The way HR works is undergoing a profound evolution.
Previously, we embraced an employee-first mindset – designing services around the people who use them.
Now, we turn to how HR’s operating model itself must change to support that mindset.
From the classic Ulrich model of HR service delivery to efficiency-driven shared services and onward to today’s emerging AI-enabled, “agentic” HR frameworks, each stage has brought new opportunities and challenges.
The fundamental question is:
How can HR deliver efficient, scalable support without sacrificing a personalized, human-centric employee experience?
In this piece, we explore the journey of HR operating models: where they started, why they must change, and how AI and autonomous agents are opening the door to a new era of HR (see below figure for a decade-by-decade view).
Along the way, we’ll see that today HR manages cases, but tomorrow HR will manage AI agents, flipping “human-in-the-loop” into “human-on-the-loop” oversight of intelligent systems. We’ll also discuss how HR knowledge and policies need to be structured for AI to interpret, and propose fresh thinking on HR’s roles, governance, and accountability.
The goal is a next-gen model that blends efficiency with empathy, harnessing technology to deliver responsive, employee-first service at scale.
For decades, the dominant template for HR organization has been the Dave Ulrich three-pillar model. Introduced in the 1990s, Ulrich’s model split HR into three components:
This “three-legged stool” was revolutionary in its time, freeing HR from purely administrative roles and aligning it more closely with business strategy.
Companies embraced the Ulrich model to increase efficiency and strategic impact: routine inquiries would be routed to a service center or self-service portal, CoE experts would craft enterprise-wide programs, and HRBPs would work on strategic people issues with business leaders.
In the 2000s and 2010s, many organizations doubled down on the efficiency aspect of this model. HR functions consolidated into shared services hubs, implemented global processes, and often introduced outsourcing or offshoring for transactional work.
Some evolved into Global Business Services (GBS), integrating HR services with other corporate functions to streamline operations.
The results were undeniable: increased standardization, lower costs per transaction, and faster processing times in many cases. HR became leaner and more cost-efficient, speaking the language of SLAs and KPIs. Internally, HR could demonstrate productivity gains by handling large volumes of employee queries and HR transactions with smaller teams.
Yet, even as this efficiency-focused model matured, cracks began to show. A common criticism was that HR’s responsiveness and personalization suffered.
Employees often had to navigate a complex web of ticketing systems and FAQs to get help, leading to frustration. HR business partners, intended to be strategic advisors, became bogged down in administrative tasks and too many direct reports, limiting their strategic bandwidth.
In fact, many CHROs have observed that the classic Ulrich model is no longer sufficient for today’s challenges – HRBPs often lack the time or skills to keep up with emerging needs, and CoEs can become inflexible silos that respond too slowly (McKinsey).
As one HR leader put it, “AI is putting traditional HR operating models under pressure”, forcing a rethink of the old structure (TI People).
What changed?
In a word, EXPECTATIONS.
The workforce has evolved – today’s employees (increasingly millennials and Gen Z) expect the same level of speed, personalization, and user-friendliness from workplace services as they get from consumer apps.
The rise of the employee experience (EX) movement has put a spotlight on moments that matter in an employee’s journey, from onboarding to career growth to everyday helpdesk inquiries.
HR can no longer succeed by only being efficient; it must also be responsive, empathetic, and tailored in its approach.
The traditional model excelled at efficiency through standardization, but employees now demand flexibility and personalization – which can clash with one-size-fits-all processes. The below figure visualises the historic efficiency-versus-experience frontier and how AI promises to shift it:
Leaders in the field recognize this shift. McKinsey’s research with senior HR executives found that excelling in employee experience is now seen as one of the most impactful ways to win the race for talent In practical terms, that means offering more individualized HR services to meet diverse needs.
For example, one of the “innovation shifts” reshaping HR is the mandate to offer individualized HR support – treating employees more like unique customers, rather than entries in a database.
It’s not that efficiency no longer matters (it does), but the balance has tipped: back-office efficiency must go hand-in-hand with front-end experience.
A process that is cheap but leaves employees confused or dissatisfied is not a win in the modern HR scorecard.
This realization is prompting HR leaders to reassess their operating models. Many organizations are extending or tweaking the Ulrich model – sometimes called “Ulrich+” – to better address agility and experience.
In an Ulrich+ model, HRBPs may take on more execution work from CoEs, CoEs themselves become smaller expert teams, and a strong digital backbone supports self-service and analytics.
The below figure contrasts the original Ulrich ‘three-legged stool’ with three newer archetypes, illustrating how roles shift as employee experience and AI maturity rise:
According to one survey, about 47% of HR leaders say they aspire to an employee-experience–driven operating model, nearly as many as those sticking with an enhanced Ulrich approach. By contrast, very few (only ~6%) have moved to a fully “machine-powered” model where automation and AI dominate.
HR must evolve beyond efficiency for its own sake.
An employee-first mindset means rethinking processes that were designed around HR’s convenience rather than the end-user’s. For instance, traditional HR service delivery often started with case management and ticket tracking (focusing on HR’s internal efficiency).
By flipping that perspective to start with the employee’s need, some forward-thinking HR tech platforms have demonstrated it’s possible to reduce issues before they become tickets.
As an example, Applaud (an HR service delivery platform built around employee experience) deliberately prioritizes tier-zero self-service for employees over back-end case management. This employee-first design empowers people to get what they need without even creating a ticket, in turn reducing overall case volume (Brandon Hall).
The impact is twofold: employees enjoy a less frustrating, more seamless experience, and HR teams are freed from handling countless repetitive queries. This kind of approach underscores the new ethos: if you improve the experience for the employee, you often improve efficiency as a by-product.
Into this landscape of high expectations comes a game-changing catalyst: artificial intelligence. Particularly in the last couple of years, AI – and in particular, the advent of powerful generative AI and large language models – has opened new frontiers for HR service delivery.
We’re not just talking about chatbots answering FAQs. We’re talking about autonomous “agentic” AI systems that can handle complex tasks, make decisions, and continuously learn with minimal human intervention (HR Executive).
In the context of HR, agentic AI might mean an AI that can manage an entire workflow (say, recruiting or onboarding) from start to finish, not only responding to inputs but proactively taking actions.
So, what is agentic AI?
Industry experts define agentic AI as AI with a high degree of autonomy – systems that don’t just analyze data and await human direction, but can perceive context, decide on a course of action, and execute tasks in a way that imitates human judgment.
As HR tech commentator Steve Boese explains, earlier “simple AI” tools in HR would, for example, scan resumes or answer basic questions and then hand off to a person.
In contrast, an AI agent could potentially run an entire hiring process: generate a job description, post the listing, source and screen candidates, schedule interviews, and even conduct initial assessments – all autonomously.
This is a leap from AI as a decision support tool to AI as an “independent” actor within HR processes.
Right now, in most organizations, when an employee has an HR issue or a task, HR manages the case, whether it’s a leave request, a policy question, or a job requisition, a human (or a team of humans) steps through the process, possibly aided by systems.
In the near future, we’re looking at a flipped dynamic: AI agents will manage many of these cases, and HR will manage the agents. In other words, HR’s role will shift from manually handling each inquiry to overseeing and refining the performance of AI-driven systems.
This is where the concept of “human-on-the-loop” comes in. Instead of a human in the loop of every transaction, the AI handles routine matters on its own; humans stay on the loop, monitoring the AI’s decisions and stepping in only as needed (for exceptions, approvals, or complex cases).
One Gartner prediction estimates that by 2024, virtual assistants and chatbots will handle nearly 69% of routine work currently done by managers, radically offloading administrative burdens.
For HR, that implies a manager (or HRBP) no longer spends hours each week updating forms or approving standard requests; an AI can do it, while the human focuses on more strategic or human-centric work.
Agentic AI in Action
How might this look in practice? Imagine an employee wants to update their parental leave status after a new child.
Today, they might search an intranet for the policy, fill out a form, and wait for HR to approve. In an AI-driven model, the employee could simply tell a digital HR assistant, “I’m having a baby, what do I need to do?”
The AI agent accesses the relevant policy, asks a few clarifying questions, auto-fills the necessary paperwork, and confirms the update – maybe even proactively schedules a conversation with a coach about return-to-work planning.
They join the HR team’s ranks, so to speak, as non-human teammates. In fact, Mercer research calls 2025 “the year of agentic AI” in HR – a time when these digital agents will take automation to the next level and become a top value driver for businesses.
Executives increasingly see AI as a priority, and those HR leaders who fail to prepare for autonomous agents may be left behind. Crucially, agentic AI isn’t just about doing the same HR tasks faster – it enables new possibilities. Volker Jacobs of TI People notes that AI in HR brings change along three dimensions: Efficiency, Innovation, and Personalization/Democratization (LinkedIn).
Yes, it can automate tasks to make HR more efficient. But it also allows HR to do things it simply couldn’t do before (innovation) and to personalize and democratize services, tailoring support to individual needs and scaling it to many people at once.
For example, without AI, a company might only offer personalized career coaching to its elite leadership due to limited HR headcount; with AI, every employee might get a personalized career development plan or learning recommendations, because an AI agent can generate those at near-zero marginal cost.
As Jacobs puts it, with AI “services that without AI could only be provided to a few people can now be provided to many or all” – truly personalization at scale.
AI has the potential to break the old trade-off between efficiency and personalization. In the past, highly personalized service meant more HR time per employee (hence higher cost), and high efficiency meant standardizing things (often at the expense of tailoring).
Agentic AI can provide mass personalization – adapting to each user’s context in real time without a human having to intervene each time. As Mercer’s analysts observe, the best implementations of agentic AI “don’t just streamline processes — they optimize experiences”.
These AI agents can sense an employee’s behavior or needs and adjust their responses on the fly, delivering a level of immediacy and personalization that would be impractical for a human team at scale.
When routine issues are handled by AI, human HR professionals are liberated to focus on relationship-building, creative problem solving, and strategic planning – the human touchpoints that really require empathy and judgment.
In essence, done right, AI allows HR to become both more efficient and more human-centric at the same time. It’s important to note that we’re still in early days for this vision. Only a small minority of organizations have heavily automated, “machine-powered” HR models today.
Many are piloting AI in limited domains – for instance, using a chatbot to answer basic HR policy questions, or AI tools to shortlist candidates in recruiting. But the trajectory is clear. Surveys show nearly half of executives believe that rethinking workforce strategies in light of AI will be a major driver of ROI in the immediate future.
And Gartner predicts that by 2026, 20% of organizations will use AI to flatten their management structures, eliminating a significant number of middle-management roles in the process.
Such shifts herald a fundamental reimagining of work and management, including how HR itself is structured and where HR professionals spend their time.
So what does an AI-driven, or agentic, HR operating model actually look like?
It’s not as simple as installing a new chatbot on your portal. It requires reengineering HR’s roles, teams, and knowledge infrastructure to fully leverage these technologies. We turn to that next.
One of the most critical and often overlooked foundations of a successful AI-enabled HR experience is how your knowledge and policies are structured. AI assistants are only as helpful as the content they can access and understand. And while AI models have become more powerful, not all are created equal… and neither is your content.
It’s easy to assume that AI tools will “just know” what’s in a document. But the truth is, their effectiveness depends heavily on how that content is structured and on the capabilities of the underlying AI platform.
Rather than rewriting all your documents, the goal is to make them comprehensive, current, and structured in a way that AI can reliably consume.
Best practices include:
Leading organizations are also using AI to improve the content itself:
Platforms like Applaud make this process continuous, using actual user behavior to fine-tune the knowledge base over time.
As AI agents access sensitive HR data to personalize responses (role, tenure, location, performance history) it’s essential to put guardrails in place.
This is why knowledge management and AI governance must go hand in hand. If HR doesn’t control the quality, security, and clarity of what the AI sees, the risk of misinformation — or worse, a data privacy breach — rises fast.
💡Final Thought
The quality of your AI assistant is only as good as the content and rules you feed it. Getting your knowledge house in order may not be glamorous — but it’s the work that makes “always-on HR support” a reality employees love and HR teams trust.
To truly capitalize on technologies and meet rising expectations, many HR leaders are adopting a product mindset.
What does that mean in practice?
Traditionally, HR has thought in terms of services and processes – for example, the service of answering employee queries, or the process of creating a vacancy.
A product mindset, by contrast, means thinking of these offerings more like a product manager would: define the “user” (employee or manager) and their journey, identify their pain points, and continually refine the “HR product” (be it the onboarding experience, the performance management program, etc.) to better solve those problems.
HR can organize around product teams that own end-to-end employee journeys.
In fact, one emerging model is to form cross-functional HR product teams aligned to key stages of the employee lifecycle or important “user journeys.” For example, an organization might create product teams for:
Each team is responsible for all the HR offerings in that domain – as if it were a product suite – and for continuously improving them based on feedback and data.
These teams bring together various skills: maybe an HRBP or subject matter expert, a data analyst, a content or knowledge manager, an HR technology specialist, and even a UX or process designer.
Notably, this cuts across the old silos. Instead of a separate recruiting CoE, a separate HRIT team, and separate HRBP doing hiring for their unit, you get a “Talent Acquisition product team” with all those people collaborating towards a common goal (e.g., make hiring fast and great for candidates and managers).
A recent co-creation study by TI People with 15 major companies illustrated what this kind of model can achieve. In one case, a European firm reorganized its 25,000-employee HR function into five product teams mirroring key user journeys: Talent Acquisition, Employee Growth & Development, Performance & Rewards, Employee Care, and Workforce Intelligence.
Each team was staffed with a Product Manager (to prioritize initiatives and measure value), an Experience Designer (to craft user-centric processes and interfaces), an HR Technology Specialist (to implement and manage AI/tools), and domain experts drawn from the former CoEs.
They didn’t eliminate expertise – they redeployed it into these agile teams. A Product Council, including HR and business leaders, provided governance by reviewing the teams’ performance and setting priorities each quarter.
Technology-wise, they retained a core HRIS (SAP SuccessFactors in this case) but layered in new AI capabilities such as AI Assistants for HR and specialized AI for recruiting analytics.
Crucially, they also revamped their metrics: instead of just tracking internal efficiency (like number of tickets closed), they used a balanced scorecard that included experience metrics (employee Net Promoter Score, manager satisfaction) and business impact metrics (e.g. time to fill critical roles, retention of top performers) alongside efficiency.
The results were striking. In the first 18 months, this company saw a 32% efficiency gain in HR operations, and at the same time improved employee satisfaction with HR services by 28 percentage points.
In other words, by reorganizing around products and journeys (and augmenting with AI where appropriate), they made HR both cheaper to run and more beloved by its customers.
This dual improvement in cost and experience is the holy grail of HR transformation, and it was enabled by a combination of structure and tech. A product mindset encouraged the HR teams to continuously iterate and solve real user problems (not just hit process SLAs), while AI and automation handled many routine tasks (like basic inquiries and data processing) so that HR team members could focus on design and problem-solving.
It’s worth noting that this doesn’t mean CoEs or HRBPs disappear entirely – rather, their roles evolve. In the product team model, what used to be CoE expertise is now embedded “on demand” via domain experts who rotate into agile squads.
And some companies still maintain a pool of HR Business Partners, but perhaps fewer in number and focused on high-level consulting (e.g., advising senior leaders on workforce strategy) rather than day-to-day fire-fighting.
Freed from routine queries (handled by AI and self-service), HRBPs can spend more time on proactive initiatives like talent strategy, organizational development, or coaching leaders. Meanwhile, some organizations create “problem solver pools” – flex teams of HR experts who can be deployed to special projects or complex cases across product teams.
An agile HR operating model can swarm resources to the most pressing workforce problems, rather than rigidly assigning people by functional silo. Treating HR offerings as “products”, such as an internal mobility platform or a learning recommendation engine, comes with version updates, feature roadmaps, and user testing. HR might even adopt agile methodologies (sprints, backlog, product owners) in managing their initiatives.
In practice, this can look like agile pods working on, say, improving the onboarding experience this quarter, experimenting with an AI-driven buddy system, gathering feedback, and iterating quickly. This is a departure from the traditional multi-year HR program cycle; it’s more continuous and responsive.
Many CHROs believe HR needs to adopt these agile principles “to prevent HR from hindering rapid transformation” in the business. When the business is transforming rapidly (new strategies, M&A, shifting workforce expectations), HR must be able to flex and respond in weeks or months, not years.
As hinted above, new operating models introduce roles that didn’t formally exist in old HR org charts. HR Product Managers are one – people who blend HR domain knowledge with product management skills, focusing on end-user outcomes and continuous improvement of HR services.
Then we have Experience Designers or UX experts in HR, who apply design thinking to craft better employee interactions (be it an app interface or a workshop format).
HR Technology Specialists or AI Specialists are increasingly crucial – these are folks who understand both HR and the tech (AI, analytics, platforms) and can configure systems or even build no-code solutions. They might manage an AI chatbot’s behavior, or refine agentic steps to streamline a workflow.
Additionally, data analysts or people analytics experts become key players embedded in teams – their insights help personalize services and prove ROI.
Even within classic roles, the competencies are shifting: tomorrow’s HR business partner might need to be savvy in interpreting AI-generated insights, and tomorrow’s compensation specialist might spend more time training an AI model on compensation data than manually crunching numbers.
The upskilling challenge is real – HR professionals will need more digital, analytical, and consulting skills. Leading organizations are already investing in training HR staff on topics like data literacy, agile project management, and AI ethics. The payoff is an HR team comfortable working alongside AI tools, not threatened by them.
Deploying AI in HR doesn’t mean you let the machines run wild. As HR adopts a more tech-centric operating model, governance and ethical considerations become paramount.
HR deals with deeply human issues – fairness, equity, privacy, well-being – and introducing autonomous agents into the mix raises new questions.
Who is accountable if an AI makes a flawed hiring decision or gives an employee bad advice?
How do we ensure AI recommendations are free of bias, comply with laws, and align with company values?
These are not just IT questions; they are HR questions. Therefore, a robust governance framework must be part of the new HR operating model.
The above figure illustrates the layered control-tower HR will need to keep autonomous systems safe and fair.
HR leaders need to establish who has the authority to deploy or modify AI systems in HR, and who oversees their outcomes.
For example, an “AI HR Council” might approve new AI use cases and regularly review performance metrics and any incidents or errors.
Decisions about where to invest (e.g., which HR “product” gets an AI upgrade next) should be made deliberately, often jointly with business leaders (as in the Product Council example).
It’s wise to develop AI ethics guidelines for HR. These could cover principles like transparency (e.g., disclosing to employees when they’re interacting with an AI vs a human), non-discrimination (auditing AI decisions for bias against protected groups), and the requirement of a “human veto” on high-stakes decisions.
For instance, your policy might state that an AI can recommend candidates, but the final hiring decision is made by a human. Or that any AI-driven termination recommendation must be reviewed by HR.
Some organizations convene ethics boards or include diverse stakeholders (legal, DEI officers, employee representatives) to oversee AI in HR.
Since AI relies on data, governance must extend to HR data management to ensure data quality, privacy, and proper usage. GDPR and other regulations impose strict rules on automated decision-making and employee data, so HR must ensure compliance.
Limiting AI’s access to only necessary data, anonymizing data where possible, and securing data against breaches are all part of this accountability.
HR should treat AI “agents” somewhat like new team members whose performance needs appraisal.
Define metrics for the AI such as accuracy of answers, turnaround time, employee satisfaction with the interaction, error rates, etc. Many organizations set up dashboards to continuously monitor how the AI is doing.
If an AI HR assistant is deflecting 50% of queries but struggling with a certain 10% (say, questions about a new policy), that signals HR to improve the knowledge content or tweak the AI’s training.
Accountability mechanisms mean you don’t “set and forget” an AI system. Instead, you actively manage it, much as you would manage a human team member.
“Human-on-the-loop” design is critical for HR scenarios. This means building in checkpoints where a human can intervene.
For example, if an AI is unsure or detects sentiment like anger/sadness in an employee’s query, it could flag a human HR agent to step in. Clear escalation paths should be defined.
Employees should also have a way to appeal or question an AI-driven outcome. E.g., if an AI informed them they’re ineligible for a benefit, they should be able to contact HR to double-check. Providing that fallback builds trust that the AI isn’t a black box judge, but rather an assistant.
Interestingly, as AI takes on more, HR’s scope in governance may expand beyond the HR function. Gartner analysts note that leaders focused on AI adoption are now accountable not just for implementing tech, but for improving the worker experience and building organizational competency in the responsible use of AI (HR Future).
HR could play a leading role here, partnering with IT and compliance to educate the workforce about AI, set usage policies (for instance, guidelines on employees’ use of generative AI at work), and model ethical AI practices.
In a sense, HR may become the steward of “responsible AI culture” within the organization, ensuring that as we automate work we do so in a way that extends trust and fairness to employees.
All told, governance in an AI-driven HR model is about maintaining human accountability for what the technology does. HR cannot abdicate responsibility just because an algorithm made the call.
Instead, HR must proactively design and monitor these systems to uphold the organization’s values and legal obligations.
This is a new competency for many teams, but an essential one. It might involve new partnerships (with data science teams or external experts) and new tools (AI auditing software, etc.), but it’s integral to sustaining an employee-first ethos in a high-tech environment.
Reinventing your HR operating model along these lines is undeniably a big undertaking. However, it doesn’t have to happen overnight. Here are some practical steps and considerations for HR leaders looking to evolve towards a more AI-enabled, employee-centric model.
Align your HR leadership team around the need to shift, whether it’s stagnant employee satisfaction scores, an overwhelmed HR staff, or new CEO expectations for agility.
Articulate a clear vision of an HR function that is both more efficient and more engaging for employees. For example, paint a picture of employees getting instant, personalized answers 24/7, or HR business partners spending 80% of their time on strategic work instead of 20%.
Data can help make the case: perhaps your surveys show low employee effort scores when dealing with HR, or external research (from Gartner, etc.) shows a ~30% efficiency gain potential with AI.
Evaluate Your Current Operating Model: Take stock of how work flows today.
Where are the pain points for employees? (e.g., Do they bounce between HR contacts? Are response times slow?)
Where are the pain points for HR teams? (Too many hand-offs? Repetitive queries consuming time?)
Map out key journeys (hiring, onboarding, resolving a policy query, etc.) and identify bottlenecks or frustration points. This mapping will highlight opportunities for both quick wins and deeper changes.
As discussed, a solid knowledge infrastructure is prerequisite to any AI or even good self-service.
Audit your HR policies, FAQs, and documentation. Form a small task force to start building an employee-facing knowledge base if you don’t have one – prioritize the top 10–20 topics that drive inquiries.
Ensure the content is easy to understand (avoid HR jargon), up-to-date, and clearly structured in scannable formats. This not only helps any AI you introduce, but even short of that, it empowers employees and reduces load on HR.
As one step, you might deploy a modern knowledge portal or an AI-ready knowledge management tool to spot coverage gaps, score content and identify issues. Some HR teams also crowdsource questions from HR staff to make sure the knowledge base covers what employees actually ask.
Rather than attempting a full AI transformation at once, identify a use case to pilot.
A common starting point is a virtual HR assistant (chatbot) for answering FAQs about, say, benefits or time off. These are relatively well-bounded topics with fewer risks.
Choose a vendor or solution that allows an easy pilot (perhaps one that can integrate with your existing HR ecosystem). Feed it your curated knowledge base and set it loose in a controlled group (maybe one office or a particular function) and gather feedback.
Measure deflection rate (what percent of questions it handles) and user satisfaction. This will give you learning on how employees interact with AI and how well your knowledge prep paid off. Early wins here help build confidence and momentum.
Try forming a cross-functional team to tackle one employee journey or HR service that’s ripe for improvement.
For example, create an “Onboarding squad” with HR, IT, a hiring manager rep, and maybe a UX designer. Give them a mandate to redesign the onboarding experience within a quarter (perhaps including an AI-driven onboarding buddy or automated paperwork processing).
Equip them to use design thinking (talk to recent hires, find out pain points) and iterate solutions quickly.
This not only yields improvements in that area but serves as a proof of concept for agile ways of working in HR. It also helps identify what roles or skills you might be missing (maybe you realize you need a dedicated HR tech analyst on such teams).
Start identifying the skill gaps between your current HR team and the team you’ll need for an AI-powered, product-centric model.
Do you have people who can serve as product managers?
Who can interpret data or refine AI agents?
Maybe some of your analytically inclined HRIS folks or forward-thinking HRBPs can be developed into these roles.
Invest in training – many HR organizations are upskilling their staff in analytics, AI basics, and agile methodology. Encourage HR team members to get curious and even participate in the AI pilot projects. You may also decide to hire a few new roles, like an HR AI lead or People Analytics expert, to kickstart the competency build.
Importantly, communicate to the team that AI is a tool to free them from drudgery, not a threat to their jobs – frame it as an opportunity to elevate their roles (because it truly can be).
Even in pilot stage, put governance on the agenda. Establish an AI oversight group (even if it’s just 2-3 people initially) that reviews how the AI is performing and addresses any issues (e.g., if the bot gives a wrong answer, the team fixes the content or rules).
Develop a draft of AI ethics guidelines for HR, perhaps adapting existing company AI principles if you have them.
Also, consider how you’ll measure success: define a set of experience metrics (employee satisfaction with HR services, NPS, etc.) and efficiency metrics (transaction cost, case volumes) that you will track as changes are implemented.
This prepares you to demonstrate the impact of your new model on both fronts.
Use the lessons from pilots and prototypes to refine your approach.
Maybe the chatbot pilot reveals employees ask a lot of nuanced questions about careers, so you might next focus on building an AI agent to guide career development.
Or your agile onboarding team’s success could be replicated for “performance management redesign team” next.
Create a roadmap for rolling out AI and product-team practices to more areas over time. This could span multiple years – that’s okay.
The important part is to keep moving forward in increments rather than assuming a big-bang reorg is the only way. As Volker Jacobs emphasizes, you don’t have to overhaul everything at once; you can start where it makes sense and gradually transition.
As with any transformation, bring others along through transparent communication. Explain to both HR staff and employees what changes to expect.
If you’re introducing an AI agent, explain its purpose, how to use it, and how it benefits the user (e.g., “ChatHR can get you answers in seconds, and if it can’t, it will seamlessly hand off to our team – it’s here to help you better!”).
Gather employee input. Perhaps involve an employee focus group to test new tools or to suggest improvements. And celebrate wins: if your new model cuts response times in half or raises your HR service NPS, share that success. It will build confidence in the journey.
In following these steps, organizations can steadily evolve their HR operating model. The end state might look different for each company – some may lean more into EX-driven models, others into data/AI-driven “machine-powered” models, or a hybrid.
The key is consciously choosing a path that aligns with your business needs and workforce culture, rather than passively clinging to a model designed for a past era. Many large enterprises may even adopt a mixed model – for instance, a stable part of the business stays on an Ulrich+ model while a fast-growing division uses an agile, AI-heavy model suited to its needs (McKinsey).
Throughout this evolution, keep the ultimate goal front and center – as one expert succinctly put it, “the goal is not to use AI, the goal is to improve the experience of people in their daily work”.
Technology and models are means to an end. By evolving from the traditional frameworks of the past toward AI-enabled, employee-first models, HR has the opportunity to become more personal, more proactive, and more impactful than ever before.
It’s an exciting time to lead HR – the old constraints are falling away, and new possibilities are emerging.
By combining the best of human empathy and judgment with the efficiency and intelligence of AI, HR can truly reinvent itself as a driver of both exceptional employee experience and business value.
The journey from Ulrich to AI-driven HR is underway – and those who embrace it will define the future of work and service in their organizations.
How Applaud Helps You Make It Happen
At Applaud, we believe employees are a company’s most important customers. That’s why our technology is built entirely from the employee’s point of view—delivering more human, intuitive, and rewarding HR experiences that empower HR teams to do more for their people.
If you’re ready to turn employee-first HR from vision to reality, we’re here to help. Get in touch to see how Applaud can transform your HR Service Delivery and create a workplace where employees truly thrive.
Duncan Casemore is Co-Founder and CTO of Applaud, an award-winning HR platform built entirely around employees. Formerly at Oracle and a global HR consultant, Duncan is known for championing more human, intuitive HR tech. Regularly featured in top publications, he collaborates with thought leaders like Josh Bersin, speaks at major events, and continues to help organizations create truly people-first workplaces.