Opens in a new tab
News, Trends, and Insights for IT & Managed Services Providers
News, Trends, and Insights for IT & Managed Services Providers
63c79131 d835 4fb9 b651 2bedaa5f364a

Cut Staff or Stop Hiring?
Before we start, the story everyone in tech is talking about. Nvidia’s board accepted Jensen Huang’s resignation as president, chief executive and board member, effective immediately, after 33 years running the company. Chief financial officer Colette Kress is interim CEO while the board searches for a permanent replacement. The board says it expects operations, customer commitments and the product roadmap to continue without interruption. The announcement gives no reason. It followed a week of reporting on Huang’s personal relationships and a board investigation. I’m tracking this for what it means for MSPs and their clients, through pricing, supply and roadmaps. It’s too early to say. When there’s something concrete, you’ll hear it here.

Companies are taking the savings from AI in one of two ways. They let people go, or they don’t hire them in the first place.

Start with Microsoft, the biggest example of the first. In April, it ran its first-ever voluntary retirement program, and about 30 percent of roughly 8,750 eligible employees took it. In July, it cut 4,800 jobs. Then it offered voluntary buyouts to about 60,000 employees, roughly a quarter of its workforce. Those who accept leave December 31.

Then, on November 12, Exchange Online went down. Outlook, Teams, SharePoint, Copilot and Defender XDR were all affected, and the disruption ran into a second day. Microsoft said servers were failing automated health checks and dropping out of service, leaving the rest to carry the load. It said there was no sign of an attack, and as of Friday it hadn’t named a root cause.

To be clear, nobody has connected the two. No one who takes the buyout has left yet, and Microsoft says some data center staff aren’t eligible. But they sit side by side: a company shrinking its workforce, and a failure across two days of the email service most of your clients run on.

The second way is quieter. Gartner estimates that executives put off creating or hiring for about 1.3 million positions in 2026. By 2030, it expects postponed hiring to affect five million people, while only 0.4 to 1.8 percent of workers lose their jobs directly to AI. Gartner’s John Lovelock put it this way: “Deferred jobs are not the same as lost jobs. They simply aren’t created.”

Hear what that adds up to. Most of what AI saves on labor doesn’t show up as layoffs. It shows up as hires that never happen. Some companies are cutting anyway.

In plain terms, there are two ways to take what AI saves. The question is which one holds up.

So why does one path hold up, and the other doesn’t?

The Knowledge AI Can’t Absorb
Underneath both is one dynamic. AI is good at absorbing work. It isn’t good at absorbing what the people doing that work know: a client’s history, why a system is set up the way it is, who to call when something breaks. Take the savings by not hiring, and that knowledge stays in the building. Take them by cutting, and it walks out the door.

Start with the MSPs doing this best. ConnectWise’s Service Leadership unit, which benchmarks MSP financials, reports that its top-quartile MSPs reached a service multiple of wages of 3.32 in the third quarter. That’s service revenue against the fully loaded cost of delivering it, and the top quartile had averaged about 2.7 for years. In plain terms, every dollar spent delivering service now brings back about three dollars and thirty cents in service revenue, up from about two seventy.

 The middle half barely moved, and the bottom quartile didn’t move at all. The top performers spend 4.6 percent more than others on tools and 6.8 percent less on labor, and Service Leadership says they hire new staff at a significantly slower pace. That’s ConnectWise’s own data, and its explanation, early adoption of service-desk automation, is a theory, not a proof. But notice what the numbers describe: slower hiring, not layoffs.

Now look at the other path. Staffing firm Robert Half surveyed U.S. hiring managers whose organizations had eliminated positions because of AI. Thirty-two percent later had to reinstate the same or similar roles. The top reason, at 40 percent, was that the job needed institutional knowledge or context AI couldn’t replace. Thirty-nine percent cited relationship management.

The big names tell the same story. Klarna leaned hard on AI chatbots for customer service, and its CEO has acknowledged it went too far. It started hiring people again for what a spokesperson called complex, sensitive cases. Ford hired and promoted 350 experienced engineers after its AI quality-control tools fell short. Ford’s explanation: “These engineers carry the hard-earned wisdom of decades of design.”

Now, a fair objection: maybe those companies just deployed AI badly, and better tools would have held. Maybe. But look at the reasons at the top of the list: knowledge, context, relationships. Those don’t improve with the next model release, because they live in people, not in the systems the tools read.

In plain terms, AI replaces tasks, not what someone knows about your clients. The top quartile kept that knowledge and stopped adding headcount. The companies that cut had to buy it back.

And buying it back isn’t cheap.

The Premium to Buy It Back
For the MSP, the consequence is that what your technicians know about your clients is the expensive part to replace, and the market is pricing it that way.

Start with what returning workers ask for. In a survey by talent assessment firm SHL, 19.1 percent of U.S. respondents, nearly one in five, said they’d been replaced by AI and then asked to come back. Of those, 51 percent wanted a substantial raise before returning. Forty-two percent wanted written guarantees about future job security, and 40 percent wanted to know how AI would be used in their role. That’s SHL’s survey, and the reporting doesn’t give its sample size, so take the numbers as a direction, not a precise measure. But the direction is plain: the person who comes back knows what they’re worth.

Then there’s the market they come back into. CompTIA says tech unemployment fell to 2.8 percent in October, down from 3.3 percent in September, against 4.1 percent nationally. Let a senior technician go and need that knowledge again, and you’re shopping for it in a market with very little slack.

Put those together. Cutting doesn’t remove the cost of knowledge. It pushes it to later, adds a premium, and collects it in a tight market. You end up buying back your own client history, at a markup.

Slower hiring has its own trap. If automation takes the entry-level work and you stop bringing in juniors, nobody is learning your clients behind the people who know them now. The knowledge stays in the building only as long as those people do.

So here’s the choice. Take AI’s savings through your hiring plan: make automation try each piece of work before you add a seat, keep the technicians who carry your clients’ history, and give each of them someone newer to teach. Or take the savings through cuts, and when that knowledge turns out to matter, pay a premium in a tight market to buy it back.

And that choice gets made one hire, and one departure, at a time.

For most MSPs, the next one is already on someone’s desk.

Why Do We Care?
Because the hiring decision usually gets made one ticket queue at a time, by whoever’s most underwater. Before the next job posting goes out, list the work that seat would do and let automation take a real run at it for a month; whatever it can’t handle defines the role. But don’t hand automation everything it can handle. Keep a deliberate share of that work with junior technicians, paired with the senior people who know your clients, because that’s how the knowledge finds a second home before anyone leaves.

What to Consider

  • Run every new seat past automation first. Write the job description as a list of the work, then pilot automation against that list for a month before you post it. Hire for what’s left, so the role you fill is the one automation can’t do.
  • Reserve the training work on purpose. Pick the ticket categories juniors learn on, and keep them manual or in approval mode even where automation could close them. That’s the cost of building the next person who knows your clients, so budget it as training, not as inefficiency.
  • Find the knowledge that lives in one head. For each client, name the person on your team who’d be hardest to replace and what only they know. Then pair them with someone newer on that client’s work, starting with your largest accounts.

If this trend continues: By the end of 2027, the MSPs that took AI’s savings by not hiring, while keeping a training path open, will be the ones posting top-quartile service multiples. The ones that cut senior technicians will be paying above-market rates to rehire for the client knowledge they let go.

This content is made possible by community-minded viewers like you

Join the Small Biz Thoughts Community to support independent journalism and get the raw Business of Tech interviews, Channel Intelligence briefs, monthly member calls and more. Member support is one way we sustain the costs of investigative work — from court filings to video editing — while keeping the core reporting open to all.

Thank you

Join the Small Biz Thoughts Community

Choose your upgrade:

Get the full benefits of Business of Tech Plus

Insider Access

$12/month

Perfect for MSPs and ITSPs that want full interviews, early access, and ad-free listening

  • Programmatic Ad-free private podcast feedSame show, little interruptions
  • Channel Chatter previews1–2 topics with light insights
  • Early access to interview episodesHear it days before public release
  • Monthly Insider BriefTighter analysis you can share internally
  • Extra audio segmentsCut interviews, behind-the-scenes commentary, quick competitive notes
  • Become an Insider for $12/month

    Leadership Access

    $149/month

    Perfect for MSPs and Vendors that run a team and need the extended tactics, executive summaries, and weekly alignment brief

  • All Insider Access benefits plus . . .
  • Invite your teamIncludes access for 5 team members with option to add more
  • Vendor Strategy BriefsThe entire library, plus new analysis every month
  • Channel ChatterAll topics, full insights, complete vendor discussion + sentiment list
  • Quarterly State of the Channel Briefing
  • Monthly AMA submission priorityAsk Dave direct questions, and skip the line
  • Get the Leadership Edge for $149/month

    Vendor Partner

    $500/month

    Perfect for channel companies or vendors looking to deepen their engagement with the show.

  • All Leadership Access benefits plus . . .
  • Get highlighted as a show sponsor You'll get placement in the show notes, throughout the website, and on our dedicated sponsors page.
  • Enjoy regular shout outs You'll be featured in a rotating format during the show
  • Become a show sponsor for $500/month

    Search all stories