The CHRO's field guide to workforce transformation

ServiceNow is rewiring its 30,000-person organization around an AI-native go-to-market motion, and that shift is forcing role-level redesign across key parts of the business. The role did not exist in the catalog before AI reshaped what customer engagements demand.

As Nur Duygun, who leads ServiceNow's Skills Centre of Excellence and Skills Intelligence, has publicly described, the work is not a once-every-five-years job-architecture refresh; it is a focused, role-specific redesign triggered by an AI-driven shift in the underlying task profile. Designing it required mapping the task reality of customer-facing delivery and building the role profile around the capability combination AI-native delivery actually needs. ServiceNow uses Work Intelligence to read the role's task reality, identify where AI has shifted that reality far enough to warrant a redesign, and then refresh the role profile.

That role is the Forward Deployed Engineer (FDE). VP of Workforce Skills and Talent Readiness Josh Newman describes the FDE as a full-stack engineer who is also an expert in client-facing skills and “absolutely loves the green space,” sitting with a customer to co-build solutions to unanswered problems in real time. Who can step into the redesigned roles is a separate question, and the way ServiceNow approached it is covered in section 3.

A leading semiconductor company was planning its AI rollout across engineering and operations. The assumption going in: the highest-value opportunity was in automating repetitive manufacturing processes.

Work Intelligence told a different story. Task-level analysis revealed that engineering design review, not manufacturing, had the highest concentration of augmentable tasks; the engineers were spending 40% of their time on documentation and compliance review that AI could accelerate, freeing capacity for the design work that actually differentiated the company.

The transformation strategy shifted entirely. Instead of investing in manufacturing automation, the company prioritised engineering augmentation, and the ROI timeline shortened by a full quarter.

A global life sciences organization mapped 38 bottleneck roles that were constraining project delivery.

Work Intelligence decomposed each role into 25 tasks and scored them at 82% accuracy, revealing that AI augmentation could unlock tens of millions in stalled backlog revenue. The insight was not about cutting headcount; it was about freeing scientists and analysts to do higher-value work while AI handled the tasks consuming their capacity.

The company is now using this intelligence to de-risk over $100 million in CapEx and plan for 500 AI agents by 2027. Market Intelligence added the external context: the skills these roles required were in high demand on the external market, so internal augmentation was not only faster, it was cheaper than trying to hire the capacity.

A North American financial services company assumed it needed to hire hundreds of AI-skilled workers to staff its transformation program, but Skills Intelligence showed otherwise.

30% of the required AI-adjacent capabilities already existed inside the organization, in departments nobody had thought to look. Work Intelligence mapped which tasks in the target roles could be augmented immediately with existing talent, and Market Intelligence confirmed that hiring externally for those same skills would cost significantly more.

The premium economics made the decision obvious. The company reskilled internally, time-to-productivity dropped, and the transformation program came in under budget.

Before the mapping, internal mobility decisions relied on manager networks and self-nominations: the talent was there, but the data to find it was not. As the bank's then Group Head of Workforce Strategy and Insights described it, the trigger was concrete and market-specific: HSBC needed more wealth managers in key Asian markets, and "we know in some markets we're simply not going to find enough people to fill our ambition." That forced the question inward: "What skill sets do we have internally that we can channel towards that specific requirement in the future?"

HSBC mapped its 250,000 employees into 1 taxonomy in five months using Skills Intelligence. With skill profiles in place across critical technology populations, HSBC could identify non-obvious mobility pathways: employees in operations with the analytical skills to move into risk management, relationship managers with the data fluency to transition into AI-augmented advisory roles.

Market Intelligence is layered on the external context, showing which of these internal moves avoided the highest external hiring costs.

With the new Forward Deployed Engineer profile defined (covered in section 2), ServiceNow faced a hard market reality: FDE hiring has historically been run roughly 70% external and 30% internal, because the combination of AI engineering depth, customer-facing chops, and platform expertise is rare on the open market. VP of Workforce Skills and Talent Readiness Josh Newman put the squeeze plainly: “The companies out there that have the mass of these, they can spend whatever they want on talent.”

Working with TechWolf, the team used Skills and Work Intelligence to read the actual task and skill signal beneath every employee and surface internal candidates the org chart alone would miss. The funnel Josh walked through: 29,000 employees, narrowed to roughly 15,000 with some skill adjacency to the FDE profile, to about 1,000 high performers, to around 500 with expert AI proficiency, and finally to roughly 100 who also carry deep ServiceNow platform expertise.

That last cohort is small enough for the FDE talent team and subject-matter experts to review individually, and the first internal hires are now running through a 12-week rotational program. As Josh framed the shift: "We've moved from intuition to intelligence."

Belgium's Federal Ministry of Health, part of a 60,000-person federal government, started its skills journey in 2020 during COVID. Leaders were searching by phone and address book for employees with epidemiology or crisis communication skills, because the official ‘who's who’ only showed job titles. As Carolien Sonck, Director of Support Services, put it: "We have thousands of employees. It's impossible to know from all our employees what their skills are." Recruiting has tightened across government, and Carolien framed the operating reality plainly: "We need to do more with less, so we are looking inside our organization." The most pointed example sits at the intersection of AI impact and internal mobility. The Ministry had 35 in-house translators whose work is being absorbed by Copilot, DeepL, and similar tools.

With skill profiles now in place across the workforce, built by TechWolf from existing HR data and validated by employees, internal mobility became the lead use case. Rather than letting that drift into attrition, the team has run each translator's skill profile against open functions, identifying the closest-adjacent roles, and proactively nudging individuals toward them.

The same pattern is being applied to employees returning from long-term illness and to internal project staffing, where leaders can query the dashboard for skills like change management, GDPR, or Power BI instead of relying on personal networks.

Ericsson faced a transformation that would test any skills model: 100,000 employees across 140 countries, and rapidly shifting tech demands requiring new skill profiles at scale. Peter Sheppard, Ericsson's former Global Head of L&D Ecosystem, framed the design goal as 'visible simplicity': a system sophisticated enough to model supply and demand globally, yet simple enough that employees and managers would actually use it.

"No matter where you go, job ad, project board, learning system, you see the same skills, same logic," Sheppard explained. This was not a top-down mandate. It was a multi-tiered, business-led effort: an internal resourcing team from the business was involved from day one, a cross-functional steering group included HR, L&D, Resource Management, and IT, more than 20 skill area owners from the business defined skills at the ground level, and legal, IT, and platform teams ensured scale and compliance. Each year, Ericsson's top strategy team updates critical skill priorities while more than 90 global job role owners map job-specific skills from the ground up. Skill priorities are now refreshed annually from the top, while job-level skills are mapped from the ground up, giving Ericsson a continuously updated view of skill supply across the business.

97% of pilot users found value in the new skill profiles, with more than 77% accepting their inferred skills, but the deeper shift was how Ericsson made it stick.

The team took the data & analytics office workforce plan, recoded it in skills language, and produced two views the planning function had not had before: the future skills profile of the organization "if we do nothing" on the current trajectory, and the gap between that and the strategic ambition. The strategic anchor sits one level above the data layer: what the bank's then Group Head of Workforce Strategy and Insights described as "blueprints", high-level, business-agreed articulations of what the future workforce needs to look like, refreshed annually and used as the "north star" against which actual capability is measured.

HSBC mapped 250,000 employee skills into one structured taxonomy, with 30,000 technology employees receiving granular, individual skill profiles, all in five months. Across more than 2,500 roles and 18 skill domains, the planning architecture produces what HSBC calls a "Fulfilment Playbook": a decision engine that compares the organization's 2028 workforce blueprint against its live talent supply. At that scale, internal data without market context is meaningless, so Market Intelligence sits underneath the playbook, surfacing regional talent supply, competitive compensation dynamics, and how external skill markets will shape options over the next 18 months.

Rather than posting roles and hoping the right person applies, HSBC proactively identifies where to redeploy, reskill, or restructure.

A global life sciences organization needed to validate which skills existed, which were missing, and where to direct development investment, fast.

Using Skills Intelligence, they processed 95,000 skill approvals from employees in just 24 hours.

That level of workforce engagement replaced months of survey work with a single day of data collection, giving L&D leaders real-time validation of their skill gap assumptions.

At a North American financial services firm, the L&D impact showed up in an unexpected place: career conversations.

After deploying Skills Intelligence, managers reported that development discussions shifted from vague ambitions to specific, data-backed plans. 97% proficiency approval on the skills model confirmed the data quality, and when employees trust the data, they engage with the development paths it suggests.

As one VP HR Digital Solutions noted, managers said the data pushed away "career awkwardness", and employees came to conversations "much more focused and prepared than ever before."

Atlas Copco's CHRO, Cecilia Sandberg, described the shift they were after: "We didn't want a static taxonomy. We wanted a living skills map rooted in our own data and language, validated and benchmarked against external data. TechWolf was the one who made our data work across the whole stack." Skills data that lives in one system is an inventory; skills data that connects across recruiting, development, mobility, and workforce planning is infrastructure.

Atlas Copco also deployed a Skill Assistant in Microsoft Teams, letting employees update their profiles in the natural flow of work.

The hiring impact showed up within months of the early pilots. Dorna Shafiei, former VP Talent & Learning, confirmed: "Our pilot recruiters saw more qualified candidates and a broader mix of applicants." More diverse applicant pools emerged naturally: when hiring criteria shift from credential proxies to actual capabilities, the talent pool expands.

Workday mapped 15,000 employees to 300 roles in three months, a process that traditionally takes 12 to 18 months manually.

Josh Tarr, Director of Skills-based organization, called it "a game changer": "By working with TechWolf, we identified skills for 85% of our workforce in just three months, saving us time and resources while setting a foundation for our future talent strategy." The skills data created a shared language between hiring managers and recruiters, and both sides felt it.

The outcomes: 15% faster time-to-hire, 39% fewer low-performing hires, and 14% faster ramp to productivity. Marianne Herrmann, Principal Skills-Based Hiring Strategy & Enablement, saw the impact from the hiring manager side: "Hiring managers tell us they're more confident because they finally know what to look for and how to evaluate it." With the skills foundation in place, Workday's hiring outcomes improved across multiple dimensions, the result of a single data layer rather than a set of disconnected projects. Performance management, internal mobility, and career growth are the next areas in scope.