🤖 AI Prompt Pack - Human Resources KPIs
Prompt topics for novel Human Resources KPIs
Human Resources KPIs are stale. Almost every website and textbook that you read has the same KPIs that focus on the performance of individual HR team members, or numbers relating to a topic (e.g. how many people under 21 years of age work for us).
Here’s an example from one of the 28 websites linked above:
All pretty standard. Using measures like these is fine, and one benefit of using the same measures as everyone else is you can benchmark performance.
But that’s not really the main purpose of KPIs. We should be using KPIs to measure progress towards a strategic or operational objective. That’s where a conversation with AI can be helpful.
In this prompt pack, two topics that are focused on:
The future of employee experiences and the value of the ‘Employee Lifetime Value’ KPI in measuring them, and
Generating clever and high-impact ways of both better measuring and getting greater value from Human Resources technology systems and transformations.
In this prompt pack, you’ll learn:
How to setup your own prompts that are relevant to your context;
How to create a relevant synthetic persona to represent different stakeholders, and
How to use the personas to augment the quality and relevance of your AI interactions.
As usual, I’m not going to be prescriptive about what AI model you should use. Use what you have access to, what you’re comfortable with, and the model that has the features you need.
✏️ Fundamentals for writing great prompts
Give the AI a specific role and specific goal. For example, “You’re a Human Resources Business Partner with 10-years’ experience across multiple industries. Your goal is to find insights from process data.”
List step-by-step instructions, where necessary.
Share examples. For example, “An example of a previously successful business argument is …”
Ask for variations. For example, “That was a good effort but it hasn’t hit the mark. Please try again with more focus on …”. Feel free to have a conversation with the AI model.
Add constraints. For example, “provide a list of five KPIs”.
💡 You’re encouraged to play around with these prompts - your unique context will generate more suitable responses.
🗈 Note - Different AI models have different token sizes. The token size of an AI model represents the amount of information that it can consume, the amount of context it can hold, about a query. A rough rule-of-thumb is that 1,000 tokens equates to around 750 words. The larger the token size, the more context the AI model can handle.
💼 Case One – Generating Novel Strategic Human Resources KPIs
Prompt one:
Imagine a brainstorming session about the future of employee experiences and the value of the ‘Employee Lifetime Value’ KPI in measuring, valuing and assessing them.
This brainstorming session’s purpose is to come up with novel, clever and high-impact ways of both better measuring and getting greater value from ‘ELV’.
The participants for this imaginary ELV brainstorming session will be Human Resources guru Dave Ulrich professor of business at the Ross School of Business University of Michigan and co-founder of The RBL Group, famed Human Resources analyst and innovator Josh Bersin, branding innovator and copywriter Mary Wells Lawrence, internet innovator and AI pioneer Kai-Fu Lee, and openAI founder and entrepreneur Sam Altman.
Please present a transcript of their provocative, high-energy and productive exchange and conclude by presenting their three most profound ELV KPI insights and recommendations.
Example response one:
Absolutely! Here’s a transcript of the brainstorming session focusing on Employee Lifetime Value (ELV), followed by their top ELV insights and recommendations.
Setting: A bright, open conference room, with whiteboard walls and tech-enabled easels for collaboration.
[Scene Start]
The group enters the conference room buzzing with excitement.
Dave Ulrich: Alright team, let’s transform how we think about employee value! Employee Lifetime Value – it’s the cornerstone of the future HR landscape.
Josh Bersin: Absolutely, Dave. Traditional HR metrics are lagging. ELV provides a holistic, forward-thinking approach.
Mary Wells Lawrence: Exactly! It’s not just about output, it’s about the brand the employee embodies. How do they make a lasting impact that amplifies what the company stands for?
Kai-Fu Lee: Data is key! We need new AI-powered tools to measure sentiment, problem-solving patterns, and innovation potential that traditional tools miss.
Sam Altman: I couldn’t agree more. Can we gamify elements of ELV? Give employees a sense of ownership and progression within the metric?
Dave Ulrich: Excellent point, Sam! Now, remember, we’re aiming for truly novel insights here. Let’s push boundaries!
Josh Bersin: What if we extend ELV beyond the employee’s tenure? Measuring how they influence customers, alumni networks, and industry thought leadership…
Mary Wells Lawrence: Love it! Like the residual value a celebrity endorsement brings, there’s a halo effect from top employees. We need to quantify that.
Kai-Fu Lee: Could we build predictive models? If AI identifies high-ELV potential at hiring, imagine the impact on tailored development and long-term retention.
Sam Altman: Hold on, this raises an interesting point. What if high-ELV employees are more likely to become successful entrepreneurs? Is there a point where we encourage that rather than fight it?
Dave Ulrich: Now you’re thinking outside the box, Sam! It’s a paradigm shift: maybe high-ELV means the company is a springboard, not a final destination.
[Two hours of heated but collaborative discussion follows]
Dave Ulrich: Okay, time to distill our wisdom! What are the top three game-changers to revolutionize ELV?
Josh Bersin: Number one: ELV is not one-size-fits-all. Personalize it! Understand employee goals, and map their career journey contributions to ELV.
Mary Wells Lawrence: Number two: Brand alignment IS ELV. We need to quantify how employees live the mission. Storytelling, internal and external, around the impact they make.
Kai-Fu Lee: Number three: AI-powered early identification of high-potential employees. Proactive development becomes ELV investment, with greater returns and loyalty.
[Scene Close]
Top 3 ELV Insights and Recommendations
Personalized ELV Journeys: Recognize employees’ aspirations and connect the dots between their evolving roles and overall company value-creation.
The Power of Brand Storytelling: Quantify employee “brand ambassadorship” and the ripple effects stemming from their embodiment of the company’s values.
AI-Augmented Talent Spotting: Proactive ELV investment means leveraging AI to spot high-potential at hiring, leading to tailored development, higher return on investment, and increased retention.
Let me know if you’d like to brainstorm more ideas or dive deeper into any of these specific recommendations!
Prompt two:
Whilst a good first effort, this brainstorming is too superficial, dispassionate and disconnected from both the professional experiences and technical capabilities and skill sets of its participants.
Present a richer, more nuanced, more sophisticated transcript and set of recommendations. Do this for a leadership readership of top Boston Consulting Group clients.
Example response two:
Absolutely! Here’s a revamped version of the brainstorming session, designed to resonate with a BCG client audience by focusing on strategic implications, innovative methodologies, and the unique expertise of the participants:
Setting: A sleek, high-tech meeting space with panoramic views, interactive whiteboards, and subtle background music to promote focus.
[Scene Start]
Dave Ulrich: Colleagues, ELV is a game-changer, and we owe BCG clients a roadmap. The old ROI mindset is obsolete – it’s about long-term impact, agility, and the employee as asset.
Josh Bersin: Right, Dave. But most ELV attempts are still just rebranded HR metrics. Clients need tools to see employee value through the lens of business strategy.
Mary Wells Lawrence: I couldn’t agree more. We must capture the “intangibles”: brand resonance, problem-solving audacity… the stuff that fuels innovation and disrupts competitors.
Kai-Fu Lee: Data is our ally, but traditional analytics fall short. Imagine AI sentiment analysis revealing not just engagement, but pre-emptive indicators of entrepreneurial flight risk, or untapped creative potential.
Sam Altman: Disrupting from within is key. Can we borrow from venture capital models? An internal ‘ideas marketplace’ where high-ELV employees get seed funding and mentorship to pitch business solutions.
Dave Ulrich: Brilliant, Sam! ELV isn’t just about retaining talent, it’s about capitalizing on it in unforeseen ways, transforming the employee/employer relationship.
Josh Bersin: This raises a strategic dilemma: How to balance this with loyalty? Perhaps ELV-driven compensation models include equity-like vesting, even if the employee eventually spins off an idea?
Mary Wells Lawrence: Or the opposite! High-ELV alumni become brand ambassadors. Imagine a network effect: the impact on recruitment, on attracting like-minded talent…
Kai-Fu Lee: We need predictive models far beyond hiring. AI mapping of collaboration patterns can reveal hidden innovation hubs, cross-functional synergies that boost collective ELV.
Sam Altman: But beware of perverse incentives. If ELV is too tightly quantified, we may stifle the risk-taking that made the company successful in the first place.
Dave Ulrich: A critical point. This requires a new leadership mindset. CEOs must champion a culture where ‘smart failures’ with high potential payoff are factored into the ELV equation.
[Fast-paced whiteboard session ensues, diagrams and keywords flashing to life]
Dave Ulrich: Let’s synthesize! Clients need these takeaways:
BCG Recommendations for Reframing ELV:
Strategic Lens on Talent: Shift from cost-center thinking to ELV as a competitive advantage. Integrate employees’ career trajectories into long-term business model evolution.
Embrace Ambiguity, Measure the Unseen: Leverage AI and qualitative insight tools to mine the hidden value drivers: maverick thinking, cultural influence, and ‘x-factor’ innovation sparks.
ELV as Ecosystem, not Endpoint: Build robust alumni networks, track their success. Factor spin-offs into the talent lifecycle and recognize the long-term marketing value of a high-ELV reputation.
[Scene Close]
Let me know if you’d like to explore a particular recommendation in more detail or add new complexities to this scenario!
Prompt three:
Now present a transcript where they disagree about the role, purpose and value of ELV and summarise the two key elements and aspects of that disagreement.
Example response three:
Absolutely! Here’s a transcript highlighting a fundamental disagreement about the nature of ELV, aimed at a BCG leadership audience:
Setting: The meeting space is sleek, but there is a palpable tension in the air. The tech aids seem superfluous as the group is focused solely on each other.
[Scene Start]
Dave Ulrich: Colleagues, ELV is the metric of the future. It’s our key to maximizing human capital investment and securing competitive advantage.
Mary Wells Lawrence: Dave, I admire your enthusiasm, but let’s be blunt. ELV risks turning people into commodities. It misses the soul of what makes a company truly tick.
Josh Bersin: Mary, I sympathize, but we can’t ignore market realities. We need data-driven tools to align HR with business outcomes. ELV is a way to make that happen.
Kai-Fu Lee: Perhaps, but the current tools are too crude. How do we quantify the ‘spark’ of creative collaboration? AI models can try, but miss the nuances of human brilliance.
Sam Altman: I think the fundamental problem is how ELV is used. If it becomes a rigid metric for performance reviews, resentment will build. It needs flexibility, an open-ended ‘what if’ factor.
Dave Ulrich: So you’re focused on the motivational pitfalls, Sam. That’s valid, but with careful design, ELV can become aspirational, not punitive.
Mary Wells Lawrence: But even the word ‘value’ turns me off! We’re not talking about widgets. Companies thrive on intangible loyalties, gut-instinct brilliance that no metric can capture.
Kai-Fu Lee: There’s a potential middle-ground. We use AI not for prescriptive ELV scoring, but to spot emergent patterns in high-potential teams. Then we use qualitative methods to understand why they succeed.
Josh Bersin: Hybrid approach – I like it! ELV as a starting point for strategic inquiry, not an end-all answer. That’s something BCG clients can wrap their heads around.
Dave Ulrich: We may have a philosophical split here. Mary, what’s your alternative to ELV if we abandon it altogether?
Mary Wells Lawrence: A renewed focus on the employee experience – the journey, not just a number. A culture where risk-taking is celebrated as much as immediate wins, and leaders are masters of reading the room, not just a dashboard.
[The discussion continues passionately, but with respect]
Scene Close
Key Disagreement in the ELV Debate:
ELV as Measurement vs. Motivation: Does ELV reduce employees to quantifiable assets, killing the elusive spirit of innovation? Or is it a necessary tool to track and cultivate talent while balancing emotional drivers for long-term success?
AI’s Role: Savior or Underminer? Can AI truly capture the nuances of human interaction, the hidden spark that drives exceptional contributions? Or should it be limited to identifying patterns, leaving qualitative insights to guide the full understanding of ELV potential?
Let me know if you want to delve into the implications of this disagreement for BCG leaders or explore potential resolutions!
💼 Case Two – Generating Novel Human Resources Technology KPIs
Prompt one:
Imagine a brainstorming session about the future of Human Resources technology in large multi-national organisations and the value of both integrated business processes and the employee experience in measuring, valuing and assessing them.
This brainstorming session’s purpose is to come up with novel, clever and high-impact ways of both better measuring and getting greater value from Human Resources technology systems and transformations.
The participants for this imaginary brainstorming session will be Human Resources guru Dave Ulrich professor of business at the Ross School of Business University of Michigan and co-founder of The RBL Group, famed Human Resources analyst and innovator Josh Bersin, branding innovator and copywriter Mary Wells Lawrence, internet innovator and AI pioneer Kai-Fu Lee, and openAI founder and entrepreneur Sam Altman.
Please present a transcript of their provocative, high-energy and productive exchange and conclude by presenting their three most profound insights and recommendations.
Example response one:
Absolutely! Here’s a transcript focused on next-generation HR technology for large multinationals, followed by key insights tailored for impactful change:
Setting: A bright, expansive room. Interactive whiteboards are coupled with multiple screens displaying real-time data visualizations and global company dashboards.
[Scene Start]
Dave Ulrich: Colleagues, HR tech has the potential to transform multinationals, but too often, it’s fragmented, failing to deliver on strategic goals. We need a major rethink!
Josh Bersin: Agreed, Dave. Legacy systems are clunky, compliance-driven…it’s time for tech to focus on empowering the employee experience, not just administrative tasks.
Mary Wells Lawrence: User experience is EVERYTHING! If employees find the tech intimidating, engagement suffers, and we lose out on valuable insights. It needs to feel like the consumer tech they use daily.
Kai-Fu Lee: Data integration is paramount. Global companies have treasure troves of information scattered across systems. Imagine AI unlocking patterns around talent mobility, or predicting skill gaps before they become crises.
Sam Altman: What if HR tech becomes more proactive? Like a recommendation engine, suggesting upskilling opportunities aligned to employees’ goals, creating a sense of self-guided development.
Dave Ulrich: That’s powerful, Sam. But multinationals have unique challenges. How do we ensure tech scales globally while being responsive to local regulations and cultural norms?
Josh Bersin: A modular approach could work. A core platform with adaptable modules for different regions. Local HR teams have a say in customization for their context.
Mary Wells Lawrence: And don’t forget the power of storytelling! Imagine internal dashboards that highlight how employees’ development journeys align with the company’s mission, across borders. That builds unity.
Kai-Fu Lee: Can we use sentiment analysis to detect early signs of disengagement? AI identifying not just attrition risk, but where employees feel unheard, even with different cultural communication styles.
Sam Altman: This raises an ethical question: Does constant monitoring invade privacy? We need transparency, and the ability for employees to partially opt-out if desired.
Dave Ulrich: Excellent point. Now, let’s identify the big-picture takeaways that truly move the needle for multinationals:
Top 3 Next-Gen HR Tech Insights
Employee-Centric Design: HR tech MUST prioritize an intuitive, engaging user interface. Adopt design-thinking principles to tailor systems to the employee journey, not just HR bureaucracy.
Integrated Intelligence: Break down data silos across functions and geographies. Harness AI for predictive analytics on talent flows, emerging skill needs, and the impact of HR initiatives on business outcomes.
Balancing Customization and Control: Enable a flexible ‘core and modules’ architecture. This empowers local HR to tailor solutions while maintaining global data integrity and compliance.
[Scene Close]
Let me know if you’d like me to add more complexity, like potential roadblocks to implementation, or focus on strategies for winning over senior leadership to invest in such HR tech transformations!
Prompt two:
Whilst a good first effort, this brainstorming is too superficial, dispassionate and disconnected from both the professional experiences and technical capabilities and skill sets of its participants.
Present a richer, more nuanced, more sophisticated transcript and set of recommendations. Do this for a leadership readership of top Boston Consulting Group clients.
Example response two:
Absolutely! Here’s a revamped version of the brainstorming session, designed to highlight the strategic implications for multinationals, leveraging the unique insights of the participants, and tailored for a BCG client audience:
Setting: A modern, expansive workspace with sleek furniture and seamless tech integration. Screens display real-time visualizations of global workforce data and market trends.
[Scene Start]
Dave Ulrich: Colleagues, HR tech transformation within multinationals is a make-or-break proposition. Legacy systems hinder agility. We need insights that resonate with BCG clients.
Josh Bersin: Right, Dave. Clients are tired of vendors overpromising. We need to map tech directly to bottlenecks in talent management, especially on a global scale.
Mary Wells Lawrence: Messaging matters. This isn’t about tools, it’s about unlocking the brand potential trapped within a fragmented multinational workforce.
Kai-Fu Lee: Agreed. Data silos are immense. Imagine AI mapping internal career paths. Hidden talent pools, succession planning… BCG clients need to see the ROI in that.
Sam Altman: We undervalue the network effect. Can HR tech facilitate cross-border mentorship, skill exchange programs? This breaks down silos faster than any org chart can.
Dave Ulrich: You’re hitting on cultural transformation, Sam. But this raises the issue of compliance. GDPR, regional data laws…how do we build in flexibility?
Kai-Fu Lee: It’s not just legal compliance, but cultural sensitivity. AI sentiment analysis in one region may be met with suspicion in another. We need adaptable algorithms.
Josh Bersin: This goes beyond HR – partnership with IT and legal is crucial. A modular, highly secure platform is non-negotiable. BCG clients need that reassurance.
Mary Wells Lawrence: Imagine the case studies! HR tech enabling a multinational to launch in a new market at lightning speed due to predictive talent analytics. That’s the story we need to tell.
Sam Altman: We can’t forget the ethics question. Transparency is key. Allow employees to see the data their work generates, give them agency in how it’s used.
Dave Ulrich: Now, let’s deliver takeaways with that strategic edge for BCG clients:
BCG Recommendations: Next-Gen HR Tech for Multinationals
Talent Agility as Competitive Advantage: Map current tech to critical hurdles in global talent mobility, succession planning, and rapid skill-gap response. Frame the investment as agility insurance in a volatile market.
Data-Driven, Culturally Aware Transformation: Emphasize AI solutions with adaptable algorithms and modular design for compliance across regions. Balance data insights with respect for cultural nuances in communication and perception of technology.
Brand Amplification, Not Bureaucracy: Quantify the impact of HR tech on internal brand ambassadorship, seamless cross-border collaboration, and faster localization time for new market entry.
[Scene Close]
Let me know if you’d like to explore a particular challenge in more depth or focus on strategies for overcoming resistance to this level of HR tech investment!
Prompt three:
Now present a transcript where they disagree about the role, purpose and value of Next-Gen HR Tech for Multinationals and summarise the two key elements and aspects of that disagreement.
Example response three:
Of course! Here’s a transcript highlighting a heated debate on the true value of next-gen HR tech, specifically designed to generate rich insights for BCG leadership:
Setting: The expansive workspace hums with a subtle tension. Data visualizations flicker, highlighting global disparities.
[Scene Start]
Dave Ulrich: Colleagues, HR tech can’t just be a shiny veneer. We must prove it generates strategic value for multinationals, or clients won’t invest.
Josh Bersin: I agree, Dave, but let’s not reduce it to ROI spreadsheets. Next-gen tech transforms the employee experience, which has long-term brand impact.
Kai-Fu Lee: Experience is crucial, but AI is the game-changer. Unlocking hidden talent pools, predicting market-sensitive skill gaps…that’s what CEOs care about.
Mary Wells Lawrence: Hold on. AI is cold. How do we ensure tech actually fosters the human connections that make multinationals work despite geographic and cultural divides?
Sam Altman: Mary, you raise a critical point. There’s a risk of over-reliance on tech. Could it stifle organic innovation, the chance encounters that spark brilliance across borders?
Dave Ulrich: Sam, that’s valid, but what’s the alternative? Relying on inefficient, gut-instinct hiring and promotion, especially across a global workforce?
Josh Bersin: It’s about finding the balance. Tech should augment human judgment, not replace it. HR leaders need skills to interpret data and tell the story behind the numbers.
Kai-Fu Lee: But even with the best interpreters, how do we tackle the bias in AI itself? Data reflects our existing workforces, potentially perpetuating inequities.
Mary Wells Lawrence: Precisely! Focusing on skills prediction may blind us to the need for diversity in thought. Tech must incentivize scouting for atypical talent, even if initial success markers are unclear.
Sam Altman: We’ve hit a deadlock. Perhaps the real value of tech is in revealing these very tensions, forcing strategic discussions that wouldn’t happen otherwise.
Dave Ulrich: A messy, but necessary process! Let’s outline this for BCG clients:
Key Disputes: Next-Gen HR tech in Multinationals
Human-Centric vs. Data-Driven Transformation: Does tech primarily empower employees, boosting engagement and fostering connection, or does it prioritize optimizing talent flows and predictive analytics for leadership-level decision-making?
AI Bias vs. the Diversity Imperative: Can tech truly overcome its inherent biases? How to balance the need for efficiency with the long-term imperative to cultivate diverse thinking styles and backgrounds, potentially invisible to traditional analytics models?
[Scene Close]
Let me know if you’d like to explore potential solutions to bridge this divide, or dive deeper into the implications of each side of the argument!
📈 Level up
These prompts are designed to work across the variety of commercial GPTs without any specific configuration, and without needing to host and run your own models.
In a business setting, Simon recommends to ‘train’ your AI model on your relevant corporate data. If security is a concern then only use public data.
Imagine using your favourite GPT model, say OpenAI’s ChatGPT-4, but have it trained on public information about your company, and any relevant documents that you have from your company. This is not fantasy, this is available now and for relatively little effort.
How to do this?
Get started with AnythingLLM. AnythingLLM is a self-hosted AI chat bot that you can run on your personal desktop (Mac, Windows, and Linux) or in the cloud from source, in Docker, or via a subscription if you don’t have your own servers. It’s fully private, supports custom models, and can be ‘trained’ on documents or websites that you provide.
Choose the LLM that suits your context the best. You can use many of the popular commercial models in AnythingLLM including OpenAI, Microsoft Azure, Anthropic AI, Google Gemini etc. You can also download and install LocalAI to run open-source models from Hugging Face with AnythingLLM (although this does require a fair bit more disk space, and potentially GPU infrastructure).
Make sure that embedding selection suits your use case. This is the ‘context’ part where you embed documents and URLs in the AI model to provide better and more accurate context. AnythingLLM’s default ensures that a local open-source model is used so your data never leaves your instance, however models are available for OpenAI and Microsoft Azure, and for LocalAI if you want to run your own model.
AnythingLLM is shipped with a local vector database running on LanceDB. This will be fine for most applications, however you can use a number of other popular vector databases running alongside your AnythingLLM instance, or on other servers. For reference, a vector database is a collection of data stored as mathematical representations. Vector databases make it easier for machine learning models to remember previous inputs, allowing machine learning to be used to power search, recommendations, and text generation use-cases. Data can be identified based on similarity metrics instead of exact matches, making it possible for a computer model to understand data contextually.
Point AnythingLLM to your corporate website and upload documents relevant to the context that it’s operating in. You can have multiple workspaces if you need multiple contexts. With AnythingLLM having access to information about your company, its responses will be much more suitable. For example, if you’re creating a “talent acquisition” workspace you can enter the URL for your public corporate Employee Value Proposition page, and upload relevant documents such as corporate plans, annual reports, strategies, and recruitment-related policies. All queries via AnythingLLM will obtain responses with this corporate context.
Still skeptical about how simple this is?
Watch the video below where Tim Carambat, founder of AnythingLLM, sets up an AythingLLM instance in the cloud, integrates with OpenAI (ChatGPT), trains the instance on data from the public AnythingLLM website, and embeds a chatbot on a website, all in a matter of minutes with no code at all.
If you don’t need to run AnythingLLM on a server (e.g. you don’t need a chatbot embedded in a website) then download AnythingLLM to your computer and watch up to the 4:48 mark. Yep, a custom, contextual, local AI in less than five minutes.