A true revolution is here, not just in rewards but in economies and societies.
AI, machine learning, deep learning, natural language recognition and unstructured data processing have seen such a leap in quality and processing capability in the past two or three years, that a myriad of possibilities so far unheard of are being opened – and more is to come in the very near future.
All these advances come with inherent risks and even ethical concerns, which means regulation is necessary. However, rewards practitioners now have a radically new set of tools and frameworks, such as ChatGPT, to utilize.
In that vein, Zereon Associates decided to put ChatGPT to the test to see if we can generate truly relevant pieces for reward practitioners and their companies. To do so, we asked a rewards question to ChatGPT expressed in the form of a prompt and received the corresponding answer in two or three seconds.
Compensation Question to ChatGPT
“What are the most important requirements of an effective fixed compensation set-up for the employees of a multinational company, from the perspective of a global headquarter design, implementation, and administration? Answer should fit into a one A4-sized page.”
“An effective fixed compensation set-up for employees of a multinational company plays a crucial role in attracting, retaining, and motivating employees. The following are the most important requirements of a fixed compensation set-up from a global headquarter perspective:
- Fairness and Equity: Fixed compensation should be fair and equitable across all locations and levels. It should align with the company's business strategy and market competitiveness, to ensure that the employees are appropriately rewarded for their contributions.
- Consistency: The fixed compensation structure should be consistent with the company's policies and procedures. It should provide clear guidance to managers and employees on the payout process, eligibility criteria, performance expectations, and career advancement opportunities.
- Transparency: The compensation structure should be transparent and should be communicated effectively to employees. The employees should be aware of the compensation philosophy, the various components of fixed compensation, and how different elements of their compensation package are determined.
- Flexibility: The fixed compensation set-up should be flexible enough to accommodate varying business needs, budgets, and changing market dynamics. The company should regularly review, update, and adjust the compensation structure to ensure that it remains relevant.
- Compliance: Compliance with legal and regulatory requirements across all locations is critical. The company should ensure that the fixed compensation set-up is compliant with local laws and regulations, including any employment- and tax-related requirements.
- Alignment with Corporate Culture: The fixed compensation structure should align with the company's corporate culture, values, and mission. It should encourage teamwork, collaboration, and performance excellence.
- Performance-based: The fixed compensation set-up should be linked to individual and company performance. Employees should be incentivized to contribute towards the company's strategic objectives and should receive appropriate recognition and rewards for their efforts.
In summary, a successful fixed compensation set-up for multinational employees should be fair, transparent, flexible, compliant, aligned with the company's culture, and performance based. By fulfilling these requirements, the company can effectively attract, retain, and motivate its employees, which ultimately drives the company's success.”
Evaluating ChatGPT’s Performance
What do you think? In our eyes, quite good overall. We would perhaps be missing the following:
- Some comment on the need to achieve a balance of centralization vs. decentralization of decision making in multinational companies regarding role grading, market surveys, definition of salary structures themselves, and of pay grades and salary movements, across geographies and divisions.
- More reference on how fixed compensation specifically affects performance internationally vs. other total rewards elements — short- and long-term variable pay, recognition programs.
- In some parts of the text, “fixed compensation” is used somewhat interchangeably with “total rewards,” which is not completely accurate.
However, apart from sharing some international reward insights, the point of the exercise was to try the value of ChatGPT or similar tools and anticipate the possible value of future ones.
At this stage, we would qualify the performance of the tool regarding fixed compensation as the equivalent of a “clever total rewards junior consultant” or specialist, for “whom” we did not need to do a lengthy debrief, but just 5-10 seconds of instruction.
We are convinced that this virtual “junior specialist” may become, at least fixed-compensation qualitative knowledge-wise, the equivalent of a total rewards manager in one year or so. And a director in just one or two more, probably.
Quantitative knowledge progression is harder to say, possibly slower than the above.
This includes using grades and surveys, creating structures and salary ranges, managing salary increases, perhaps suggesting new ideas, all by inputting some simple prompts and numeric data — it will all depend on what we will previously feed into the tool’s open-source databases, so that the machine learns.
A Meta Discussion on the Experiment Results
In any case, if the part on true quantitative knowledge happens, it will mean effectively replacing several parts of the historical total rewards departments’ activities, and therefore a shift in the required competencies/skills of their professionals.
In this regard, and in our opinion, there will be knowledge and skills that will be more needed than ever — as opposed to the purely technical rewards subtleties, and as corresponds to a workforce that will be totally different in no time. They will include: asking effective questions/writing and evaluating effective prompts, strategic business and human capital thinking; business and financial acumen; the core concepts of compensation (i.e. grades, surveys, midpoints, variable pay payout scales, etc.); intercultural communications; holistic/systemic/ connecting-the-dots thinking; creativity; technology mindset; meta-analysis; and project and change management.
That is, we will perhaps see a slight return to “rewards generalism,” given that reward specialisms and technicalities will be done by the AI — what will matter is the general kind of human intelligence necessary to:
- Effectively generate ideas and direct the powerful “machine researcher” and then “machine designer” in the overall right direction
- Evaluate the results of such research and design,
- Apply human ingenuity to find new relationships and applications to the findings of the machine,
- Reintroduce the human-improved data into the AI to take it to new heights
- Apply the refined results onto the business and the people, to their advantage.
Contrary to what some other people may think, we do not need to wait 20 or 30 years for the future; it is here, and we are building and living it, right now.
During the certification courses we deliver in Central Europe, we extensively discuss these topics — to great intellectual delight and practical takeaways for both teachers and students.
Let’s all contribute to this bright rewards future.
Editor’s Note: Additional Content
For more information and resources related to this article see the pages below, which offer quick access to all WorldatWork content on these topics: