Building a Strategic HR Scorecard: Connecting Human Capital Investments with Business Outcomes

A general Balanced Scorecard places people at the base of its causal chain and then, in most implementations, measures them with whatever HR already collects. Articles 07 and 09 examined the scorecard as a strategy tool, including Hanne Nørreklit's objection that its causal links are assumed, and that ground is not repeated here. The narrower problem for HR is that "learning and growth" is a container, not a theory. It does not say which employees matter, what they must be able to do, which practices would enable them, or how long any of this should take to reach customers or profits.

Kaplan and Norton tried to close the gap themselves. In a 2004 article on the "strategic readiness" of intangible assets, they proposed measuring human capital by first identifying strategic job families, "the positions in which employees with the right skills, talent, and knowledge have the biggest impact on enhancing the organization's critical internal processes," and then measuring the competency gap in those positions. In one consumer bank, only 40 of the 100 financial planners the strategy required were proficient: a readiness of 40 percent. In three other companies, strategic success depended on competencies in less than 10 percent of the workforce (Kaplan and Norton, 2004).

Two ideas in that answer run through this article: measurement should start from the strategy's critical processes, not from the HR department, and it should concentrate on a small part of the workforce instead of averaging across all of it.

Becker and Huselid came to the problem from research on high-performance work systems, the subject of Article 19. Summarizing five national surveys of US listed firms between 1991 and 2000, covering more than 3,200 firms, they wrote that "the effect of a one standard deviation change in the HR system is 10-20% of a firm's market value" (Becker and Huselid, 2006, quoting Huselid and Becker, 2000). Article 15 explained why such associations cannot be read as returns on HR investment. The scorecard grew from the practical question they raised: how would a particular firm manage and measure that contribution?

Their answer, set out in The HR Scorecard (Becker, Huselid and Ulrich, 2001), began by changing what was to be measured. The object was what they called the HR architecture: the HR function (its professionals and their competencies), the HR system (the policies and practices for selecting, developing, rewarding and organizing people), and the strategic employee behaviours those produce. Value, in their argument, lies in the third element and its connection to the firm's strategy map. Becker and Huselid (2006) later restated the point: "it is the fit between the HR architecture and the strategic capabilities and business processes that implement strategy that is the basis of HR's contribution to competitive advantage."

The distinction for which the book is best known, credited by the authors to Ulrich, separates HR doables from HR deliverables. Doables are the activities HR performs, such as courses run and vacancies filled. Deliverables are outcomes of the HR architecture that directly drive strategy implementation, such as the share of a utility's field engineers who fix a fault on the first visit. The authors placed deliverables in employee behaviour, not in its drivers such as skills or motivation, because behaviour is what line managers' goals depend on most directly. Existing HR measures, in their diagnosis, captured doables.

The book's seven steps run from strategy down to measures: clarify the business strategy; build the business case for HR as a strategic asset; create a strategy map; identify the HR deliverables within it; align the HR architecture with those deliverables; design the strategic HR measurement system; and implement management by measurement. The order carries the argument: measures arrive only at the sixth step, after the strategy map has said which behaviours matter. Efficiency remains one of the scorecard's four dimensions, because cost discipline still matters, but it no longer stands in for effectiveness, the confusion Article 15 examined.

Four years later Huselid, Becker and Richard Beatty (2005) argued that the HR Scorecard had too often been read as a scorecard for the HR function. In The Workforce Scorecard they separated a strategy for the business, a strategy for the workforce and a strategy for the HR function, measured respectively by a balanced scorecard, a workforce scorecard and an HR scorecard. The workforce scorecard is built from the outside in, and its four elements are workforce success, leadership and workforce behaviours, workforce competencies, and workforce mindset and culture. Its most significant change was ownership: workforce success belongs to line managers, while HR answers for the systems that support them. Becker and Huselid (2006) described this as "a shared responsibility for strategic workforce performance between line managers and HR professionals."

Comparison of the three scorecards. The first three rows follow Huselid, Becker and Beatty (2005) and Kaplan and Norton (2004); the "characteristic failure" row is this article's interpretation.

The Workforce Scorecard also made explicit the principle implied by Kaplan and Norton's job families: differentiation. In a Harvard Business Review article, Huselid, Beatty and Becker (2005) argued that companies "simply can't afford to have A players in all positions." They proposed identifying "A positions," which must meet "the dual criteria of strategic impact and performance variability," and investing disproportionately in them. Because "there are no inherently strategic positions," a job is strategic only in relation to a particular strategy; such positions, they argued, are usually under a fifth of the workforce.

Performance variability is the criterion most often missed. A role can be essential without being one where better performance creates much more value. For scorecard design the consequence is that a workforce measure averaged across all employees can hide the result that matters most. An engagement score that rises across 10,000 people can coexist with losses among the 200 people the strategy depends on.

Scorecard guidance usually labels workforce measures as leading indicators and financial measures as lagging ones. The labels describe a hoped-for relationship, not a property of the measure. Employee turnover is a leading indicator only if, in this organization, changes in it are followed by changes in something the strategy needs, after a delay that can be estimated. Heskett and his colleagues were careful about this when they set out the service-profit chain: its links, they wrote, "should be regarded as propositions" (Heskett et al., 1994).

Figure 1 shows the chain in the general form shared by the HR Scorecard and by Becker and colleagues' earlier model of HR and shareholder value (Becker, Huselid, Pickus and Spratt, 1997). Each arrow implies a delay as well as a cause, and the delays are long. Patrick Wright and John Haggerty (2005) cited a survey in which more than 70 HR vice-presidents estimated, on average, 9 to 10 months to design a new HR system and a further 10 to 12 months to deliver it. "If the first link in the causal chain takes almost two years," they reasoned, "it is not unreasonable to expect a 3-4 year total time lag before the relationship would be observed."

Two practical rules follow. Every leading indicator should carry a stated prediction and a stated lag: if retention in the first 90 days rises by ten points, error rates should fall within two quarters. A scorecard without predictions cannot be wrong, and so teaches nothing. The review cycle must match the lag. A workforce measure judged each quarter against a link expected to take two years will look like failure long before it could look like anything else.

The service-profit chain is the most thoroughly tested version of this logic. James Heskett, Thomas Jones, Gary Loveman, Earl Sasser and Leonard Schlesinger (1994) proposed that profit and growth are driven mainly by customer loyalty, loyalty by satisfaction, satisfaction by the value of the service, value by satisfied, loyal and productive employees, and employee satisfaction by "high-quality support services and policies that enable employees to deliver results to customers." Much of their evidence compared units at one point in time: Taco Bell, for instance, had found that the 20 percent of its stores with the lowest staff turnover had double the sales and 55 percent higher profits of the 20 percent with the highest turnover. The authors called their data on what service employees value "preliminary at best."

The best-known attempt to quantify the chain inside one company came from Sears. Anthony Rucci, Steven Kirn and Richard Quinn (1998) described an employee-customer-profit model built from store data with causal pathway modelling. The figures most often quoted from it, reported here from a summary that reproduces the article's exhibit, are that a 5-unit improvement in employee attitude drove a 1.3-unit improvement in customer impression, which drove a 0.5 percent increase in revenue growth. It was serious internal measurement, and it also shows the limit of any scorecard. A well-calibrated chain inside the stores could not tell Sears whether its stores were the right business to be in; Sears Holdings filed for Chapter 11 protection in October 2018. A scorecard tests how well a strategy is executed, not whether it is sound.

Later research qualified almost every link. In a study of one leading UK grocery retailer, Rhian Silvestro and Stuart Cross (2000) found correlations among profit, customer loyalty and satisfaction, service value, internal service quality and productivity, but "no support for the claim that these are driven by employee satisfaction and loyalty." In fact, "there was a strong correlation between employee dissatisfaction and store profitability." Silvestro (2002) argued that the employee link depends on context, holding best where staff-customer contact is close and central to the value offered. Benjamin Schneider and colleagues (2003), analysing attitude data from 35 companies over eight years, found overall job satisfaction and satisfaction with security predicted by return on assets and earnings per share "more strongly than the reverse": success may produce satisfied employees as much as the other way round. A Gallup-led study of 2,178 business units in 10 organizations reached a more favourable conclusion: employee perceptions predicted later outcomes, and reverse causality "existed but was weaker" (Harter et al., 2010). Four of its five authors worked for Gallup, which sells engagement surveys.

The most comprehensive test, a meta-analysis by Jens Hogreve and colleagues (2017), found that "all the proposed links are statistically significant and substantial," but that effect sizes vary considerably, partly with the type of service, and that the results "challenge the implicit SPC rationale that firms should always maximize employee satisfaction and external service quality." A study of more than 500 branches of a Brazilian bank concluded that "superior satisfaction alone is not an unconditional guarantee of profitability" (Kamakura et al., 2002). The evidence argues neither for abandoning the chain nor for copying it. The links hold on average and are unreliable in any particular case, so a scorecard should treat each one as a local hypothesis and expect some to fail in some units.

The break inside HR. One link is weak before the chain even leaves the HR function. Patrick Wright and Lisa Nishii (2013) distinguished intended, implemented and perceived HR practices, and employees' reactions to them; practices that are implemented "may often be implemented in ways that differ from the initial intention." As David Guest (2011) put it, "It is line managers, not HR managers who implement HRM." A scorecard that records whether a policy exists measures intention. In the AMO logic of Article 18, it is employees' experience of a practice that changes their ability, motivation and opportunity to perform.

The scorecard's authors and their colleagues have been unusually candid about how the tool is used. John Boudreau and Peter Ramstad (2002, p. 14) reported that "when we work with scorecard designers, they note that the majority of scorecards measure only HR operations and activities," and that even strategic-looking links were "often superficial," pairing an organizational goal of "speed to customers" with an HR measure of "faster time-to-fill." Other scorecards applied the four Balanced Scorecard perspectives to the HR function itself, turning HR's customers into internal clients and its financials into programme budgets, with little link to organizational outcomes. These are observations from consulting practice, not survey findings, but they match what Becker and Huselid said themselves.

In a 2003 HR Magazine article titled "Measuring HR? Benchmarking Is Not the Answer!", Becker and Huselid argued that relying on benchmarked cost and efficiency comparisons "not only fails to measure HR's important contributions to firm success, it also can encourage an approach to human capital management that is counterproductive." The habit persists, they suggested, because benchmarking is doable, looks credible and lets HR report on its performance without rethinking what HR is for. Strategic measures, they conceded, are slower, harder and less crisp than cost per hire, but "it is better to be approximately right than precisely wrong." By 2006 they were criticizing their own field too: attention to the internal character of HR systems was "the natural comfort zone of HR scholars," and research needed "less emphasis on the 'black box' within the HR architecture" and more on what lay between it and firm performance (Becker and Huselid, 2006).

Article 15 set out the general reasons activity measures dominate: they are available, controllable and quick. The scorecard adds a mechanism of its own. Once a template exists, every box must be filled by the next reporting date, and the nearest available number fills it. Strategic deliverables need new data, often from operations or customers, and agreement with line managers about what counts; in Becker and Huselid's comparison, benchmarking reconfigures existing data while strategic metrics require new measures. A drifted scorecard is usually less a design failure than a data-collection decision deferred, quarter after quarter.

A different critique questions the whole chain. Guest (2011) concluded that "after hundreds of research studies we are still in no position to assert with any confidence that good HRM has an impact on organisation performance." Others in the same tradition observed that "the vast majority of studies has defined performance from a strongly managerialist perspective, using performance measures most important to shareholders" (Paauwe, Wright and Guest, 2013). The Harvard model discussed in Article 17 would put the same question to any HR scorecard: whose outcomes is it designed to report, and are employees' own outcomes among them?

Bangladesh's listed companies already have a board-level body with HR responsibilities. The Corporate Governance Code issued by the Bangladesh Securities and Exchange Commission (BSEC) on 3 June 2018 requires every listed company to have a Nomination and Remuneration Committee (NRC) of non-executive directors, which is to oversee, among other matters, "identifying the company's needs for employees at different levels and determine their selection, transfer or replacement and promotion criteria" and "developing, recommending and reviewing annually the company's human resources and training policies" (conditions 6(5)(b)(v) and (vi)). The wording concerns policies, criteria and remuneration; nothing in condition 6 asks the committee to review evidence of what the HR system produces.

BRAC Bank's annual report for 2025 is a useful case because it goes further than most. Its section on "Human Resource Accounting" describes 2025 as the year the bank began "treating training costs as a strategic capital investment rather than an operational expense," and reports five years of training and per-employee data (BRAC Bank, 2025, pp. 177-179). Figures are the bank's own (report version dated 9 July 2026); the analysis is this article's.

The report draws a causal conclusion from the first two figures. The growth in operating profit per employee, it says, "demonstrates a high yield on our training investments," a "Knowledge-to-Profit" conversion that "confirms that our skilled workforce is driving significant value." The section ends by calling its results "empirical evidence that human capital investment is the primary engine of our sustainable growth."

Read through the frameworks above, this is a scorecard with the middle of the chain missing. Training hours are doables. Operating profit per employee is a financial ratio whose numerator moves with interest margins, balance-sheet growth and loan quality as well as with people. No measure of what trained staff can now do, or of which processes improved, connects the two. Wright and Haggerty (2005) identified a further problem with exactly this pairing: measures such as "the number of hours training received are heavily influenced by the past and current financial performance of a firm." BRAC Bank's training expense per employee rose from Tk 1,128 in 2021 to Tk 10,518 in 2025 (p. 162) while the bank was also becoming more profitable; profit may be funding training as much as training produces profit. Nor is a customer link visible: the satisfaction index was flat across the three years reported.

The most informative workforce figure sits elsewhere in the report. The ratio table records that turnover among the bank's highest performers rose from 3 percent in 2024 to 9 percent in 2025, the highest in the five years shown. The Human Resource Accounting section does not mention it, and presents the overall turnover rate, also 9 percent, as part of "a healthy turnover profile." The bank's own SWOT analysis lists "Talent Attrition: Intense competition for specialized skills in fintech, data analytics, and high-level compliance" among its weaknesses (p. 186). In Section 3's terms, the narrative averages across 10,293 employees while the disclosed measure closest to a strategic workforce deliverable moved the wrong way.

Three cautions apply. An annual report is a public disclosure, not the internal scorecard management may use, and BRAC Bank discloses more than many listed companies, including the high-performer figure that makes this analysis possible. A one-year rise in a small group's turnover may be noise; the figure was 8 percent in 2022. And nothing here shows that the training failed, only that the evidence presented cannot show it worked. The measures that could, such as capability and retention in the fintech, data and compliance roles the bank itself names as scarce, are not reported. An NRC applying the HR Scorecard's logic would ask for those measures before approving the next training budget.

Constructed teaching case. Gorai Superstores Ltd. is a fictional company, and its people, figures and events are invented for teaching; a search found no real company of that name.

Gorai Superstores runs 34 supermarkets in Dhaka. Its three-year strategy is to become the chain Dhaka households trust most for fresh fish, meat and produce, measured by how often loyalty-card members return. It employs about 2,100 people, including 900 cashiers and sales assistants, 600 fresh-section staff and 68 fresh-section supervisors, two in each outlet.

The HR department's monthly report shows staff hired, days to fill a vacancy, training hours, HR cost per employee and an annual engagement score, and all five have improved this year. Loyalty-card members, meanwhile, are visiting less often. Fresh-section wastage ranges from 3 percent of sales at the best outlet to 9 percent at the worst, and the share of fresh lines fully stocked at opening time from 95 to 70 percent. The weakest outlets have the most fresh-section staff in their first two months and the most supervisors appointed within the past year; about 35 percent of new fresh-section staff leave within 90 days. The chief executive wants a scorecard the board can use.

Worked through in this order, the steps produce a scorecard along the following lines.

Illustrative scorecard for the fictional case. The lags are assumptions to be tested, not estimates drawn from research.

What the exercise is designed to show. The finished scorecard keeps almost none of the original HR report, yet HR's work is more visible in it, because each HR measure now sits beneath the behaviour it should produce. The sequence did the work: deliverable, positions, measures, test. Starting instead from the HR data already available is how scorecards drift back to activity. And because grocery is the setting in which Silvestro and Cross found the employee-satisfaction link failing, the scorecard deliberately tests capability and retention in fresh sections, not engagement, as its first link.

Module 04 has approached strategic HRM from several directions: as a claim about fit between HR and strategy (Article 16), as competing models of what HR should serve (Article 17), as a theory of how practices work through employees' ability, motivation and opportunity (Article 18), and as bundles of high-performance practices (Article 19). The HR scorecard is where these ideas become measurable: external fit becomes the alignment dimension, internal fit and the high-performance work system a dimension of their own, and AMO describes the drivers behind the deliverables. The scorecard adds no new theory. It forces an organization to state its own theory of how people create value precisely enough to be checked.

That is also where its limits lie. Links that hold on average can fail, or run backwards, in a particular firm, and a lag of three to four years means many scorecards are judged before their hypotheses could be tested. The unresolved issue is one of patience and power more than design. A scorecard that tests its links will sometimes show that a favoured programme does not work, and that the most useful workforce measures depend on data HR does not own. Whether senior leaders want that information, and will wait for it, decides whether an HR scorecard becomes part of management or one more monthly report.

Select an organization you know well, or can research through dated public sources. You may disguise its name.

Prepare a report of 2,500 to 3,000 words that:

1. states the organization's strategy in one paragraph and builds a simple strategy map that identifies its critical internal processes;
2. identifies two to four HR deliverables and the A positions in which they are performed, justifying each against the criteria of strategic impact and performance variability;
3. designs a scorecard covering all four HR Scorecard dimensions, giving for each measure its definition, data source and owner, and the hypothesis and time lag it represents;
4. audits the organization's current HR reporting, classifying existing measures as doables or deliverables and explaining which you would keep, demote or drop;
5. evaluates the strongest and the weakest causal link in your scorecard, using the service-profit chain evidence and at least one alternative explanation (reverse causation, an implementation gap or a confounding factor); and
6. sets out how the scorecard would be tested in its first eighteen months, including a comparison design, the result that would show a link is failing, and how the measures would be protected from becoming distorting targets.

Use at least five academic sources, including at least three from this article's reference list. Every figure about the organization must come from a document or dataset you could produce if asked. Where the data do not exist, say so and explain what you would collect.

Assessment rubric (100 marks)

1. Becker, B. E., Huselid, M. A., & Ulrich, D. (2001). The HR scorecard: Linking people, strategy, and performance. Harvard Business School Press.
2. Becker, B. E., & Huselid, M. A. (2003). Measuring HR? Benchmarking is not the answer! HR Magazine, 48(12), 57–61. Author-hosted copy: https://www.markhuselid.com/pdfs/articles/2003%20Becker-Huselid%20HR%20Magazine.pdf
3. Becker, B. E., & Huselid, M. A. (2006). Strategic human resources management: Where do we go from here? Journal of Management, 32(6), 898–925. https://doi.org/10.1177/0149206306293668
4. Heskett, J. L., Jones, T. O., Loveman, G. W., Sasser, W. E., Jr., & Schlesinger, L. A. (1994). Putting the service-profit chain to work. Harvard Business Review, 72(2), 164–174. Quotations checked against the 2008 HBR reprint: https://hbr.org/2008/07/putting-the-service-profit-chain-to-work
5. Hogreve, J., Iseke, A., Derfuss, K., & Eller, T. (2017). The service-profit chain: A meta-analytic test of a comprehensive theoretical framework. Journal of Marketing, 81(3), 41–61. https://doi.org/10.1509/jm.15.0395
6. Huselid, M. A., Beatty, R. W., & Becker, B. E. (2005). "A players" or "A positions"? The strategic logic of workforce management. Harvard Business Review, 83(12). Author-posted reprint: https://www.markhuselid.com/pdfs/articles/2005%20Huselid-Beatty-Becker%20HBR%20Paper.pdf
7. Huselid, M. A., Becker, B. E., & Beatty, R. W. (2005). The workforce scorecard: Managing human capital to execute strategy. Harvard Business School Press.
8. Kaplan, R. S., & Norton, D. P. (2004). Measuring the strategic readiness of intangible assets. Harvard Business Review, 82(2), 52–63. https://hbr.org/2004/02/measuring-the-strategic-readiness-of-intangible-assets
9. Schneider, B., Hanges, P. J., Smith, D. B., & Salvaggio, A. N. (2003). Which comes first: Employee attitudes or organizational financial and market performance? Journal of Applied Psychology, 88(5), 836–851. https://doi.org/10.1037/0021-9010.88.5.836
10. Silvestro, R., & Cross, S. (2000). Applying the service profit chain in a retail environment: Challenging the "satisfaction mirror." International Journal of Service Industry Management, 11(3), 244–268. https://doi.org/10.1108/09564230010340760

Other sources cited

• Becker, B. E., Huselid, M. A., Pickus, P. S., & Spratt, M. F. (1997). HR as a source of shareholder value: Research and recommendations. Human Resource Management, 36(1), 39–47.
• Boudreau, J. W., & Ramstad, P. M. (2002). Strategic HRM measurement in the 21st century: From justifying HR to strategic talent leadership (CAHRS Working Paper 02-15). Cornell University. https://hdl.handle.net/1813/77386
• Fahed-Sreih, J. (2009). The HR scorecard: Linking people, strategy and performance [Book review]. Management Research News, 32(3), 297–299. https://doi.org/10.1108/01409170910943147
• Guest, D. E. (2011). Human resource management and performance: Still searching for some answers. Human Resource Management Journal, 21(1), 3–13. https://doi.org/10.1111/j.1748-8583.2010.00164.x
• Harter, J. K., Schmidt, F. L., Asplund, J. W., Killham, E. A., & Agrawal, S. (2010). Causal impact of employee work perceptions on the bottom line of organizations. Perspectives on Psychological Science, 5(4), 378–389. https://doi.org/10.1177/1745691610374589
• Kamakura, W. A., Mittal, V., de Rosa, F., & Mazzon, J. A. (2002). Assessing the service-profit chain. Marketing Science, 21(3), 294–317. https://doi.org/10.1287/mksc.21.3.294.140
• Kaplan, R. S. (2010). Conceptual foundations of the balanced scorecard (Harvard Business School Working Paper 10-074).
• Paauwe, J., Wright, P. M., & Guest, D. E. (2013). HRM and performance: What do we know and where should we go? In J. Paauwe, D. E. Guest & P. M. Wright (Eds.), HRM and performance: Achievements and challenges. Wiley.
• Rucci, A. J., Kirn, S. P., & Quinn, R. T. (1998). The employee-customer-profit chain at Sears. Harvard Business Review, 76(1), 82–97.
• Silvestro, R. (2002). Dispelling the modern myth: Employee satisfaction and loyalty drive service profitability. International Journal of Operations & Production Management, 22(1), 30–49. https://doi.org/10.1108/01443570210412060
• Wright, P. M., & Haggerty, J. J. (2005). Missing variables in theories of strategic human resource management: Time, cause, and individuals. Management Revue, 16(2), 164–173. https://doi.org/10.5771/0935-9915-2005-2-164
• Wright, P. M., & Nishii, L. H. (2013). Strategic HRM and organizational behaviour: Integrating multiple levels of analysis. In J. Paauwe, D. E. Guest & P. M. Wright (Eds.), HRM and performance: Achievements and challenges (pp. 97–110). Wiley.

Bangladesh sources

• BRAC Bank PLC. (2025). Annual report 2025 (version updated 9 July 2026), pp. 162, 177–179, 186. https://www.bracbank.com/en/page/annual-report-2025
• Bangladesh Securities and Exchange Commission. (2018). Corporate Governance Code, Notification No. BSEC/CMRRCD/2006-158/207/Admin/80, 3 June 2018, condition 6. Text checked against the verbatim reproduction in GSP Finance Company (Bangladesh) Ltd., Annual Report 2022, Annexure C: https://www.gspfinance.com/assets/media_files/pdf_docs/COMPLIANCE_REPORT_ON_CORPORATE_GOVERNANCE_CODE_BY_BSEC.pdf

Verification notes: The HR Scorecard (2001) and The Workforce Scorecard (2005) are described from the authors' own summaries of them (including Becker and Huselid, 2006, and Huselid's published description of the 2005 book) and from publisher search-inside text; they are paraphrased, not quoted. The Huselid and Becker (2000) market-value estimate is quoted as it appears in Becker and Huselid (2006). The Sears figures are reported from a summary that reproduces the 1998 article's exhibit, because the original was not accessible. Silvestro and Cross (2000), Silvestro (2002), Schneider et al. (2003), Hogreve et al. (2017), Harter et al. (2010) and Kamakura et al. (2002) are cited from their published abstracts. The BRAC Bank figures were checked against the published PDF; BRAC Bank's analysis is quoted, and the reading of it is this article's. The Gorai Superstores case, its scorecard, the comparison table in Section 3 and the quiz scenarios are original teaching material for this series.

Next in The Human Capital MBA series: Module 05 (Organizational Behaviour) opens with Article 21.