Digital transformation

The Magnum Ice Cream Company is preparing for a digital renaissance following its spinoff from Unilever, an undertaking that Michael Friedlander, CIO of the Americas, described as a once in a lifetime opportunity.

Home to some of the largest ice cream brands globally, including Ben & Jerry’s, Breyers, Magnum and Klondike, the company completed its exit from Unilever in 2025, becoming a standalone company.  

Over the next two years, Friedlander said The Magnum Ice Cream Company will deploy new IT systems under a phased approach — including new ERP and CRM systems, website management and supply chain capabilities. In the meantime, the company is operating under transition services agreements with Unilever, relying on the former parent company for IT functionality and support.

Looking to The Magnum Ice Cream Company’s technological future, Friedlander said he’s excited for the possibilities of building a new tech stack — unencumbered by legacy systems — and positioning the company to take advantage of the latest AI developments.   

“We’re taking this opportunity to standardize the processes where we can and invest in platforms that are also investing in AI capabilities,” Friedlander told CIO Dive. 

Data, partnerships and people  

As The Magnum Ice Cream Company prepares to implement new technologies, Friedlander said he is focusing on the critical underpinning of its future tech stack — data. 

“None of this works without clean data,” Friedlander said.

The Magnum Ice Cream Company’s digital transformation isn’t Friedlander’s first modernization rodeo. 

Before joining The Magnum Ice Cream Company in August 2025, Friedlander spent nearly four years at PepsiCo as senior director of strategy and transformation. Before that, he was the head of U.S. IT at Toll Group, leading the logistics and transportation company’s efforts to deploy Oracle financial systems and evaluate digital technologies that could transform business processes.  

As Friedlander brings his experience to bear at The Magnum Ice Cream Company, he views clean, accurate and available data as critical to deploy future AI capabilities. The right partnerships with large technology providers are also key.

CIOs are fielding a flurry of vendor pitches offering the latest and greatest AI tools and platforms on the market. Competition for enterprise customers is heating up for hyperscalers, ERP providers such as Oracle and Salesforce, and even large language model providers OpenAI and Anthropic, which are looking to earn a greater slice of enterprise spend. 

When assessing companies to partner with, Friedlander said he’s largely focused on how they’re building AI into their services and platforms. He’s also steering clear of highly customized systems, which he said can limit scale and slow upgrade capabilities. 

“We’re very much partnering with the companies that are investing heavily in AI capabilities and are sound and proven, with some outliers here and there,” Friedlander said. “We’re trying to make sure we’re integrating big platforms together. That’s really going to help us unlock AI faster, primarily because we’re going to do more configuration than customization, which I think is the huge unlock for the business.” 

AI will touch every function across The Magnum Ice Cream Company’s operations, including finance, HR, supply chain and marketing, Friedlander said. In supply chain, for example, Friedlander said he’s considering how to use AI-powered camera recognition to prevent safety incidents in factories. Meanwhile, in HR, he’s focused on bettering the employee experience and building an HR assistant into workflows. 

To help steer future adoption, the company will be focused on training employees in proper AI use and looking out for AI-related threats. 

“It’s all about change management,” Friedlander said. “People are the main part of our company…and we have to have those people armed with the right resources to be effective.” 

Time invested in learning and preparing now will help when new systems go live, he added. 

“We have to make sure we’re operating in a way that the data is clean, the processes are understood, so when the tech does land, we’re in a strong position to have a great kickoff event rather than work backwards and say, ‘Well now let’s learn everything,’” he said.

If CIOs need a use case to convince businesses why modernization is a critical investment, they got it with AI. Today, old-school infrastructure can slow data- and compute-hungry enterprise AI ambitions. 

But building the modernization roadmap can be tricky for technology executives, who must balance what processes to upgrade now while keeping legacy systems operational — and all of it in budget. Sometimes, the best plan of action is to determine what not to modernize.

To help ease decisions, Liberty Mutual Insurance puts the focus on key metrics and the outcomes that a modernization effort can yield, said EVP and Global CIO Monica Caldas, speaking during a panel at the 2026 MIT Sloan CIO Symposium. 

"We did a technical debt radar as an example, to understand what we need to do now, versus what we need to monitor versus what is stable," Caldas said. The framework includes variables to help the company understand, from a business and product perspective, where the focus should lie.

Travelers Insurance took a similar approach, focusing its modernization efforts on systems that elevated customer experiences or boosted operations. 

"You want to modernize what's in the critical path," said Mojgan Lefebvre, EVP and chief technology and operations officer at Travelers, speaking on the same panel. "We've chosen to not modernize some things and actually isolated subsystems behind APIs and then invested around it where the customer value or productivity was."

AI adoption plans are influencing modernization recipes. Travelers is aiming to identify priority areas, starting where AI deployment is critical, Lefebvre said. For businesses struggling with data accessibility and AI adoption plans, modernization is key, she said.

"At the end of the day, the goal is to, in the right order, modernize as much as possible," Lefebvre said.

AI-focused upgrades

Insurance industry businesses such as Travelers and Liberty Mutual have pushed forward to adopt AI, with an eye on applications that can help lower operational costs and improve customer service. Much like their peers in the banking sector, modernization of legacy systems has been a prerequisite to tackle use cases.

Travelers is reaping the rewards of earlier cloud modernization efforts, which have helped establish a foundation for AI, according to Lefebvre.

"We were lucky at Travelers that, eight years ago, we very aggressively started modernization on our platforms," Lefebvre said. "We've got 70% of our compute in the cloud."

That foundation has enabled the insurer to pursue AI adoption initiatives including an opportunity to rethink parts of the claims process, Lefebvre said.

"We're thinking very differently about how you handle a call," said Lefebvre. The company plugged automation into the way personal auto insurance customers can enter their first notice of loss after physical damage. The change is allowing most customers to complete the effort with support from AI agents.

Modernization decisions also require clarity around alignment at the executive level, Caldas said. 

"You have to be aligned at the top of the house, because you're making decisions about investments," Caldas said. "But then it doesn't stop there — it has to actually trickle down."

Sponsored content

By nvisia

Most technology leaders have some version of the same conversation on their calendar right now. AI progress needs to be demonstrated. A timeline has been set and somewhere between the directive and the delivery, someone has to account for the infrastructure underneath it — the systems that were built before the architecture AI actually requires existed.  

The ongoing AI hype cycle is giving this fact short shrift and it’s creating confusion around what modernization actually means, causing transformation efforts to stall in a predictable place.  

The plain truth is that legacy systems and applications aren't suddenly going away — their codebases aren’t being magically transformed into modern, sustainable, AI-ready building blocks. They accumulate cost, complexity and dependency quietly, until a new initiative tries to build on top of them and reveals that they're the ceiling. 

You can't build intelligent systems on top of fragile, undocumented, architecturally brittle infrastructure. Not in a way that holds. 

What "legacy" actually means 

Part of what makes this hard is that the word "legacy" tends to get used as though it means "old." It doesn't. A system becomes legacy when it's difficult to change, expensive to maintain, or no longer aligned with what the business needs to do next — and that can happen to a system built three years ago as easily as one built 30 years ago. The diagnostic question isn't how old the system is. It's what the system is preventing the organization from doing. It’s a matter of capability.  

That reframe is extremely important for AI readiness. The systems that block AI adoption aren't always the oldest ones in the environment. They're the ones with data that isn't structured or accessible for AI consumption, integration surfaces that AI tooling can't work with and architectures that weren't designed for the workflows AI agents will need to participate in. In other words, legacy is less a description of vintage than a description of capability gap. That’s always been true. But the drive for AI is exposing it like never before. 

The discovery problem nobody admits 

The second reason modernization programs stall is one most vendors won't acknowledge directly: no discovery process finds everything. Legacy systems of any consequence contain undocumented integrations, informal dependencies and institutional knowledge that lives in the heads of a few engineers who've been maintaining the system for years. When those engineers leave, that knowledge goes with them — and any modernization program that starts after that departure starts from a much harder position. 

What's changed is that AI-assisted analysis can now surface complexity that traditional discovery either misses or takes months to find.  

The goal of discovery, properly understood, isn't to find everything. It's to build enough confidence in an approach to commit to it, while structuring the program so that what discovery doesn't find becomes an input rather than a crisis.  

What modernization actually produces 

A modernization program designed for AI readiness produces something different from one designed simply to replace old infrastructure. The distinction is in what the organization inherits at the end: not just a modernized system, but one designed from the beginning so that AI is a first-class operational capability in the future state — with data structures, APIs and integration patterns that AI workflows can actually use. 

This is the difference between a system that was built using AI and a system that's ready for AI-enabled operation. The first is a statement about how the work was done. The second is a statement about what the organization can build next. Both need to be true for a transformation to really take hold. 

The confidence question 

The practical challenge for any organization in this position isn't that the case for modernization is unclear. It's that the cost of committing to the wrong program — or committing before there's enough information to commit well — is high enough to produce inaction. The standard offer from the market is a large commitment against an estimate built on discovery that left most of the risk undefined. 

A more defensible approach starts with a fixed-cost diagnostic phase that produces something specific: an architectural blueprint, a modernization roadmap, a risk profile, a delivery strategy and an AI readiness assessment. Evidence the organization can use to make a fully informed decision about whether and how to proceed — before committing to the full program. 

For organizations carrying both the AI mandate and the legacy infrastructure problem, that's often what the conversation needs to become. Not two problems running on parallel tracks. One decision, made with the right information. 

Multinational retail group Amer Sports spent the last few years revamping its technology backbone, upgrading key systems that support operations. In 2026, the company plans to boost its IT investments with an eye toward embedding AI and amplifying its employees’ strengths. 

The Helsinki, Finland-based company, which operates outdoor and sports brands such as Salomon, Arc’teryx and Wilson, expects its capital expenditures to reach $400 million in 2026, up from $310 million in 2025. The increase is driven in part by strategic investments in IT infrastructure, according to a February Q4 2025 earnings call. 

The company’s key IT investments are concentrated in enterprise systems modernization, digital commerce infrastructure and data and planning capability, Donghai Chen, chief digital officer at Amer Sports, told CIO Dive in an email.

AI is a significant part of the company’s infrastructure investments, with an active AI program operating in parallel with the company’s core systems. Chen said he frames the relationship between infrastructure and AI investments as “deliberately complementary.” 

“Core platforms and systems establish the foundation that enables AI to be applied meaningfully at scale,” he said. “AI initiatives, in turn, accelerate and amplify the value created by those foundational investments. The two reinforce each other, and the sequencing matters. When AI is introduced without sufficient foundational capabilities in place, the impact tends to remain localized and experimental rather than translating into sustained productivity gains.” 

Replatforming efforts

Amer Sports is in the process of migrating its finance, supply chain and HR business platforms onto a modernized, cloud-based infrastructure. 

The centerpiece of the company’s enterprise systems modernization is its SAP S/4HANA program, which Chen said will “serve as the operational backbone across our brands and geographies.” Amer Sports began its RISE with SAP journey in May 2025 to speed the company’s digital transformation, according to a press release. 

The company is also in the first phase of deployment of a human resources information system implementation to modernize HR data and processes across the group’s brands, Chen said. 

Digital commerce infrastructure is another critical piece of the group’s IT investments, with brands at different points in the transition from wholesale-oriented technology stacks to purpose-built, direct-to-consumer architectures, Chen said. 

“Building the commerce, data and supply chain infrastructure that premium brands require is a significant part of the investment,” he added. 

The group’s third IT modernization pillar focuses on data and planning capabilities. Chen said Amer Sports is investing in integrating demand and supply planning tools, enabling brands to run more advanced forecasting and inventory management. 

Embedding AI

Retail firms have turned to AI in search of operations and customer experience improvement as weakening employment rates and high fuel costs create challenging shopping environments for consumers.

Companies pursuing AI deployments are often focused on use cases in marketing, IT and digital functions, digital commerce and merchandising strategy and pricing, according to a BRG report. 

AI is already playing a part within Amer Sports’ IT infrastructure, with generative AI tools embedded in daily workflows across productivity, content, analysis and operational support, Chen said. 

When implementing AI, Chen said his goal is to remove friction and repetitive work that gets in the way of employees applying their expertise and deep knowledge of the group’s brands, products, customers and markets. “AI creates the most value when it helps people focus on what they do best,” Chen said.

Last year, Amer Sports partnered with Personal AI, a small language model platform based on identity-based memory architecture that learns from a company’s data, memory and other patterns to provide a more personalized AI experience. 

When it comes to successful AI use cases, Chen said what matters more than deployment breadth is depth of integration into workflows. He said an AI use case counts when it is regularly used and “measurably changing how work gets done, not when someone has logged in once.” 

Getting thousands of employees onboard with new systems and adapting to new tools is a challenging part of any business transformation, he said. 

“I want to be direct about the state of AI in our industry: It will be transformative, but how exactly and on what timeline is still genuinely uncertain,” he said. “The organizations that will win are those that build the governance and the agile capability to move quickly when the right use cases become clear, rather than those that over-commit to a specific vision today that may need to change.” 

The responsibilities of CIOs continue to shift, with tech leaders listing the alignment of IT strategy to business objectives as the role’s most important job, overtaking 2025’s priority of managing cybersecurity, the CIO 2026 Outlook report from Experis found. The tech services firm surveyed 1,900 CIOs and other tech leaders on how they’re navigating technological change and the future of work. 

Amid the ongoing role changes, AI investments are coming into focus for IT chiefs, with more than half reporting positive ROI and adequate investment levels in their organizations. However, nearly one-third believe their companies are overinvesting in the technology. 

In this transition period, the leaders that treat technology as a business leadership function with complementary investments will succeed, said Kye Mitchell, president of Experis U.S., in a statement. "CIOs are being asked to lead AI transformation, drive growth, improve productivity, and manage risk all while facing significant talent shortages,” she said. 

Enterprise tech leaders are balancing shifting responsibilities and talent acquisition challenges while trying to prove financial viability for their tech projects. 

The report found tech leaders feel pressure to make timely decisions, manage cybersecurity risks and stay competitive with new technologies. Nearly half of respondents said the top barrier for CIOs is to keep up with the pace of change. An April survey from Writer found 61% of tech leaders fear for their jobs if they fail to navigate tech transformations well. 

As marrying tech projects with business outcomes becomes more important to the C-suite, 61% of CIOs said they feel their peers do not fully understand the CIO role, up from 49% last year. 

"Many CIOs are finding themselves in a balancing act right now,” Mitchell told CIO Dive in an emailed statement. “On one hand, there is tremendous pressure to invest in AI and keep pace with innovation. On the other, boards and executive teams are increasingly asking where the business value is coming from.”

Despite plans for adoption, many organizations are not adequately addressing the talent needs within their teams, the report found. Fewer are conducting programs for rising talent than last year, and IT leaders are looking to external candidates to fill open roles rather than promoting from within, Experis found. 

In an era where tech leaders will need to prove their AI and IT spending is worth it, CIOs should avoid deploying AI for AI's sake, Mitchell told CIO Dive. It’s easy to get caught up in the excitement and competition, but it’s a waste without training people and defining the business problem at hand, she said. 

“AI has enormous potential, but the organizations realizing the greatest ROI are the ones aligning technology investments with clear business priorities and helping their people adopt new ways of working,” Mitchell said. 

The shift has thrown some CIOs — and their budgets — for a loop, grabbing headlines in the process.  

GitHub shifted from offering a flat price for its premium request model to charging based on token usage for “input, output, and cached tokens, according to the published API rates for each model,” per the company’s announcement. Zendesk and Workday also rolled out changes to their pricing structure as AI features expand. 

“Many people look at AI as a technological upgrade,” said Marko Markov, technology, media and telecommunications senior analyst at RSM. “It’s not just a technological upgrade. It’s really a fundamental shift in how SaaS companies will operate.” 

Where before, SaaS vendors would price on a per-person model, they’re now moving towards charging either partly or totally on consumption. More than half of technology executives say they expect usage-based revenue to grow by 2027, according to a Revenera report.

Dhaval Moogimane, software industry lead at West Monroe, said that charging on a per-head subscription model may not make sense to AI companies if the technology is used to automate processes. For example, if a software provider works in the customer service space, the number of agents will reduce over time, he said. Those vendors have to “find another metric for pricing, typically moving to a consumption or outcome based pricing.” 

For enterprise technology leaders, the updated pricing structures can mean going back to the drawing board in terms of strategy. 

Michael Corrigan, CIO at World Insurance Associates, told CIO Dive that AI pricing models with their vendors have changed "very rapidly" over the last three to six months. “We’re a big SaaS user for most of our systems, and we’re used to paying a per-seat cost based off of growth of organizational headcount.” 

Many vendors the company works with are switching to a combination of seat license and consumption based pricing, Corrigan said. And while seat-based costs are easy for them to plan for, token burn is not. 

“What we have to do is a little bit differently than in the past and establish those baselines and forecasts,” he said, then work with vendors to “really flesh out what your expected costs are going to be out front.” 

Corrigan also keeps in mind that while cost for AI embedded in SaaS may increase, the tools are “giving you and your employee either time back in their day or quick processes and more consistent processes.” 

Getting through pricing transitions

The pricing model updates signal a moment of transition, Markov said. CIOs will need to get an understanding of the tokens they require. That’s especially true for CIOs who find that their organization either has surprisingly high bills or runs out of tokens too quickly. “They’ll be able to forecast token usage and token cost a little bit better.”

Understanding usage will also help CIOs adjust software costs based on seasonality instead of paying a per head license for an entire year, Markov said.

“If your hotel chain is using more tokens in the summer, you can forecast for that and as a CIO, you can adjust your budget,” he said. The strategy would also limit token use in the slower winter months. 

Corrigan anticipates that forecasts will help them understand the cost of different scenarios, which will in turn help the IT department understand how employees are burning through tokens — not just in IT but across the organization. 

“We’re trying to get our arms around that right now," Corrigan said. "It’s a moving target.”

Rethinking spend strategies

While this pricing shift affects everyday business tasks, it can also affect innovation, which could create a process and financial headache for CIOs. 

Companies with in-house developers running AI-driven pilots, might need to reframe their approach to experimentation. “AI adoption is an R&D process,” said Alex Bakker, director and primary research lead at ISG. Some of those developments aren't going to work, which might make AI token spend seem a waste. But if something hits, it could yield big results. “If you’re comfortable with the fact that some of your research has no outcomes, you’re going to be in a better position.”  

For companies that aren’t comfortable taking that risk, the best approach is to rely on commercial vendors with built-in AI features. 

CIOs are still setting expectations around what AI can do, and what it should cost, Bakker said. If someone uses a free ChatGPT account and sees it works really well for basic tasks, they might expect AI at work to be able to do complicated ones with the same ease, and cost little to nothing for it. 

“That positive surprise probably created a bit of irrationally high expectations,” he said. “If you go in with low expectations there's room for positive surprise. If you go in with high expectations that this thing is going to cure cancer and should also be able to do my accounts payable processes, it can’t do that.”

The 2,000 largest public companies globally are collectively sitting on trillions of dollars in untapped value from AI investments, with progress stymied by internal operational weaknesses such as poor data quality and inefficient processes, according to a report from technology firm Genpact and global research and advisory firm HFS Research.

Many companies have yet to address the foundational problems limiting AI investment returns, the study found.

“AI is exposing every weakness enterprises have spent decades learning to live with,” Phil Fersht, founder and CEO of HFS Research, said in a statement. “Poor process discipline, fragmented data, legacy technology and talent gaps are no longer operational nuisances. They are now direct barriers to growth, productivity and competitiveness.”

The study estimated that the world’s top 2,000 public companies collectively have nearly $18 trillion in unlocked value tied to unresolved “enterprise debt” — a combination of outdated technology, poor-quality data, inefficient processes and workforce readiness gaps.

HFS and Genpact said they calculated the aggregate value at stake by applying respondent-reported revenue uplift and cost reduction estimates across the combined revenue base of the Global 2000. 

The report argued that technology, data, process and talent debts are interconnected and often reinforce one another. Poor data quality can hinder process improvements, while outdated technology can make it more difficult to deploy AI tools effectively. Workforce skill gaps further complicate efforts to address the other challenges.

“These interconnected enterprise debts do not appear on financial statements, yet they are quietly keeping agentic AI trapped in pilot purgatory,” the report said.

Organizations that successfully address these issues could achieve about 8% faster annual revenue growth and reduce costs by 16% annually, according to the study.

The research found that proven organizations do not address enterprise debt in a linear sequence. Instead, they operate at what the report calls “dual velocities,” working to fix foundational weaknesses while simultaneously pursuing higher-impact transformation initiatives.

The report noted that while these investments rarely deliver immediate quarterly results, they build the organizational capacity needed for other initiatives to scale and compound over time — balancing the CFO’s focus on near-term performance with the CEO’s mandate for longer-term transformation.

Eighty-five percent of leaders surveyed by HFS and Genpact said enterprise debt is actively constraining the value generated by AI initiatives, while more than half said their organizations lack a funded plan to address the issue.

Only 6% of respondents were classified as “proven debt resolvers” — organizations that have established, implemented and measured enterprise debt reduction programs at scale.

“Proven resolvers treat debt resolution and agentic transformation as one program, owned at the top, funded as a portfolio, and sequenced to build capability, not just fix visible pain,” the report said.