Introducing the Glassdoor Employee Confidence Index

Key Findings

I. Introduction

The Glassdoor Employee Confidence Index (GECI) provides a real-time pulse on employee confidence in their company’s six month business outlook based on data from Glassdoor, one of the world’s leading sites for insights on jobs and companies with millions of reviews from employees. In aggregate, these employee voices provide a valuable window into the business outlook for the broader economy.

With the world growing busier all the time, it’s more important than ever to see the economy in real time. Most conventional confidence indices survey consumers or senior corporate leaders. By contrast, our index focuses on the perspectives of employees, providing a unique ground-up view of business outlook. There are several advantages to focusing on employees’ business outlooks: They may be more informed about business conditions than respondents in general population surveys and more immune to the natural optimism of executives in CEO or business surveys. 

Additionally, as described in Section II, past research has found that Glassdoor’s employee business outlook ratings can prove incrementally useful on top of management forecasts for predicting future business performance. Aggregating the knowledge of employees allows us a more decentralized view of what’s happening in the economy, complementing other confidence measures.

The GECI aggregates and reweights Glassdoor’s business outlook ratings to construct a nationally representative time series index, producing the monthly percentage of employees who believe that their company’s six month business performance is positive. This white paper details the methodology behind the GECI. Section II covers our data in detail, Section III details how we constructed the index, Section IV describes insights from the index, and Section V concludes. 

II. Our Data

Glassdoor’s data provides unique insight into employee sentiment. As the world’s largest site for career and workplace insights, Glassdoor collects millions of ratings and reviews from employees every year on different aspects of their workplace experience and job satisfaction. For example, reviewers rate their overall satisfaction with a company but can also provide ratings for various workplace factors such as Culture & Values, Compensation & Benefits or Senior Leadership. Reviewers name the employer they are providing the review for, enabling us to attach employer-level information like industry and employer size. Reviews include job characteristics like job title, employment type (e.g., full-time, part-time) and whether they are a current or former employee.

The component of the review we use for the Glassdoor Employee Confidence Index is the Business Outlook workplace factor rating. Reviewers can rate their employers on their “6-Month Business Outlook,” with three options: “Positive”, “Neutral” and “Negative”. Figure 1 shows the user interface of the survey as of July 2023. On a monthly basis, Glassdoor collects tens of thousands of Business Outlook ratings.

Figure 1: The Glassdoor Company Review Survey

It is important to note that reviewers on Glassdoor are not surveyed from a nationally representative sample. Instead, reviewers can elect to leave ratings and reviews when they visit the Glassdoor site. To address concerns about representativeness, we highlight two solutions: (1) the give-to-get model and (2) industry reweighting.

Firstly, one concern about representativeness is selection bias, particularly polarization bias—the idea that reviewers who choose to visit Glassdoor and leave reviews may be unusually negative or positive. To help address this, Glassdoor’s platform uses a give-to-get model, which requires that users provide reviews or other content to access and read other users’ provided content for up to a year’s time. This approach helps ensure that reviewers with moderate opinions, who may feel less inclined to provide reviews, do share their opinions. This approach has been demonstrated to reduce polarization bias (Marinescu et al. 2021).

Relatedly, there may be a selection bias in who is likely to elect to submit reviews. In a forthcoming paper (Wilmers forthcoming), the authors conduct a survey and compare the demographic characteristics of respondents who have reported writing an online job review against those who report that they have not, finding that there is no statistically significant difference in the proportions of respondents in different demographic groups across gender, highest level of education and race/ethnicity. Reviewers are less likely to be at least 60 years old (though this group represents a small share of reviewers and non-reviewers).

Secondly, as an online platform, Glassdoor’s user base may skew towards more educated or higher income workers. In Figure 2, we compare the industries of Glassdoor reviews in 2022 against employment data from the Bureau of Labor Statistics (BLS) and find that, while the information industry (which includes the tech sector) is overrepresented on Glassdoor, a broad array of other industries are represented in Glassdoor data.

Figure 2: Comparing Glassdoor Reviews against the U.S. Workforce in 2022

To address this potential source of bias, we reweight our data by industry to provide a nationally representative measure by approximating the industry mix from the latest workforce data from the BLS. We also only consider ratings from full-time and part-time employees, excluding interns and contractors.

The Power of Glassdoor’s Business Outlook Ratings
Business outlook ratings on Glassdoor offer insight into companies from those who know companies the best -- their employees. Independent researchers have demonstrated the utility of Glassdoor’s business outlook rating in predicting future business performance, business news events and stock market performance. We highlight some examples of this research below.

III. Constructing the Index

The Glassdoor Employee Confidence Index’s value is the share of Business Outlook ratings where the reviewer says the outlook is “Positive”, i.e., the number of “Positive” ratings divided by the total number of ratings. This simple “Percent Positive” value gives an intuitive interpretation for the index: “The share of employees who believe their employers have a positive business outlook over the next 6 months.”

Equation 1

Percent Positive_m = # of  “Positive” ratings_m / # of Business Outlook ratings_m where m refers to the current month. 

To adjust for any differences in composition between Glassdoor data and the national workforce, we reweight the Glassdoor data at two levels. First, we reweight based on an internal definition of the source of reviews to mitigate the impact of changes to the platform that may disproportionately attract different types of reviewers. Second, to better approximate a nationally representative workforce, we reweight at the industry level using seasonally adjusted BLS non-farm payroll employment from the Current Employment Statistics (CES) each month.

We mapped Glassdoor’s internal industry categories to the major industry sectors the BLS uses in the CES. In Glassdoor data, employers are assigned to canonical industry categories at the firm level, with 89 percent of reviews having an industry associated with them as of 2022. We aggregated some BLS industry sectors (like durable goods manufacturing and nondurable goods manufacturing into manufacturing) to ensure mapping was possible between Glassdoor’s internal industry categories and the BLS major industry sectors.

We compute the “Percent Positive” for each Glassdoor industry category and then compute a weighted average, using the share of employment in each industry from the BLS data as the weights. 

For missing data, like for the latest month when Glassdoor business outlook rating data is available sooner than the CES release, we use the most recent available data to calculate weights. For more granular cuts of the data, we reweight by industry when possible. For example, we do not reweight our industry-level Employee Confidence Index measures but we do reweight our metropolitan area-level measures by local industry employment data from the BLS. Our metropolitan area-level measures are also smoothed using a 3-month trailing average. The data is not seasonally adjusted. 

IV. Insights from the Glassdoor Employee Confidence Index

The Glassdoor Employee Confidence Index shows that employee business outlook has declined significantly in recent months as the labor market has cooled. 47.6 percent of employees reported a positive six-month business outlook for their employers as of July 2023, down 2.2 percentage points from 49.8 percent in June 2023 and down significantly (-5.8 percentage points) from a year prior when the index was at 53.4 percent in July 2022.

Starting at the end of 2022, a wave of tech and finance layoffs gripped headlines, driving a deterioration in employee confidence in their employers. While that wave of layoffs has slowed, employee sentiment has continued to decline, falling in July to the lowest level since 2016.

Figure 3: The Glassdoor Employee Confidence Index

The Employee Confidence Index has largely tracked the strength of the labor market over the data available going back to 2016. As the long expansion following the 2008 recession made slow but steady progress, employee confidence gradually improved. In 2019, however, improvement plateaued, albeit at a solid level, as the threat of trade wars and other headwinds raised concerns of an economic slowdown.

Interestingly, when the pandemic began, our measure of employee confidence was volatile but did not deteriorate as sharply as hard labor market indicators like the unemployment rate or payroll employment, perhaps as employees perceived the pandemic as a temporary public health disruption rather than a typical recession caused by enduring economic problems. In early 2021, as the recovery began in earnest and labor shortages began to dominate economic news, employee confidence began rising, peaking in early 2022 before the labor market began to cool. 

A. Glassdoor Employee Confidence Index by Industry

Underneath the headline trend, the index can reveal how employee confidence is evolving in individual industries. In Table 1, we show the current Glassdoor Employee Confidence Index level by industry and in the section below, we’ll explore a few interesting trends.

Table 1: Glassdoor Employee Confidence Index by Industry

Tech Falls from Grace

Most notably, the information sector has seen the steepest drop with the Employee Confidence Index for information falling to 48.6 percent in July 2023, a whopping 11.9 percentage point decline from 60.6 percent in July 2022.

This is a fall from grace as information nearly had the highest employee confidence of any industry in February 2022 when it peaked at 65.5 percent. While tech expanded aggressively during the pandemic recovery, the recent waves of layoffs have more than reversed employee confidence, resulting in plunging sentiment falling to the lowest level since our index’s start in 2016.

Similar to information, professional and business services (-8.7 percentage points year-over-year) and financial activities (-5.1 percentage points year-over-year) have also seen declines in employee confidence as layoffs have also hit these sectors since the end of 2022, though their declines have not been as steep as tech.

Figure 4: Employee Confidence Index for Professional Service Sectors

In-Person Service Sectors Show More Resilience

In-person service sectors like leisure and hospitality, retail trade and transportation and warehousing have been more stable over the last year. Despite concerns about economic headwinds, retail and leisure spending remain fairly strong, buoying these sectors.

Leisure and hospitality is only down 2.9 percentage points year-over-year and similarly, retail trade is down 4.1 percentage points. Transportation and warehousing rose more during the pandemic as the shift to e-commerce increased demand for shipping but as the pandemic has ended and spending patterns are normalizing, employee business outlook has declined, falling 5.2 percentage points year-over-year.

Figure 5: Employee Confidence Index for In-Person Service Sectors

B. Glassdoor Employee Confidence Index by Seniority

Employee confidence varies across seniority levels at American employers. Directors and executives are generally significantly more positive about their companies' business outlook than entry level or mid-senior level employees. 68.4 percent of directors and executives in our data report a positive business outlook for their employees, compared to 50.4 percent for mid-senior level employees and 52.1 percent for entry level employees.

Notably, employee confidence has diminished much more for mid-senior level employees, falling 6.2 percentage points year-over-year as of July 2023, than for either directors and executives or entry level employees which both have seen little change over the last year (-0.1 percentage points). This may reflect that in particular, mid-level managers may be feeling the squeeze as businesses pull back on pandemic-era expansion.

Directors and executives may see the bigger picture of business performance and have more information about business outlook. But conversely, Dunham et al. 2023 found that less tenured employees' business outlook ratings are actually more strongly related to financial distress measures, perhaps suggesting that less-tenured employees are more objective and less likely to see their employer through rose-colored glasses. Because entry and mid-senior level employees dramatically outnumber directors and executives, our Employee Confidence Index better captures the sentiment of everyday employees.

Note this data is reweighted by industry but may not be representative because we cannot always identify the level associated with a job title and because the distribution of seniority within and across industry may differ between Glassdoor and BLS data.

Figure 6: Employee Confidence Index by Seniority

V. Conclusion

The Glassdoor Employee Confidence Index offers a new and valuable perspective on the state of the economy, leveraging the unique perspective that workers have on their employers’ business outlook. Our data points to an economy that has made substantial strides in recovering from the pandemic, but is now facing strong headwinds. Industries like tech that exploded during the pandemic are now facing a hangover as employees report waning confidence in their employers’ business outlook.

The index builds on a body of research that highlights the utility of employees as a crucial source of information for managers and economic decision makers alike. By analyzing hundreds of thousands of employee business outlook ratings available annually on Glassdoor, we present a near-real time picture of business health in the economy at a time when economic volatility makes timely data all the more important. Ultimately, the Glassdoor Employee Confidence Index serves to highlight information and perspectives on the economic outlook from those who know businesses best—their employees.

References

deHaan, Ed, Nan Li, and Frank S. Zhou. "Financial Reporting and Employee Job Search." Journal of Accounting Research, vol. 61, no. 2, May 2023, pp. 571-617, doi: https://doi.org/10.1111/1475-679X.12469.

Dunham, Lee M., John Garcia, Jaime L. Grandstaff, and Sijing Wei. "Do Employees Waive Financial Red Flags through the Glassdoor?" Journal of Forensic Accounting Research, 2023, doi: https://doi.org/10.2308/JFAR-2022-008.

Green, T. Clifton, Ruoyan Huang, Quan Wen, and Dexin Zhou. "Crowdsourced Employer Reviews and Stock Returns." Journal of Financial Economics, vol. 134, no. 1, 2019, pp. 236-251. ISSN 0304-405X, doi: https://doi.org/10.1016/j.jfineco.2019.03.012.

Hales, Jeffrey, James R. Moon, and Laura A. Swenson. "A New Era of Voluntary Disclosure? Empirical Evidence on How Employee Postings on Social Media Relate to Future Corporate Disclosures." Accounting, Organizations and Society, vol. 68-69, 2018, pp. 88-108, doi: https://doi.org/10.1016/j.aos.2018.04.004.

Huang, Kelly, Meng Li, and Stanimir Markov. "The Information Asymmetry between Management and Rank-and-File Employees." AAA 2019 Management Accounting Section (MAS) Meeting, SSRN, doi: https://doi.org/10.2139/ssrn.3233855.

Huang, Kelly, Meng Li, and Stanimir Markov. "What Do Employees Know? Evidence from a Social Media Platform." The Accounting Review, vol. 95, no. 2, 1 Mar. 2020, pp. 199-226, doi: https://doi.org/10.2308/accr-52519.

Marinescu, Ioana, Andrew Chamberlain, Morgan Smart, and Nadav Klein. "Incentives Can Reduce Bias in Online Employer Reviews." Journal of Experimental Psychology: Applied, vol. 27, no. 2, 2021, pp. 393-407, doi: https://doi.org/10.1037/xap0000342.

Symitsi, Efthymia, and Panagiotis Stamolampros. "Employee Sentiment Index: Predicting Stock Returns with Online Employee Data." Expert Systems with Applications, vol. 182, 2021, article 115294, doi: https://doi.org/10.1016/j.eswa.2021.115294.