Impacting Organizational Capability - Data & Analytics

3.7.1.2.1 Increase Data-Fluent Expertise

Although organizations recognize the value that the data holds, many lack skilled staff who know how to collect, analyze, and interpret available data. Gartner's annual chief data officer (CDO) survey cited "poor data literacy" as the second highest barrier to organizational success (Richardson et al. 2017). Research also shows a 65 percent increase in the expected emphasis on technical skills in the workplace (Agarwal et al. 2018). As organizations become more data-driven, they will need employees to become experts in artificial intelligence, analytics, and advanced technology. Intelligent machines are an integral part of the analytics solution, because AI can help learners as they struggle with new skills. However, when on-the-job learning is replaced by "the introduction of sophisticated AI, analytics, and robots," this reduces learning opportunities (Beane 2019). TD professionals will need to resolve this. Another study found that 85 percent of TD leaders want to use big data and analytics to improve learning (Miller 2017). This means that talent development must step up to assert their desire for more accountability in this discipline by developing individuals who are proficient in a combination of both business skills and analytics. The fusion of analytical and business skills in one curriculum is one solution to ensure more employees have the expertise required (Chin et al. 2017). Talent development must lead the effort to build analytical capability.

To better grasp these processes and applications, TD professionals should understand the following related terminology:

Analytics. TD professionals should be familiar with four types of analytics:Descriptive analytics. A summary of historical data that provides the details of what happened.Diagnostic analytics. Examines the data to explain why something happened and may use data mining or data discovery.Predictive analytics. Any approach to data mining with four attributes:an emphasis on prediction (rather than description, classification, or clustering)rapid analysis measured in hours or days (not the stereotypical months of traditional data mining)an emphasis on the business relevance of the resulting insightsan emphasis on ease of use, thus making the tools accessible to business users.Prescriptive analytics. A form of advanced analytics that examines data or content to answer the question "What should be done?" or "What can we do to make X happen?" It is characterized by techniques such as graph analysis, simulations, complex event processing, recommendation engines, heuristics, and machine learning. Big data. Made up of data sets that are too large to capture and process using common methods for analysis, big data's significance is dependent upon the 5 Vs (volume, velocity, variety, veracity, and value). [See 2.4.15.7] Business intelligence. This umbrella term includes applications, infrastructure, tools, and best practices enabling access and analysis of information to optimize decisions and performance. Data management. Practices, architectural techniques, and tools for achieving consistent access to and delivery of data, to meet the data requirements of all business applications. Data mining. A process of discovering meaningful correlations, patterns, and trends by sifting through stored data. Data visualization. The graphic images or symbols that depict the data. Machine learning. Algorithms that are composed of many technologies (such as deep learning, neural networks, and natural language processing), and that are guided by lessons from existing information. Predictive modeling. This common statistical technique is used to predict future behavior. Predictive modeling solutions are a form of data-mining technology that works by analyzing historical and current data and generating a model to help predict future outcomes. A model is validated (or revised) as additional data becomes available. People analytics. Also called talent analytics, this is the application of statistics, expertise, and technology to numerous large sets of data that provide a way to combine information and data to make better organizational decisions (Gartner 2019).

3.7.1.5 Talent Development Contributes to the Bottom Line

As TD professionals use analytics to gain knowledge about what and how learners learn, they can use analytics, dashboards, and AI to help leaders see the influence learning has on the organization. With the right data and analysis, talent development can prove its contribution to the bottom line. For example, they can measure data related to employee frustration with a lack of information to do their jobs. Panopto's (2018) Workplace Knowledge and Productivity Survey showed that the average large U.S. organization loses $47 million in productivity annually due to insufficient knowledge sharing. Using an analytic approach to determine what information is needed, when, and by whom, increases the availability of information and decreases employee frustration. This gives TD professionals a way to contribute to the bottom line.

3.7.1.2.2 Ensure Data Integrity

As more organizations become more dependent upon the complex analytics that support their decisions, it is increasingly critical to ensure the data is accurate and complete. According to Forbes, 84 percent of CEOs are concerned about the quality of the data they use to make decisions (Olenski 2018). In fact, this is one of the most critical issues in data and analytics today (Onay and Öztürk 2018; Spotless Data 2017; Reynolds 2016). TD professionals should ensure that the data they use comes from the correct audience segment, that it originated from humans, and that it has been properly curated. TD professionals should assess their function's analytical capabilities and focus on what is most important to their organizations and their success. Talent development should develop its own longer-term plan and find ways to automate the delivery of information for themselves and the rest of the organization.

3.7.4.2 Pitfalls of Initial Data Analysis

Before beginning to interpret the results, TD professionals should make sure they didn't fall into any data analysis traps. Data must be skillfully analyzed and TD professionals should understand the consequences of poor data analysis. Several of the most common include: jumping to conclusions—or worse, starting with the conclusion unconscious bias overusing the mean and avoiding the mode and median incorrectly defining the sample size hypothesis testing without accounting for the Hawthorne effect or placebo effect. [See 2.8.6.2] It is important to examine the processes used and the assumptions made before starting to interpret results.

3.7.1.3.1 Dashboards

Dashboards distill the most important data so that talent development and leadership can view indicators at a glance. They are often used to monitor how the organization is executing on strategy, and are equally valuable in strategic, tactical, operational, or analytical formats. In addition to viewing indicators at a glance, dashboards can also be used to interact with the data and drill down into information needed at any time. This provides TD professionals with the ability to better advise as well as inform leadership.

3.7.7.3 Ethical Use of Data and Analytics

Data analytics is a relatively new field, and talent development's part, people analytics, is one of the newest segments of the discipline. As a result, there are few principles to guide its use. The European Union's General Data Protection Regulation (GDPR) enforces regulations for using personally identifiable information (PII). PII is data that could be used to identify an individual, such as social security number, bank account number, passport number, driver's license number, or even an email address. GDPR requires organizations to request consent from people to use their data and to tell them how it will be used, among other things. In situations where data is being transmitted between an organization and its customers, Data Processing Agreements are the legally binding agreements that outline the terms associated with the data's treatment. Using people analytics has expanded concerns about privacy, transparency, and how PII can be used in the workplace. As TD professionals begin to use statistical theory and methods in data analytics for talent development and the organization, they should know what protocols their organizations have established regarding processing and security for using personal data. This becomes more important when it is segmented into gender, race, age, performance, or other sensitive data. Talent development needs to understand its role in balancing privacy issues with data for decision making. Application of both is a requirement to provide positive results for employees and organizations alike. [See 1.6.3.5 and 2.8.6.3]

3.7 Data & Analytics

Data and analytics are key drivers for organization performance and should be drivers for talent development. This is about the ability to collect, analyze, and use large data sets in real time to affect learning, performance, and business. Discerning meaningful insights from data and analytics about talent, including performance, retention, engagement, and learning, enables the talent development function to be leveraged as a strategic partner in achieving organizational goals.

3.7.5.1 Visualization Principles

Data representation is the form in which data is stored, processed, and transmitted for use, usually in digital format. When TD professionals are asked to display information in a data visualization format, they should follow several basic principles: Consider how and where the data visualization will appear to the user. For example, should the design consider mobile-first indexing? Balance the design by choosing: symmetrical visuals—both sides are the same asymmetrical visuals—each side is different but has a common visual weight circular visuals—use the center as an anchor, placing the components around it. Plan for a contrasting background color (white is the easiest); then use a consistent color palette throughout. Optimize images before uploading. Use color, form, spatial positions, and other properties to enhance clarity. Emphasize key areas. Maintain consistency using themes, but also include some variety to keep the audience engaged (Healy 2019).

3.7.5.4 Data Visualization Methods

Many decisions are involved when designing a complete visualization system. Example methods TD professionals can use include: Color mapping, which is the choice of colors that are linked to the specific data. The goal is to easily and accurately recognize the original data set, and could include: telling the absolute data value at all data points identifying which of two data points is greater showing the difference between two data points telling which points represent selected data values showing the speed of change or the value of data at specific points (Telea 2014). Texture mapping, which is a visualization technique used to define high-frequency details in two-dimensional images that correspond to a three-dimensional model. They can be created using photos painted in a graphics software program or application. Plots can be shown in two ways. Line plots display data points, called markers, which are connected by straight lines. Scatter plots allow each data point to have its own x or y axis value. This shows trends or correlations, and the data points aren't connected with lines. Word clouds are viewed by the data visualization experts as limited at best. However, people often want to use them more seriously than intended, even though they lack insight into the data (except how many words were used how often). Word clouds are best used if they can show data in a comparative sense.

3.7.4.4 Using Data Visualization to Tell the Story

Once TD professionals interpret the results and determine they are accurate and serve the purpose for which they were intended, they can determine which visualization tools will make the most important point—what the data says. Although the numbers are important, listeners will want to know the story the data tells. TD professionals can create a story by: using the percentages to create a narrative providing context with the statistics, such as comparing to a previous year showing which benchmarks were used for comparison when interpreting results including quotes from open ended questions or interviews, if possible, to help interpret numbers. TD professionals should tap into the data they analyzed and consider the audience and the message they want to deliver. What's the story they want to tell? Each tool tells a different story; for example, charts can show relationships, distribution, comparisons over time or among items, or static composition. [See 3.7.6.2] Also available are crosstab tables, which show a pictorial comparison of the results of two or more questions. For example, comparisons could be made based on demographics such as respondents' age, organizational position, between departments, or longevity. Two guiding principles for these tools are scaling and integrity—scaling shows proportions and relationships, while integrity focuses on the presentation's truthfulness and accuracy. [See 3.7.4.3] TD professionals should select graphs that present the results in the most useful format—one that clarifies the point that was intended (Evergreen 2020). TD professionals will deliver their message best if their graphs: Induce viewers to think about the message rather than methodology, graphic design, and technology used to create the graphic. Avoid distorting what the data say. Make large data sets coherent. Encourage the eye to compare different pieces of data. Reveal the data at several levels of detail, from a broad overview to the fine structure. Serve a reasonably clear purpose, including description, explorations, tabulation, or decoration.

3.7.4.1 Process for Data Analysis

Once questions have been asked and the right data collected, TD professionals should use a deeper data analysis to identify useful information and initial conclusions. They can begin by manipulating the data by plotting it out to find correlations or creating a pivot table that allows data to be sorted and filtered using different variables. They should also calculate the mean, maximum, minimum and standard deviation of the data. As they sort and filter data, TD professionals may find that they have the necessary data; or, they may need to collect more data or ask different questions. In either case, this knowledge and what they learn about correlations, trends, and outliers helps to focus the analysis and get closer to an accurate conclusion.

3.7.2.2 Steps to Gather and Organize Data

Once the project is selected, TD professionals should begin by using a clear process that includes these steps: Define the question that needs to be answered. Set clear measurement priorities, including determining what to measure and how to measure it. Collect the data. TD professionals should consider what information exists in current databases, how the data is stored and filed, and how to collect the data. Cost, bias, and confidentiality will be factors. Analyze the data. This might mean manipulating data in several ways, such as creating pivot tables, plotting it, or finding correlations. A number of analysis tools and software can be used. Interpret the results. As TD professionals interpret the results, they should circle back to the original question to determine if it has been answered. By helping defend against any objections, the data ensures that a productive conclusion has been found. TD professionals should organize the data to answer the original question and use visuals to present it.

3.7.1.1 Principles and Applications

People analytics describes the use of data to do such tasks as improve performance, predict turnover, measure the organizational impact of leadership development initiatives, hire the best employees, or determine the effectiveness of onboarding programs in improving time-to-performance metrics. The data available vary based on the tool being used, and may include typical statistics (such as turnover, cost per hire, and the number of learners attending courses) and business impact (such as the correlations between performance improvement and profit margins). These types of data can be used to predict which experiences will help employees advance in their careers or which new hires are more likely to succeed. With analysis, TD professionals can also provide direction to senior leaders about budgeting for the fiscal year or to employees about their career development paths. By analyzing the impact of talent initiatives on the organization, TD professionals can help inform decisions about which efforts are worth pursuing for another fiscal year. Data can be captured through many methods, including questionnaires, focus groups, manager surveys, and human resource information systems.

3.7.7.2 Methods to Analyze Data

TD professionals can use several methods to analyze data: Correlation analysis evaluates the data, resulting in a statistical direct correlation, inverse correlation, or zero correlation. TD professionals should consider all factors to ensure that the statistics represent more than a causal relationship. Correlation analysis can show a relationship between courses taken and employee advancement. Multiple regression analysis quantifies how certain behaviors are tied to outputs. For example, if customer satisfaction surveys indicate that car insurance customers rate their experience as satisfactory (output) when service time on the phone is eight minutes, regression analysis helps define the service behaviors that drive the rating (such as use of the customer's name, times placed on hold, politeness, and so forth). Significance testing verifies correlation and multiple regression analysis. Because statistics can have many sources of errors (such as sampling errors, research bias, and validity), complete certainty that a relationship exists between two variables is difficult to achieve. Statistical significance means that the probability of a relationship existing between two variables is true. The findings from these types of analytics can help TD professionals guide organizations in determining what talent initiatives to implement, which types of training yield the best return on investment, and how to prioritize the use of limited resources

3.7.2.1 Selecting Initial Talent Management Analytics Projects

TD professionals require a basic understanding of analytics and the ability to communicate the results of a project to stakeholders. When they first begin to create an analytically enabled function, early project selection will be important, because it benefits both talent development and the identified stakeholder if done right. In addition, these early projects offer a way to determine whether the organization will accept the use of TD dashboards. TD professionals should identify a business problem with a clear outcome if this is one of their first experiences with analytics. To increase the chances of success, TD professionals should ensure that they will have access to the required data, and identify a project that has a good probability of success, isn't too complicated, and avoids anything that may be politically charged. An early project should also be something important to the C-suite, such as reducing costs or increasing sales. An ideal TD project will have two qualities. First, that it is a quick win, such as something that improves an existing process. This means that talent development will be able to demonstrate success rather than trying to improve something in the future and waiting for the results. Second, the project should be something that uncovers an insight or has a business impact that will create interest by senior level leaders.

3.7.5.3 Optimizing the Visualization of Data

TD professionals should ask questions about what they want to show using the data. Once they know the point they are trying to make, they can select the right visualization technique to display it. Options include: distribution of a single continuous variable relationships between two or more variables comparisons distribution of multiple variables connections composition of parts of a whole location. Each option has several types of charts from which to choose. For example, comparisons can be displayed as slope graphs, side-by-side column charts, back-to-back bar charts, dot plots, or others.

I. Selecting a Project for an Analytics Initiative

TD professionals should be able to gather and organize data to use for an analytics initiative.

I. Selecting Data Visualization Techniques

TD professionals should be able to select data visualization techniques based on what the data need to show. They should also have the skills to create and use them appropriately.

3.7.3.2 Conduct Stakeholder Analysis

TD professionals should be adept at conducting a needs assessment, because the stakeholder analysis may be broader and deeper than typical. Identifying the stakeholders and determining their power and influence are critical first steps. It is also important to segment the stakeholder group, which can be done in three ways: hierarchy, such as team leads, department heads, or directors function or department, such as sales, marketing, or operations decision-making authority, which differs from hierarchy; for example, if there is a unique situation where the stakeholder group has responsibility across departments (Anand 2017).

I. The Importance of People Analytics

TD professionals should be knowledgeable about analytics and their importance to talent development. Data and analytics refer to the management of data for all uses.

I. Data Visualization Principles

TD professionals should be knowledgeable about the principles and applications of data visualization as it relates to what the data show. They need to know how to make their point using data visualization to communicate information by displaying data in the best, most accurate way possible.

3.7.1.2 Analytics in the Workplace

TD professionals should be prepared to help their organizations as they increase the use of analytics to predict what will happen in the future. Although this is the most important focus, organizations also use analytics to increase efficiency, improve performance, and make better, more strategic decisions. This has been fueled by the availability of more data and technology, which allows organizations to capture and analyze the data faster and more economically. The additional dependence on data raises two requirements that talent development can support: a need for skilled data-fluent employees and a need to ensure the integrity of the data used to make decisions.

I. Analyzing Data and Interpreting Results

TD professionals should be skilled in analyzing results so they can identify trends and relationships among variables. They do this in two steps: analyzing data and interpreting what it means.

I. Developing a People Analytics Plan

TD professionals should be skilled in identifying stakeholder requirements so they can develop a people analytics plan.

3.7.6.2 What to Display

TD professionals should determine how they want to display the data. The visual they choose will depend on the audience and what will be most meaningful to them, as well as which will provide the best aesthetic and visual effectiveness. The following visualization techniques can be used to show different relationships: distribution of a single variable: columns, histogram, scatter chart, bar chart relationship: bubble charts, scatter chart comparison: bars and columns, timeline, line chart, scatter plots distribution of multiple variables: heat maps, bubble charts connection: relationship or connection maps, heat maps, Venn diagrams composition of the whole: pie chart, stacked bar chart location: maps, building diagrams, processes.

3.7.6.1 Presenting Data to Stakeholders

TD professionals should have an understanding of how data driven their organization is prior to planning projects and selecting or using data visualization techniques. Five factors define a data-driven organization: a strong company culture an experimentation mindset and objectively learn from failures a digital technology influence a focus on the future are organizationally agile (Sinar 2018). These characteristics help TD professionals understand the organization's preparedness level to use data for decision making. If there isn't a specific request from leadership or another stakeholder, TD professionals should define an outcome they would like to influence. They should consider the main purpose of talent development and identify which TD reports leadership uses to make their decisions. They can also look for business metrics that are tied to performance, such as increased sales. TD professionals can determine the decision they would like to impact, which becomes a likely candidate for gathering and analyzing data for a visualization first step. Before moving forward, TD professionals should make sure that their selection is not in conflict with what the stakeholder might be doing or planning. This requires them to know the stakeholder's goals and take time to predict stakeholder questions before introducing the idea to the stakeholder.

3.7.4.3 Interpret Results

TD professionals should interpret results by referring back to the original plan and purpose of the data analysis initiative. TD professionals should also make sure they have this information: The number of respondents and the total number that describe the ideal respondents. This provides the sample size. If a survey was used, determine the response rate by dividing the number of responses by the number who were asked to complete the survey. Review the aggregated number of responses to each question—this is simply counting the totals. TD professionals next need to make sense of the quantitative and qualitative information: Quantitative data is the best place to start. Working with the numbers first provides an initial focus or direction for the results. When working with numbers, it is easier to make comparisons of percentages than whole numbers, so TD professionals will want to turn most of their data into percentages. Qualitative data should be compiled after the numbers are quantified. This will be used to provide a rationale for the quantitative data message. TD professionals will want to provide context and meaning to their analysis. One way to do this is by using benchmarks, which are standards or reference points against which things can be compared or assessed. Potential benchmarks include a comparison to the last survey, other organization's data, or best practices. However, when benchmarking it is critical to compare exact questions from one to the other to avoid misinterpretation of the data. It is also possible to make comparisons within the data using cross-tabulation by breaking data out into different categories. Crosstab or cross-tabulation is a multidimensional table that records the frequency of respondents that have specific characteristics defined in each table cell. These tables show valuable data about the relationships of all the variables to one another and help to analyze cause-and-effect or complementary relationships. For example, the cross-tab table between a question about age and a question about professional development might lead to a conclusion that 20 percent of employees over age 50 want more professional development opportunities. [See 3.7.4.4] While interpreting the data, TD professionals should use good judgement and question any results that don't "feel" right. They should also dig deeply to determine correlation—when two variables move at the same time and causation—if one variable directly causes a change in another. [See 3.7.7.2] As TD professionals interpret their analysis, they must remember that it is not possible to prove a hypothesis true. Instead, it can only fail to reject the hypothesis. This means that no matter how much data is collected; chance could always interfere with the results. Asking these questions while interpreting results helps determine the legitimacy and usefulness of the conclusions: How likely will the conclusions be beneficial? Do the results answer the original research question? Does the analysis explore all perspectives? Does the data address any objections? TD professionals have likely reached a useful conclusion if these questions can be answered satisfactorily. Armed with this data analysis and interpretation, TD professionals can determine the best course of action, make recommendations, and report on their findings.

I. Principles, Definitions, and Applications of Statistical Theory

TD professionals should know fundamental statistical terminology.

3.7.4.5 The Analytic Spectrum

TD professionals typically measure talent development using metrics. "Today, thanks to plentiful data and smart analytics, companies can gain the insights they need to diagnose problems and make sound decisions" (Dearborn 2015). Using analytics along a progressive spectrum, TD professionals can use four different analyses to measure the business value and effectiveness of various training programs and other initiatives (Pease and Brant 2018): Descriptive analytics. Use it to explain what happened. This is the analysis TD professionals are most familiar with, such as assessment scores, summary activities, opinions, satisfaction, and evaluation surveys. Diagnostic analytics. Use it to explain why something happened using a variety of techniques, such as data mining and data discovery. It provides correlations for focusing on the reason something did or did not happen as expected. It can save time by knowing where to apply and concentrate next steps. Predictive analytics. Use both descriptive and diagnostic data to predict what will happen in the future. By having a sense of what the data has uncovered, TD professionals can take this information one step farther and build models that prescribe support to increase success. Prescriptive analytics. Use it to show how to make something happen. It offers the best opportunity to influence a different outcome. This is the least developed analytic because each organization has different requirements. TD professionals can use prescriptive analytics to personalize learning by matching a learner's preferences to make something happen.

3.7.5.2 Rationale for Using Visualization

TD professionals use visualization when they create a dashboard for leaders that displays the relationships between multiple data sets. Data visualization is a combination of computer science, mathematics, cognitive science, and engineering. It ensures that an audience can gain insight from and form a mental image of large, complex data sets, which might represent events, processes, concepts, or objects (Telea 2014). Data visualization helps TD professionals communicate complex information in a way that is more easily and more quickly comprehended and remembered, because the human brain processes information faster when it's presented in a visual format. Visual graphs add credibility to the data and lead to a more persuasive argument. TD professionals should use data visualization: as a means for communicating large amounts of data that may be highly complex to help audiences glean insights and patterns that they would not otherwise discern from the raw data to summarize the data in a way that brings relationships and themes to light. "Research tells us that data are more persuasive when shown in graphs. One factor may be that we are primarily visual beings and that most of us, most of the time, are skimming the narrative for things that pop out at us and catch our attention. Data visualization does just that—it provides the pop" (Evergreen 2020).

Talent development also uses analytics beyond learning and development; for example, in:

Talent development also uses analytics beyond learning and development; for example, in: Hiringimproving hiring decisionspredicting candidates' successanalyzing video interviewspredicting future workforce requirementsmatching past experience to job success. Ongoing Feedbackoptimizing the employee engagement experienceevaluating team performancedetermining if departments have the right skill set and right talentdemonstrating correlations between coaching and engagementanalyzing employee time management patterns. Organizational Supportevaluating the effectiveness of employee policiesquantifying safety and accident riskstying talent development actions to organizational resultsidentifying overtime and other forms of payroll leakagedetermining improved practices in workforce management. These options all lead to increased employee satisfaction and engagement, while at the same time increasing retention and productivity. [See 3.3.10]

3.7.7.1 Fundamental Statistical Concepts and Processes

Talent development and organizations have ample data at their disposal. In fact, there's often too much data for them to be able to make a clear decision. TD professionals should be able to sort through the data to draw accurate conclusions using fundamental statistical knowledge, including: Algorithms are a tool that depicts physical steps or behaviors and the rationale that supports them. Business analytics are solutions used to build analysis models and simulations to create scenarios, understand realities, and predict future states. Business analytics includes data mining, predictive analytics, applied analytics, and statistics, and is delivered as an application suitable for a business user with prebuilt industry content (Gartner 2019). Causal inference is the process of drawing a conclusion based on the conditions of the occurrence of an effect. Clustering is the capability to define resources on one or more interconnected midrange systems as transparently available to users and applications from within a specified group of loosely coupled systems in a local- or metropolitan-area network (Gartner 2019). Computation is a mathematical calculation. Data set is a collection of separate elements or related information that can be manipulated as one group. Mean is the average of a group of numbers. This is the most affected by the presence of rare extreme values (outliers). Normal distribution is the way observations tend to pile up around a particular value rather than spreading evenly across a range. Range is a series of values between two outer values. Standard deviation is a measure or indicator of the amount of variability of scores from the mean. The standard deviation is a statistic that's often used in formulas for advanced or inferential statistics.

3.7.1.3 Data Available in Real Time

Talent development supports the organization as it works to reach its goals and mission, and thus should provide data that bolsters stakeholder needs. Dashboards can provide stakeholders with insights about engagement, skills, and learning paths. In the past, collecting and analyzing people analytics data was a labor-intensive process that took a significant amount of time. The advantage of using analytics and AI is that the data can be analyzed and delivered to the organization's leaders in real time, which allows better decisions to be made faster. Analytics and AI deliver more powerful searches. One way they do this is by eliminating extraneous content. They can also: ensure alignment of the content to what's needed to contribute to the bottom line be used to match mentors and coaches to employees provide real-time data about safety and where the organization might be at risk potentially address how, when, and what people are learning from one another, as well as pinpoint the gaps predict turnover or how to best incentivize the workforce.

People Analytics Components

The organizational purpose of analytics is to reduce cost, drive improvements, identify new products and services, and make faster, better decisions. The purpose of analytics for talent development is similar. ATD research has found that talent development's interest in big data includes being able to: Better evaluate effectiveness of learning and development initiatives. Improve learning delivery methods. Better evaluate the impact of L&D on organizational results. Enhance decision making within talent development (ASTD 2014).

3.7.2.3 Talent Development's Role in Selecting Future Analytics Projects

Ultimately TD professionals will want to establish a portfolio of analytics projects. They should meet with senior leaders or other stakeholders to show how these projects could deliver short- and long-term value for the organization. TD professionals must be prepared to make recommendations about the projects based on both the level of impact and the ease of implementation. They should remember to align projects with organizational success, engage multiple stakeholders, and determine what support they will need: Ensure the talent strategy aligns with organizational success. Organizations that are most successful have created a talent strategy that enhances their business strategy. Goals, strategy, and outcomes should have a corresponding plan for how employees will help achieve what the organization wants to accomplish. This strategy should align with how the organization succeeds. This is important because it is the direct connection between talent development and the organization. This means that TD professionals should have a clear understanding of the organization and how it makes money (or if it's a nonprofit, what it is expected to accomplish). This ensures that leadership will be interested in the results and helps demonstrate that talent development is a valuable resource. [See 3.1.2] Engage multiple stakeholders. Talent development's key stakeholders are the organization's leaders and the business department leaders they serve. During future projects, additional leaders may be added, which is why it is important to identify consistent definitions of the problem, the measures, and the assumptions. Determine the support needed. Since every project will likely require support beyond talent development, TD professionals should determine what support they will need and from where. To do so, they'll need to obtain buy-in from others. For example, data they require might be located in departments or functions outside talent development, or experts throughout the organization may be needed to gather additional data (extant data or through interviews) and obtain input to build a measurement plan or analyze what was learned. Building these relationships will continue to be valuable in the future.

3.7.3.1 The Stakeholder's Desired Purpose

When working with a stakeholder, the purpose is what the stakeholder wants and needs to know—their goal, need, or requirement. This forms the framework for the data analysis plan. Talent development owns valuable information for other departments, such as skills needed to improve business performance, how to predict turnover, data to measure the business impact of leadership development programs, or how to determine the effectiveness of onboarding programs