Cut, Then Rehired: CX Staffing Is Set to Rise 20% as AI Efficiency Starts Costing Loyalty

For two years the customer experience function has been one of the easiest places in the business to find savings. Contact volumes could be deflected to chatbots, frontline teams could be thinned, and the efficiency gains showed up neatly in the cost line. Now the analysts who tracked that wave are calling the turn. Forrester's 2027 predictions say CX headcount is about to grow again, and its reasoning should concern every revenue leader who relies on renewals: companies cut too deep, and customers noticed. The cost of that lesson does not land in the service budget. It lands in retention, expansion and the forecast that depends on both.

The Correction Forrester Sees Coming

In its 2027 predictions for B2C marketing, CX and digital, released October 21, Forrester says CX staffing will rise by 20% after overzealous cuts. The hiring will not simply restore the old org chart. Forrester expects companies to add natural language processing engineers, data scientists and AI experience designers, the people who identify which AI use cases are worth pursuing and make them work. "The challenge for leaders in 2027 will be gaining greater control without limiting growth," said Sharyn Leaver, Forrester's chief research officer.

The regret behind that rebound has been building for a while. Forrester's Future of Work research, as reported by ITPro, found that 55% of companies that cut staff in favor of automation regretted the move, and predicted that much of the rehired work would go offshore or to lower-paid roles. The rebound, in other words, is not a return to how service looked before AI. It is a second attempt at getting the mix of people and machines right.

None of this means frontline service jobs are coming back in volume. In a separate forecast, Forrester's Kate Leggett and Laura Ramos predict that AI will cause 49% of current customer service jobs to disappear by 2030, with high-volume contact centers seeing proportionately more reductions. The picture is fewer generalist agents and more specialists who design, tune and govern the automation. That is a different workforce, not a bigger version of the old one.

Where the Money Has Been Going

The spending data explains how the function got here. A Gartner survey of 199 service and support leaders, conducted in April and May 2026 and covered by CX Today, found that AI spending had risen 38% while overall service and support budgets grew just 2%. With the total budget nearly flat, that AI money has to come from somewhere else in the function, and people are the largest line. Leaders in the survey expect generative AI chatbots to become their most valuable customer service channel within two years, ahead of live chat and voice bots. "The challenge is ensuring those investments produce measurable business value," said Kim Hedlin, a director analyst in Gartner's customer service and support practice.

That measurement gap is where the risk sits for revenue teams. A deflected ticket shows up as a saving the day it happens. A customer who felt brushed off by a bot shows up months later as a softer renewal conversation, a smaller expansion or a quiet churn decision, and by then nobody connects the two. Leggett and Ramos note that 85% of decision-makers in Salesforce's State of Service research expect service to contribute a larger share of revenue. Service is being asked to grow revenue at the same moment it is being measured almost entirely on what it saves.

Efficiency Is Not the Same as Loyalty

Forrester's CX predictions for 2027, written by Rich Saunders and Shar VanBoskirk, put the problem bluntly: efficiency strategies will eclipse meaningful connection, weakening customer loyalty. Organizations will deploy customer-facing automation mainly to cut costs, the authors argue, and many will discover that removing friction does not build loyalty when the experience lacks differentiation and human connection. Their advice is to study the return on AI versus human experiences and to identify the high-stakes moments where human contact pays off.

The same report sees the measurement itself changing. Forrester predicts one in four CX teams will transform legacy CX measurement with AI, and notes that 29% of CX decision-makers said in 2026 that their teams planned to boost customer insight and analytics capabilities. It also expects one in five companies to implement fully agentic journey orchestration, with 25% of AI decision-makers already working on AI for CX journey orchestration. For revenue leaders, the opportunity is to get into that measurement work early, so that the new systems track whether customers stay and grow, not just how cheaply they were served.

The CRO Playbook: Protect the Renewal Book From the Cost Line

• Map the high-stakes moments. Work with CX to name the interactions where a human must be available, such as onboarding escalations, outages and pre-renewal issues, and keep them out of deflection targets.
• Tie service metrics to revenue outcomes. Report retention, expansion and churn by service experience, so a deflection gain that costs a renewal is visible on the same dashboard.
• Get a seat in the 2027 rehiring plan. If CX is adding data scientists and AI experience designers, make sure some of that capacity is pointed at the journeys your largest accounts actually use.
• Ask churned and renewed customers directly. Exit interviews and renewal conversations reveal what a bot experience cost you in a way ticket data never will.
• Challenge chatbot ROI that stops at cost. Before a generative AI chatbot becomes the primary channel for key accounts, ask for evidence on satisfaction and retention, not just handle time saved.
• Treat service cuts as forecast risk. When service headcount or coverage changes, flag the affected accounts in the renewal forecast rather than waiting for the churn to explain itself.

The first wave of AI in customer experience was judged on what it removed. The second will be judged on what it keeps, and the customers it keeps are the revenue team's to lose. CROs who wait for the service function to correct course on its own will be reading about the results in their net retention numbers. The ones who step in now get to help decide where the humans go.