• One new stream of this data has been the adoption of new worker monitoring tools. Between the beginning of the pandemic and late 2022, around one in three medium and large companies had adopted new tools for monitoring their employees
• their organization leverages data and analytics to make talent decisions
• Between the beginning of the COVID- 1 9 pandemic and late 2022, approximately one third of medium and large companies surveyed adopted new worker monitoring tools that analyze key strokes, mouse activity, and more to determine how much a person is working and on what—a rate of adoption that outpaces even the steepest part of the curve of smartphone adoption in the U.S.
• Passive data is generated and collected without the direct input, and sometimes even awareness, of an individual. For example, location information is passively generated when an individual keeps their phone with them while running errands
• A quantified organization takes a strategic approach to measuring what it should, not just what it can. It takes a responsible approach to using new data sources and AI tools to create value for stakeholders across the organization, improving workforce trust and driving the organization forward to new levels of financial, reputational, and operational performance.
• The growth in passive data from common workplace applications like email, collaboration and social tools, and shared calendars may already be outpacing many organizations’ capacities to process and use the data, resulting in too much data and too little insight.
• We think that there is much more insight and value in unstructured data than there is in structured data.
• Biased algorithms that scale the wrong decisions, risks to privacy, worker rights violations—poorly planned approaches to quantification have the potential to undermine trust and fairness
• The growth in passive workforce data—combined with other sources of information, analytics, and AI— is surfacing new opportunities to help create shared value at multiple levels of an organization: the individual worker; the team, group, or function; the enterprise; and greater society.
• When an organization uses the data they collect about their workforce to benefit everyone— individual workers, teams and groups, the organization, and society as a whole—we call that creating shared value
• Identify the activities, roles, actions, and workers that create the most value Use data from workers and managers— with their consent—to identify the portfolio of activities, people, roles, or skill sets that determine 80% of effective outcomes
• Implement fact-based performance reviews Quantify performance based on data from work applications or products to inform a manager’s assessment of performance, leading to more objective and less biased evaluations
• Unlock collaboration across boundaries Use internal social network analysis to help determine presence of crossfunctional leadership teams
• Improve physical well-being Use wearables, sensors in the environment,
• Use emotional data to help guide interventions Deploy emotion detecting software that mines biometrics
• Access diverse and hidden talent Identify hidden talents during recruitment and hire based on skills, cultural fit, or team fit instead
• These opportunities for shared value creation can cut across multiple dimensions. Some aim to use data on work and the workforce to improve outcomes like agility, speed, quality, or customer satisfaction while also improving the worker experience. Others, such as redesigning work processes, can point toward opportunities to develop evidencebased approaches to creating successful human—and machine—work teams
• Simulate work processes or human/machine interactions Collect data on how workers interact with others and/or machines to create a digital twin of the workforce to simulate work redesigns and their impact.
• Identify bottlenecks and supplement formal processes with informal conversations Use organizational network data to help identify employees who are bottlenecks or spots where more interaction can improve a process, adjusting communication patterns accordingly
• Understand hidden work patterns Use process mining of enterprise transaction systems, workflow data, or video analytics to help identify root causes of issues, the tasks underlying each job, and the strongest opportunities for improvements.
• Optimize team composition Mine data on skills, work styles, values, collaborative abilities, and more to optimize team composition and predict team performance based on the impact of each newly added member .
• Enhance team performance Mine data from team collaboration and project management sites to determine factors like interpersonal relationships that require special attention, points of conflict, and project decision makers using AI as a coach to suggest improvements.
• Make meetings more effective Use audio and video analytics to automatically summarize meetings and surface relevant information realtime.
• Improve emotional and relational intelligence Use organizational network data to reveal relationship patterns and interactions
• Personalize communications Use data from communication applications to suggest the optimal way to communicate with others.
• Improve agility Use data from embedded sensors in plants that track workers and equipment to create self-adapting, non-linear assembly lines or work orders.
• Capture continuous improvement Evaluate the digital trail of people’s routines and actions to capture the details of how workers creatively try new approaches to continuous improvement.
• Reduce time to market Mine communications or work application data to spot and improv
• Increase inclusion Use calendar and communications data to measure indicators of inclusion, including formal and informal interactions with senior leadership
• Understand how attitudes and behaviors spread Based on the strength, diversity, and number of connections with others as determined through organizational network data, identify influencers who can drive greater DEI and Belonging efforts,
• Spot and mitigate implicit bias Mine text in work, communications, and performance feedback platforms to identify biases, gaps in inclusion efforts, and toxic pockets of conversation that undermine broader DEI and Belonging efforts.
• Spotlight: DEI beyond representation In addition to mining collaboration, communication, and voice data to identify bias or microaggressions, data and technology can also provide insights into inclusion based on people’s interactions with leaders, mentors, and resource or affinity groups.
• “If you never go to a social event, never go to a Christmas party, never go to a coffee talk, lunch-and-learn thing and your performance is poor, is that a direct reflection or is it something else? We have that data, but I’m really trying to identify answers to questions like: Are managers doing the right thing? Are they discriminatory? Do they bring in the right development to the right staff? Are we supporting our high performers, or are they just high performers regardless of what you do for them?
• By aggregating data across the enterprise, organizations can gain new insights on how practices, behaviors, and decisions may impact the organization. New tools are going beyond a qualitative and imprecise understanding of culture and can now quantify how subtle elements of culture—such as the use of different terms and phrases across the enterprise—highlight potential issues and opportunities for improvement.
• Determine “as is” state Use process mining of enterprise transaction systems to start with an “as is” picture of the way people perform work through a process, identifying root causes of issues and opportunities for improvements.
• Match people to work Human decision-makers can use AI as input to analyze a change project and map the skills required at each stage based on learning from previous projects.
• Create a heat map to localize solutions Take an employee-centric, end-to-end view of change to understand how each employee is being asked to change across multiple change initiatives. Adjust accordingly to prevent change fatigue.
• Identify change agents Use organization network data from work applications to help identify influencers
• Analyze culture Use natural language processing, text analysis, computational linguistics, audio analysis of business meetings, and biometrics to identify and quantify potential declines in sentiment across the workforce or identify microcultures.
• Spot potential deals Use automated tools to scout for potential acquisition or merger opportunities, using data from public social media sites that reveal data on engagement and culture; data on leaders or founders that is predictive of results or indicative of cultural fit;
• Drive culture change Use organizational network analysis to identify barriers to cultural change as well as target opportunities to spread new values and practices
• Integrate acquisitions Use organizational network analysis to spot and solve for organizational silos and bottlenecks.
• How can organizations chart a course toward responsible use of workforce data and technology that creates trust, while navigating trade-offs between risk and opportunity?
• 1. They’re collecting data but are not sure how to use it, effectively gathering data for the sake of having it. 2. They’re gathering and using it without a plan, governance, or responsibility practices in place (beyond meeting regulatory requirements), setting themselves up for significant risk and potential damage to workforce trust. 3. They’re underinvesting in technologies that gather or use workforce data, either paralyzed by concerns of worker backlash or unknowledgeable about how to manage data responsibly, leaving a great deal of value on the table.
• with significant variations across the globe, the regulatory landscape can be challenging for multi-nationals to navigate.
• Moving at the speed of trust Trust can be just as important as growth, profitability, and other key performance indicators on the organization’s balance sheet. It may not be enough to consider “what” a company does; it can also be equally important to consider “how” it does it.
• 79% of employees feel motivated to work for a trusted employer .
• The quantified organization journey is not without its risks, however, and it is critical to work with legal and human resources teams to manage them. Collecting the wrong data, failing to link analytic efforts to intended outcomes, and alienating employees in the process can erode worker trust and derail the best efforts. Advances in technology will likely continue to pose new questions and create additional challenges for quantified organizations to conside
• Many of these challenges—such as questions around worker privacy— may be well-known but could intensify as technical advances allow enterprises to infer detailed, highly personal information

48 more annotations...

• Il y a une mauvaise communication autour de l'IA générative, que beaucoup de gens présentent comme supérieure à d'autres techniques. Ce n’est qu'une autre technique de traitement de la modélisation et de l'inférence. La grande majorité des décisions prises par les entreprises aujourd'hui ne sont pas basées sur l'IA générative, qui affiche encore de grosses lacunes en raisonnement quantitatif.
• En fait, la démarche devrait être de résoudre le problème par le biais d’un processus, puis d'appliquer la technologie à ce processus.
• L'IA générative n'est que la dernière étape pour renvoyer cette information à l'utilisateur.
• Oui, l’utilisation de l’IA générative doit davantage être une expérience conversationnelle, plutôt que de tenter d'exploiter à l'extrême les capacités d'inférence des grands modèles de langage
• d'utiliser SAS Viya pour le raisonnement quantitatif, ou même le raisonnement en général, puis de le restituer d'une manière qui permette au grand modèle de langage de répondre à l'utilisateu
• L'IA ne pense pas, il n'y a pas de réflexion en IA. En fait, tout en IA est basé sur la prédiction de probabilité. Si vous entraînez suffisamment de données et que vous lui posez ensuite une nouvelle question sur des données similaires, elle vous donnera la meilleure probabilité de ce qu'elle pense être la réponse

4 more annotations...

• • uerying information collections, which puts a GenAI front-end on to structured repositories of important content
   
  • AI-powered chatbots, which use AI to deliver a chatbot with enough ‘smarts’ to answer many common employee questions
   
  • generating suggestions, where the AI puts forward links to potentially useful content ‘in the flow’ of an employee’s task
   
  • digital workplace information retrieval, perhaps the most challenging AI use as it requires good governance of a very broad spread of information
• The immediate challenge is that if the content is out of date, incorrect or poorly structured,
• The longer-term challenge is that most AI-based solutions will be delivered as a project, without necessarily having an ongoing business-as-usual component. The content underpinning the AI may have many owners, and almost certainly won’t be owned by the business area that built the AI solution

2 more annotations...

• This year, the survey finds that adoption has jumped to 72 percent (Exhibit 1)
• Also, responses suggest that companies are now using AI in more parts of the business.
• . Yet across functions, only two use cases, both within marketing and sales, are reported by 15 percent or more of respondents.
• espondents are much more likely to be using gen AI at work and even more likely to be using gen AI both at work and in their personal lives
• Yet in most industries, larger shares of respondents report that their organizations spend more than 20 percent on analytical AI than on gen AI. Looking ahead, most respondents—67 percent—expect their organizations to invest more in AI over the next three years.
• The function in which the largest share of respondents report seeing cost decreases is human resources. Respondents most commonly report meaningful revenue increases (of more than 5 percent) in supply chain and inventory management (Exhibit 6)
• For analytical AI, respondents most often report seeing cost benefits in service operations—
• they’re also recognizing the diverse risks associated with the technology. These can range from data management risks such as data privacy, bias, or intellectual property (IP) infringement to model management risks, which tend to focus on inaccurate output or lack of explainability. A third big risk category is security and incorrect use.

• Respondents most often report inaccuracy as a risk that has affected their organizations, followed by cybersecurity and explainability.

   
• just 18 percent say their organizations have an enterprise-wide council or board with the authority to make decisions involving responsible AI governance, and only one-third say gen AI risk awareness and risk mitigation controls are required skill sets for technical talent.
• We have found three archetypes for implementing gen AI solutions: takers use off-the-shelf, publicly available solutions; shapers customize those tools with proprietary data and systems; and makers develop their own foundation models from scratch
• their organizations required one to four months from the start of a project to put gen AI into production, though the time it takes varies by business function
• Gen AI is a new technology, and organizations are still early in the journey of pursuing its opportunities and scaling it across functions. So it’s little surprise that only a small subset of respondents (46 out of 876) report that a meaningful share of their organizations’ EBIT can be attributed to their deployment of gen AI.
• gen AI high performers are less likely to use those off-the-shelf options than to either implement significantly customized versions of those tools or to develop their own proprietary foundation models.
• For one thing, they are paying more attention to gen-AI-related risks. Perhaps because they are further along on their journeys, they are more likely than others to say their organizations have experienced every negative consequence from gen AI
• Gen AI high performers are also much more likely to say their organizations follow a set of risk-related best practices (Exhibit 11). For example, they are nearly twice as likely as others to involve the legal function and embed risk reviews early on in the development of gen AI solutions—
• Seventy percent say they have experienced difficulties with data, including defining processes for data governance, developing the ability to quickly integrate data into AI models, and an insufficient amount of training data, highlighting the essential role that data play in capturing value

16 more annotations...

• In many organizations, HR functions are divided into distinct areas like recruitment, benefits, payroll and employee relations. This traditional North-South siloed approach means each functional area deploys its own strategy and tools to achieve specific goals
• it often leads to a disjointed experience for employees
• East-West thinking encourages collaboration across these functional silos within HR. This approach emphasizes creating a cohesive employee experience by integrating strategies and tools across all HR functions. For example, when an employee has a query, an East-West-aligned HR function ensures that whether the question pertains to payroll, benefits or career development, the response is seamless and consistent.
• In the context of chatbot implementation, this might mean developing highly specialized bots for different functions without considering the holistic employee experience.
• Generative AI plays a pivotal role here. By designing chatbots that can understand and respond to natural language queries, we can create an interaction experience that feels more human and less robotic.
• First, it allows for more personalized interactions. A generative AI-powered bot can remember past interactions, anticipate future needs,
• generative AI enables the creation of more sophisticated and context-aware responses.
• Good data ensures that chatbots can provide accurate, relevant and timely information. It also enables these bots to learn and improve over time
• the old-school intranets and portals—often referred to as link farms—are becoming obsolete.
• This new approach prioritizes the needs and experiences of employees, ensuring that technology serves as an enabler rather than a barrier.
• An integrated experience architecture involves breaking down both east-west and north-south silos. Cross-functional collaboration is key. HR, IT and other departments must work together to ensure that chatbot solutions are aligned and interconnected

9 more annotations...

• work was fundamentally broken and to stay ahead of curve small changes won’t be sufficient.
• Slack found that AI use by desk workers is up 23% since January and 60% since September, and they’re reaping the rewards.

• Their work-life balance was 18% higher, they were 29% more likely to feel highly passionate about work and were 23% better at managing stress.

   

• Ultimately, they were 24% more satisfied at work.

   
• two in three workers have still not used AI at work.

• This is because of trust issues – only 7% see AI as being completely trustworthy for work – and because only 15% think they’ve had enough education and training to confidently use AI effectively at work.

   
• 37% of desk workers told Slack that their employer has no AI policy, but when workers understand how and when to use AI, they are 6x more likely to have experimented with AI at work.
• work was fundamentally broken and to stay ahead of curve small changes won’t be sufficient.
• desk workers estimate they spend a third of their day on low value work,
• The ideal situation is that AI minimizes these busy work tasks, and that enables people to focus on more meaningful work,
• hen Slack asked workers what they would prioritize with the time they got back from AI, ‘more admin’ topped the list, while innovation, learning and skills fell to the bottom.

9 more annotations...