How to Make the FEVS Better
by Peter Bonner
Recommendations to improve the Federal Employee Viewpoint Survey (FEVS) without cancelling it.
This is the second in a two-part series on the Administration’s decision to cancel the 2025 Federal Employee Viewpoint Survey. Review what we are losing because of the cancellation here.
Our first post explored how the FEVS, the Federal Employee Viewpoint Survey, helps improve efficiency, performance, and effectiveness of federal employees and the programs on which they serve. Cancelling FEVS in 2025 forgoes the opportunity to realize this tool’s benefits through the feedback loops on employee performance. Losing this data for 2025 hamstrings federal managers from making evidence-based decisions.
Bottom Line: Conduct the FEVS for 2025 (even if only for the core 16 questions required by statute).
FEVS Challenges & Recommendations
FEVS is always late – Build on recent initiatives to turn FEVS results around in less than two months and engage in pulse surveys.
FEVS is not actionable – Provide manager support tools, already in use, across government to help interpret data and identify ways to work on issues.
FEVS is meaningless – Support pulse surveys and other alternatives to employee feedback in addition to the FEVS.
FEVS puts dominant employees in the ‘driver’s seat’— Balance agency resources and capacity with FEVS improvement initiatives through authentic debate with employees.
FEVS is performative – Resist the temptation to control the narrative and overly broadcast FEVS results and activities; give managers a chance to do their jobs.
FEVS is not perfect but it can be improved. This piece details key FEVS challenges and identifies recommendations to make FEVS better with greater impact on agency performance.
One example of the value of FEVS is understanding the degree of employee commitment through Employee Net Promoter Score (eNPS). The FEVS statement “I recommend my organization as a good place to work” (Q. 46) provides the foundation for a federal eNPS. At 69% positive responses for 2024, the federal eNPS is quite high and high eNPS scores are correlated not only with improved retention and employee satisfaction, but also with profitability and customer satisfaction. Managers can use a high or low eNPS score to gauge the foundations for strong employee and work unit performance. This is but one example of how FEVS can help agency leaders and managers act on FEVS feedback.
Federal leaders, managers, and employees have criticized FEVS and its execution. NASA and the Veterans Administration have opted to satisfy Congressional requirements for surveys by using their own surveys instead of FEVS so that they can get the information in the hands of their managers faster and ask questions targeted to their specific environment. Federal Chief Human Capital Officers (CHCOs), managers, and employees have voiced some common complaints.
Common Complaints about FEVS
“FEVS is always late” – FEVS results are historically shared in a long, slow-moving waterfall throughout the government. In a typical year, results have taken months for OPM distribution to the agencies: usually there is about half a year of lag between issuing the survey in the spring to agency distribution of results in late fall. Then, it also takes time for agency CHCO offices to send the results to agency units. Though OPM has improved its part of the timing (down to seven weeks in recent iterations), this still means that at best the data is two quarters old by the time it gets into the hands of managers who can do something about it. At worst, in cases where the actual employee attitudes and opinions may have changed radically over this period, managers are still flying blind. The employee expectations of some positive action by the agency on the issues they have identified wanes and leads to skepticism. Plus, the strengths and weaknesses of the employee experience may be wholly different six months after they have filled out the survey. These delays are largely due to process and technology challenges and would not be tolerated in the private sector.
“FEVS is not actionable” – FEVS scores provide a snapshot on employee attitudes and opinions and compares those scores across previous years. They tell managers whether their scores have increased or decreased, but do not tell them what to do about it. Deciding what to do with the FEVS results for an agency or unit is inconsistent and haphazard. Some agencies provide FEVS diagnostic and support tools (e.g., NIH EVS Art tool that provides heat maps to target specific challenges and “at risk” components) agency to managers. This enables them to work with their units on identifying actions to improve their lower scores while acknowledging and continuing to build on their strengths. These tools also help agency leaders track results and aid managers. However, this is far from common. Many managers are left on their own, without training, to interpret the results and decide what to do with them. Finally, FEVS covers topics that individual managers cannot control and seldom influence – pay, budgets, resources, etc. They can take little action to affect employee experience in these areas.
“FEVS is meaningless” – After managers receive their FEVS results, sometimes they have trouble perceiving how the scores relate to their units. They find the scores unrecognizable in comparison to their lived experience and ongoing feedback discussions with their employees. This can persist even after they have conducted root-cause FEVS workshops with employees. For example, if their scores are declining on the quality of communications or the need for adequate resources and their employees have never expressed discontent with these topics, managers can doubt the veracity of the FEVS results. This disconnect can be caused by delays in distributing the FEVS results (see above), response rates that only captured a small subset of the employee population (at an average 41% response rate this means that the feedback from 6 out of 10 employees in a work unit is missing), or an influx of new employees who do not see these challenges as past employees may have. Furthermore, the definitions of “manager” and “leader” are inconsistent at agencies that have multiple management levels in their structures and also have political leaders in the mix (e.g., who is my “leader” that I am rating?). This leads to managers to not trust the results and seeing the follow-up activities as a waste of time.
“FEVS puts a few dominant employees in the ‘driver’s seat’” – Some agencies (and administrations) place such a profound emphasis on the FEVS results, that agency leaders and managers believe that it gives power over to particularly vocal employees whose views may not be representative of their peers. Effective agencies broadly announce and post FEVS results, create FEVS scorecards comparing bureaus and offices, build goals based on FEVS results into performance plans, conduct action planning sessions with employees, and publish and track action plans. Though these activities can be helpful, they can also create the perception that FEVS results outweigh other agency performance metrics. That prominence can enable vocal employees to sway management decisions rather than leaders balancing the FEVS actions with other agency needs. Leaders think this takes the focus off the mission and organizational health and puts it on employee welfare.
“FEVS is performative” – Federal leaders across administrations have sometimes worked to control the narrative regarding the FEVS results – taking credit for scores improving while explaining away, shifting blame, or criticizing the instrument when the scores go down. Though some of this can be expected in a political system, this breeds cynicism attached to how seriously political leaders take improving employee experience. Furthermore, FEVS scores have become performance elements in executive and managerial performance plans, sometimes leading to short-term actions aimed at manipulating the scores rather than working on the systemic issues the scores identify (e.g., short-term boost to budget or compensation versus handling discontent with leadership quality). Managers know that attributing FEVS scores cause-and-effect to specific workforce actions belies the impacts of budget and resource decisions, technology deployments, turnover, and other external factors that impact the scores. On the other hand, with diligence and a focus on management quality, the Veterans Administration has seen employee engagement score increase significantly.
How to Improve FEVS
Short-Term Improvements
Conduct the 2025 FEVS Before the End of the Year, if only for the core 16 questions required by statute, so that there is no year-long break in FEVS data for agencies and policy makers. (The statutory requirement requires “the agencies” to administer an annual survey with the 16 keep questions.) With this action, agency unit managers can use this information to improve efficiency. Administration and agencies can obtain a baseline on some of its workforce policies. For example, FEVS asks questions about performance management. Agencies can find out how effective their changes to the performance management process are.
Release Results Faster and Support CHCOs by sending results to agency units for their interpretation and action planning. Recent FEVS reports were sent to the agencies within seven weeks. With technology support and assistance to the agency CHCO offices, the turnaround time could be less than three months to get the data in the hands of front-line managers. This kind of speed will make the FEVS feedback actionable, giving work units the ability to improve communications, performance, the provision of resources or other areas the data targets.
Push Adoption of Existing Agency Tools that help managers interpret and act on the FEVS data. OPM and OMB should set an expectation at the CHCO Council that agencies will distribute these tools, such as the NIH tool they already have access to and follow up with agency units on their deployment. This will help managers identify areas for improvement and give them ways to work with their staffs.
Longer Term Improvements
Learn to Communicate (and Follow Through on) the Agency Needs Balanced with the Employee Wants through manager workshops. These trainings should encourage authentic conversations that balance agency resources and ability to change work environments and employee-initiated implementation initiatives. For example, if a source of employee dissatisfaction is a lack of adequate analytic tools and the agency does not have the resources to provide them, a dialogue addressing the options and timing is better than a manager making employees think they will get the resources soon.
Develop a Common and Interoperable FEVS Implementation Platform that tracks changes to agency structures and reporting relationships that serves as a central repository or clearinghouse for employee surveys and feedback results. The platform would assist with consistency across agency implementation and provide common elements for regular pulse surveys. This will decentralize FEVS to the agencies for implementation while preserving whole-of-government insights, cross-agency sharing and comparative data on agency performance.
Encourage Pulse Surveys and integrate these results into FEVS analytics and reporting. In addition to annual employee surveys, many public, private and non-profit organizations conduct pulse surveys to check on employee attitudes and perspectives in close to real time. These immediate checks give leaders real-time feedback on how employees are thinking about their work and the organization. They give managers the ability to act fast when needed. With care taken on an interoperable platform, pulse surveys can be rolled up to a composite picture at the unit, bureau, agency, and whole-of-government level.
Open FEVS Design and Outcomes to external researchers and academics who can help improve FEVS for the future. For example, in the eNPS example above, there appears to be little attempt to translate these results into data that maps to private sector insight benchmarks, such as the percentage of employees who identify, because of the scores, as “Promoters,” “Passives” or “Detractors” when it comes to recommending a colleague to work there.
Finally, and this may be unrealistic, resist the temptation to try to control the narrative and stop adding questions to the FEVS. FEVS is too long. Successive administrations have added FEVS questions to obtain employee attitudes on administration-specific policies or issues. Frequently, these additional questions are eliminated in the subsequent administration and its policies. Experts point to an inverse relationship between the time it takes/number of questions and completion rates. (As an example, see the paper straws discussion.) Political grandstanding on FEVS results, from either side of the aisle, is death to transparency. Attempts to control the narrative limit forthright action on the part of program leaders trying to improve performance on the ground. It breeds skepticism if not cynicism regarding the FEVS program, rendering the whole process counterproductive.
The utility of FEVS depends on a range of stakeholders and owners working together to administer the survey fast, get the results out to agency unit managers, help them interpret the results, and take actions that will improve program and agency performance. Agencies will not see the FEVS benefits in improved performance, innovation and efficiency without these stakeholders and owners working together. These include:
Governance & Policy: OPM, OMB, CHCO Council, Agency Leadership
Design, Administration & Communications: OPM, OMB, GAO
Distribution and Interpretation: OPM and Agency CHCOs/HR Operations
Interpretation and Action Planning: Agency Leaders and Managers
Oversight: GAO, Congress, IGs, Good Government Groups
However, federal employees are the most important owner and stakeholder. Their willingness to participate in FEVS and trust that something will be done with their answers (reflected in FEVS Question 47 at a low 49% in 2024), are an indicator of their willingness to assist in improving performance. As noted above however, the average FEVS response rate is 41%, so there is work to do to build trust in the survey and its results. As Yogi Berra said, “If people don't want to come out to the ballpark, nobody's gonna stop 'em.” Cancelling the 2025 FEVS will erode trust. Continuing FEVS for 2025 will improve engagement and communications; it will work toward uniting employees to develop solutions to resource constraints, and integrate technology effectively.
Effective leaders, political and career, talk with agency employees every day. They are checking in with employees as they make decisions, as they mentor, coach, communicate, and as they evaluate performance. This is the daily part of the employee experience performance feedback loop.
FEVS provides an additional composite feedback loop to this daily work. It adds to the performance scorecard for the enterprise – from the individual program unit up through the agency and the whole of government. It gives leaders an outside, statistically valid, objective source for this feedback loop, providing crucial information on employee attitudes, opinions, and perspectives at the agency unit, bureau/office, agency, and whole-of-government levels. FEVS gives policy makers and governance/oversight functions both a snapshot in time to assess agency functions as well as a time series across years to evaluate how well their policies and strategies are working.
Improving FEVS will lower the signal-to-noise ratio so that agency leaders and managers can make better performance decisions.
Let’s keep it going for 2025.
For monitoring of FEVS, check out the data collection on dataindex.us.
Peter Bonner is a Senior Fellow at the Federation of American Scientists. The observations and recommendations here come from interviews with principals and experts involved in FEVS and from work inside and outside of government on projects to implement FEVS results.