expensive and hard to deal with, which will feel unfair to you and completely true to them.

What you want is emotional neutrality on the ask. No small charge, no small judgement, when they raise something. So you want a flat fee with an honest scope, a stated preference that they raise things early because early is cheaper, and a small allowance of "just do it" work so the tiny asks don't become quotes.

And when something is genuinely out of scope, say so early and plainly. "That's project work, it's about this much, do you want it this month or next" is a service. The same sentence after the work is done is a betrayal, and it costs you more than the job was worth.

If it's already happened

You've read all of this and you know exactly which client I'm describing. They've gone quiet. The report goes unread. You've started dreading the renewal.

One conversation, twenty minutes, in person or on camera, four parts.

Name it before they do. "I don't think you're getting what you should be out of this, and I'd rather hear it from you than guess." Then stop talking.

Take notes, not offence. Count the specifics, not the tone. If they say "you're slow", ask for the last three examples.

Own the mechanics, not just the apology. "That's on us. Here are two things I'm changing, starting this week, and here's what you'll see when it's working." Two changes, not nine. Nine means nothing happens.

Give them a decision and a date. "If this isn't better by the end of the quarter, tell me straight and we'll talk honestly about whether we're the right fit." Terrifying, and it works, because it turns a quiet grudge into a shared project with a deadline.

Then do the thing. That's the part nobody writes about and the only part that counts.

Some clients are past it. Fatigue that's sat for two years doesn't get fixed by a nicer report and a named engineer, and the kindest thing is to help them move on without drama. But most of it is fixable, and fixing it is cheaper than replacing it.

What I'd actually measure

Five numbers and one conversation. That's the whole dashboard.

Asks per user per month, trending down. If it's climbing, you're either growing or failing.

Percentage of tickets raised by monitoring rather than a human. You want this rising, because every one is an interruption that never happened.

Repeat contacts on the same asset within 30 days. Near zero, or it's on the problem list.

Client task rate. How many tickets needed the client to do something before they could close. Count it for a month and you'll be embarrassed. This is the number that produces fatigue and almost nobody tracks it.

Time to first human response. Not time to close. Fatigue is born in the silence before a human appears, and a client will forgive a two day fix much more easily than two days of not knowing whether anyone has seen it.

Then the conversation. "What's annoying you? What have you bought without telling us?" That second one is the best diagnostic in the business. The day a client buys a tool you didn't know about is the day they stopped believing you'd fix it.

Do these three things this week

Not this quarter. The quarter will fill up, this becomes a project, and projects about internal improvement never happen.

One. Pull 90 days of tickets and find the top three repeat issues across your client base. Pick one, price the fix, send it with no expectation attached.

Two. Count the tickets that included a task for the client. Find the one thing you ask them to do every single month and absorb it. Just do it yourself from now on and see whether anyone notices. They will.

Three. Ring your quietest client. Not to sell, not to book a review, not to mention the renewal. "How's it actually going? Is there anything you've stopped telling me about?" Then listen for ten minutes.

That's the whole thing. Move the ask, not just the ticket.

Find what’s slowing your GTM.

Most GTM problems aren’t one big failure. They’re one or two weak spots quietly dragging down the rest.

The free GTM Audit scores your go-to-market across four dimensions, then sequences a 90-day roadmap for your stage.

Drop your URL. Get your score in about 10 minutes.

The mindset bit

Support fatigue is invisible in exactly the places we're trained to look. We're engineers. We look at the queue, the clock, the SLA percentage, and those are the numbers that go up when the client gives up.

What you want is a client who tells you about small problems early, trusts you with the awkward ones, and doesn't think about IT for the rest of the month. That's a relationship, not a service level.

Nine times out of ten the fix isn't more effort on your side. It's less effort on theirs, and a bit more of yours spent on the boring design work that makes their effort unnecessary.

Anyway. That's enough out of me.

Thanks for reading, guys, I genuinely appreciate it. If you're building your own managed IT consultancy, or you've got one and it's running you rather than the other way around, tell me what your quietest client is really saying. The comments on these posts are usually better than the posts.

Go and ring the quiet one. Before the renewal does it for you.

Peace out.