PerformYard · 2027 State of Performance Management
The performance review is getting smaller. Performance management is getting bigger.
The annual review is giving way to a steady stream of shorter check-ins, goals, feedback, and 1:1s. Organizations know more about how their people are doing, and know it sooner. The ones that gather signals most often have higher reviews completion percentages, better goal performance, and are the first to start using AI.
Executive summary
What we found, why it’s happening, and what to do
Performance management is turning from a once-a-year document into a steady flow of information about how people are doing.
Every check-in, goal, piece of feedback, and 1:1 is another data point, and the organizations collecting the most of them, most often, are the ones that spot problems while there is still time to act. As organizations mature they have more employee touchpoints, not less.
The rest of the market — the companies running reviews manually — has not caught up: most organizations we surveyed still run reviews once a year, mostly on spreadsheets, and grade their own process a C.
The gap is a process gap, which means it can be closed.
Treat performance data as a stream, not an annual snapshot.
There are four big takeaways here.
Cut the review form to 10 to 15 questions and keep the open-ended ones, so each check-in is quick and worth reading.
Add a second review cycle before changing anything else.
Give managers reminders and keep their spans under ten.
Decide where AI fits, because it is already summarizing all of this for the majority of organizations.
Six findings, and the reasoning behind each ones
Once an organization has a rhythm, each check-in is quick. Organization didn't learn much from the annual review, so they're leaning into a steady stream of reviews, goals, feedback, and 1:1s. This frequent feedback habit compounds, and it is the thread running through every finding below.
The numbers: Organizations in their first year with PerformYard run a median of 2 review cycles a year; by their fifth year, they're running 5 cycles per quarter.
Goals per person rise from 0.4 to 1.8 between the first and second year, and the typical written answer grows from 233 characters in the first-year to 432 by year six.
Long forms produce thin answers and late completions, so organizations are adding check-ins and shortening the form.
Asking less of people, but asking it more often, gets managers and employees to provide richer feedback..
The numbers: We looked at the the same 418 organizations from 2022 to 2026.
Review forms went from 17.7 to 14.4 questions, review cycles from 3 to 5 a year, and written answers grew 25% longer during that time frame..
Every touchpoint, whether a review, a goal, a piece of feedback, or a 1:1, is a reminder and a piece of context, so all the feedback and follow-through builds on itself.
We surveyed of organizations that don’t use PerformYard and found the same pattern.
The numbers: Organizations recording 8 or more touchpoints per employee a year complete 94% of their reviews.
Those recording one or fewer touchpoints complete 76% of reviews. Even companies that went from one review cycle to two cycles raised their review completion from 84% to 90%.
Frequent, lighter check-ins and more concrete goals give organizations more information about each employee over a year, and more chances to act on it.
The numbers: The 25% of organizations with the highest goal completion rates run 4 review cycles a year instead of 3, have 1.76 formal reviews per person instead of 1.25, and set 5× as many goals per person as the bottom 25%, on forms that are slightly shorter (15 questions instead of 17).
Managers hold the pen for every report. Some organizations have structures where the manager has 12+ direct reports. At that number, the load stops being manageable. Additionally, managers who rely on spreadsheets and PDFs run into serious issues with their review process.
The numbers: Asked why reviews don’t get finished, 45% of organizations surveyed say managers forget or deprioritize them.
Among PerformYard customers, goal completion rates fall 54% once managers average 12 or more direct reports.
PerformYard AI summarizes the information managers already have to read, and it arrived inside the platform as written answers were getting longer.
As a result, organizations using AI to assist in providing more feedback to more employees.
The numbers: 14.5% of organizations used PerformYard's AI Review Insights tool in 2025; 56% have in 2026, including 91% of large enterprise companies.
01 · Five years of change
Reviews are getting shorter and happening more often, and employees are writing longer answers.
Ask an HR leader what their performance review looks like, and most will describe a form. For the 418 organizations we followed from 2022 to 2026, that form has been shrinking (from 17.7 questions to 14.4) while the number of times a year it gets used has gone up (from 3 review cycles to 5 review cycles). Long annual forms gave companies thin answers and late completions, so companies are adding mid-year check-ins. With reviews happening more than once a year, each review needed fewer review forms.
The surprising part is what happened to the answers on each review form. Asked fewer questions, more often, employees wrote more: the typical written answer grew 25% over the period, and the organizations that have been at this longest write the most, 233 characters per answer in a first-year organization against 432 by year six. The review is turning from a once-a-year audit into a running record of how each person is doing. That is the direction to plan for.
The numbers We followed the same 418 organizations from 2022 to 2026. Their review forms went from 17.7 to 14.4 questions, their review cycles from 3 to 5 a year, and the answers employees wrote grew 25% longer.
Go deeper: the detail behind this chapter
The cleanest way to see change is to follow the same organizations. 418 organizations have run performance management on the platform in every year from 2022 through 2026.
What grew: how often reviews happen, and how much people write. Median cycles per year rose from 3 to 5, and the share of these organizations running three or more cycles went from 57% to 65%. The typical written answer grew from 323 to 402 characters, while open-ended questions rose from 28% to 33% of the form.
What shrank: the form, from 17.7 questions in 2022 to 14.4 in 2026. Across all customers the typical form has 15.5 questions in 2026, down from 17.0 in 2024, and organizations in their first year build the shortest forms of all (14.8 questions).
Answers grow with experience. Looking across organizations in 2026, the typical written answer runs 233 characters in an organization's first year on the platform, 334 in its third, 390 in its fifth, and 432 in its sixth. Form length barely changes over those years; what changes is how much people put into each answer once writing about performance has become routine.
Organizations are asking fewer questions, checking in more often, and getting longer answers back.
The longer they have been at running formal reviews, the more employees write: 233 characters per answer in an organization's first year with PerformYard, 334 by its third, 432 by its sixth. Over time, that adds up to more information about each employee arriving in smaller pieces all year long.
02 · What the best do differently
The organizations with the highest goal completion rates ask fewer questions, run reviews more often, and set far more goals.
It would be reasonable to guess that the organizations with the best results run the most straightforward programs. They don’t.
We ranked every organization that ran reviews and set goals in 2024 by its goal completion rate. The top quarter runs reviews as a rhythm rather than one event: 4 cycles a year instead of 3, 1.76 reviews per person instead of 1.25, on forms that are slightly shorter (15 questions instead of 17) yet come back with longer answers. Most of all, they set 5 times as many goals per person.
The various touchpoints reinforce each other. More goals give each review something concrete to cover, and more reviews keep goals alive. The ideal recipe is this: a 15-question review, extra review cycles each year, and three or more goals per person. Over a year it produces more information about more employee, and keeps goals top of mind.
The numbers Compared with the organizations with the lowest goal completion rates, the top quarter set 5× as many goals per person, use shorter reviews (15 questions instead of 17) and run 4 review cycles a year instead of 3.
The strongest programs are not lighter, they are more frequent: slightly shorter forms, used far more often, with more goals to talk about each time.
Go deeper: the detail behind this chapter
How we define the top and bottom tiers. We ranked every organization that ran reviews and set goals in 2024 by its goal completion rate. The top tier is the quarter with the highest goal completion rates that also completed at least 95% of their reviews. The bottom tier is the quarter with the lowest goal completion rates. Both groups run reviews and set goals. The difference is how.
Top-tier organizations run more review cycles (4 vs 3 a year) and more reviews per person (1.76 vs 1.25), yet their forms are shorter (15 vs 17 questions) and their answers longer (351 vs 286 characters). They set 5 times as many goals per person (3.5 vs 0.7), and 64% of them average two or more goals per person against 28% of the bottom tier. Every one of these gaps is wider in 2025.
Organizations whose employees set 6–10 goals a year report 85% review completion, against 77% where people set five or fewer. The survey also asked who sets the goals, something the customer data cannot see. Where goals are set together by manager and employee, review completion is 82%; where managers set them alone, 69%; where no goals are set, 61%.
For 2027: measure your program against the top tier: three or more review cycles a year, a form of about 15 questions, and three or more goals per person, set together with the employee.
03 · More conversations, more finished reviews
The more employee touchpoints you have, the more likely it is your reviews get completed.
Every review, goal, piece of feedback, or 1:1 meeting recorded for an employee is a touchpoint: a moment when someone paid attention to how that person is doing and wrote something down. Each touchpoint is also a reminder and a piece of context. A manager who set a goal and held a 1:1 has something to say in their review, and an employee who was asked for feedback is already in the habit of responding.
Follow-through and feedback builds on itself.
Organizations that record 8 or more employee touchpoints per employee a year complete 94% of their reviews; those that record one employee touchpoint complete just 76% of their reviews.
A separate survey of organizations that don’t use PerformYard shows the same phenomenon: those using three review or feedback touchpoints per year report 91% review completion, against 68% for those using one touchpoint.
For HR, touchpoints are the leading indicator. Touchpoints per employee is the best predictor of whether people complete your performance management process. If you're under two a year, you're likely letting people slip through the cracks. Four or more touchpoints, and you're likely ensuring all your employees get meaningful feedback.
The numbers We call each review, goal, piece of feedback, or 1:1 meeting recorded for an employee a touchpoint. Organizations that record 8 or more touchpoints per employee a year complete 94% of their reviews. Organizations that record one complete just k76%.
Each touchpoint is another data point about how someone is doing. A separate survey of organizations that don't use PerformYard shows the same thing from the other side: those using three kinds of review or feedback report 91% completion, against 68% for those using just one.
Go deeper: the detail behind this chapter
What counts as a touchpoint. A touchpoint is any recorded performance interaction with an employee during the year: a review they take part in, a goal they set, feedback they give, or a 1:1 meeting logged with them. Adding these up per employee measures how much performance management a person actually experiences. In 2024 the typical organization recorded 2.9 touchpoints per employee; a quarter of organizations recorded more than 6.5.
The pattern holds all the way up the scale. Organizations recording one touchpoint per employee or fewer in 2024 completed 76% of reviews; those recording eight or more completed 94%. Goal completion rates in the eight-or-more group are 1.7 times those of the lowest group. The 2025 data shows the same pattern. Reviews alone tell the same story: organizations running two to four reviews per person a year complete 93% of them, against 76% where reviews happen less than once every two years per person.
Respondents were asked which kinds of review and feedback they use regularly: annual reviews, quarterly conversations, 1:1 meetings, competency-based reviews, project-based reviews, and 360-degree feedback. Reported completion rises with the count: 68% with one kind, 76% with two, 91% with three, 94% with four. Only 49% of organizations using one kind say they complete more than 85% of reviews; 82% of those using three kinds do.
Using more of the system points the same way. 52% of customers used three or more modules (reviews, goals, feedback, engagement, meetings) in 2025. Review completion climbs from 82% for organizations using a single module to 93% for those using four, and goal completion rates are 1.7 times higher. These are correlations; the organizations that adopt more are likely those already more committed.
For 2027: count the touchpoints your program actually records per employee. Fewer than two a year is a program running on paper. The organizations at four or more finish the most.
04 · How often
Adding a second review cycle makes the biggest difference.
The most effective change that PerformYard customers make is also the smallest. Going from one review cycle a year to two raises the review completion percentage from 84% to 90%. A second cycle lowers the stakes of each review, turns a judgment into a check-in, and surfaces problems months earlier, while there is still time to address them. Companies that decide to run three or more cycles per year also get each review cycle finished 28% faster than those who just do one big cycle.
Increasing the number of reviews is the first step on a path organizations take as they mature: 2 cycles in their first year, 5 cycles by their fifth year. Among PerformYard customers, the once-a-year review is now the exception, only 17% of organizations that ran a cycle in 2025 ran just one. When we surveyed HR folks who don't use PerformYard, we found that the once-a-year-review is the norm.
If your company is running just one cycle per year, add a second review (like a mid-year check-in) before changing anything else.
The numbers Going from one review cycle a year to two raises the share of reviews completed from 84% to 90%. Organizations running three or more cycles a year also finish each review 28% faster than organizations running one.
Among PerformYard customers the once-a-year review is now the exception: only 17% of organizations that ran a cycle in 2025 ran just one. Outside the platform it is still the norm.
Go deeper: the detail behind this chapter
Among organizations that started a review cycle in 2025, 68% started three or more, while 17% ran exactly one. Frequency pays off: organizations running three or more cycles in 2024 completed 91% of reviews and finished each one 28% faster than single-cycle organizations, which completed 84%. Two cycles already lift completion to 90% and speed up reviews by 6%.
Frequency also grows with experience. Among 2024 company-years, first-year organizations ran a median of 2 cycles; organizations in their fifth year or later ran 5. Goals per person jump from 0.4 in the first year to 1.8 in the second, and reviews per person rise 31% within the same organizations between their first and second year.
52% of surveyed organizations give employees formal feedback once a year, and 22% use nothing but an annual review. Those annual-only organizations report 67% review completion, against 81% for everyone else. Moving to twice-yearly feedback lifts reported completion from 76% to 84%; quarterly and monthly cadences do not report higher completion than twice a year. Organizations that hold quarterly conversations report 87% completion, against 74% for those that do not.
For 2027: if you run one review cycle, add a mid-year check-in before you change anything else.
05 · The review form
Shorter review forms get finished. Long ones cost a week.
Every extra question on a review form is a small tax on the manager who has to answer it. Long forms get put off, then rushed, and open-ended questions end up competing for the same attention as quick multiple choice questions.
The tax shows up in completion percentages: when forms have 10 or fewer questions, 91% of reviews get finished; with 26 or more questions, just 85%.
The tax shows up in the answers too: with one or two open-ended questions the typical answer runs 376 characters; with eleven or more open ended questions, that drops to 241 characters.
When we surveyed people who aren't using PerformYard, we found that the cost is measured in time. Among organizations surveyed, 64% of those with forms of 16 or more questions say HR and managers spend a full week or more on every cycle. Form length is the lever HR controls completely. Aim for 10 to 15 questions with three to five open-ended prompts, so each check-in is quick and worth reading.
The numbers When review forms have 10 or fewer questions, 91% of reviews get completed. With 26 or more questions, 85%. And among organizations surveyed, 64% of those with forms of 16 or more questions say HR and managers spend a full week or more on every review cycle.
Ask fewer questions and you get better answers: with one or two open-ended questions the typical answer runs 376 characters; with eleven or more, 241 characters. Aim for a focused form of 10–15 questions, not the shortest form possible.
Go deeper: the detail behind this chapter
Length has a cost. When review forms carry 10 or fewer questions, 91% of reviews were completed in 2024; when they carry 26 or more, 85%. Open-ended questions also dilute one another: with one or two of them, the typical answer runs 376 characters; with eleven or more, 241, a drop of 36%.
Among surveyed organizations with review forms of 16 or more questions, 64% say HR and managers spend a full week or more on every review cycle; among those with forms of 6–10 questions, 28% do. Reported completion is also lowest for the longest forms (74% for 16 or more questions) and highest at 11–15 questions (83%). One nuance the survey adds: forms of five or fewer questions, most common among organizations running reviews on spreadsheets, do not get completed more often (73%).
For 2027: aim for a form of 10–15 questions with three to five open-ended prompts.
06 · Managers
Reviews stall when managers are stretched.
Performance reviews stall with managers, and not because managers resist them. Managers hold the pen for every direct report, and past about a dozen reports the writing load stops being manageable: where managers average 12 or more direct reports, goal completion rates are 54% lower than where they average 3 to 5. Outside PerformYard, the problem is memory rather than load. We asked people why reviews don’t get finished, and 45% of organizations said managers forget or deprioritize the reviews. That was a much more common answer than blaming a manual process or employee resistance.
Design the program around the manager’s calendar. Keep spans of control under ten, and send reminders before the cycle closes. 360-feedback from colleagues lightens the load too: at companies where 15–30% of employees gave feedback, 93% of reviews were completed, against 85% where only managers are giving feedback.
The numbers Where managers average 12 or more direct reports, goal completion rates are 54% lower than where they average 3 to 5. And when organizations outside the platform were asked why reviews don't get finished, 45% said managers forget or deprioritize them.
Performance management works to the extent managers have the time and the reminders to do their part.
Go deeper: the detail behind this chapter
The typical organization averages 4 direct reports per manager; 43% sit between three and five, and 10% average more than eight. Taking organizations at three to five reports as the benchmark, goal completion rates are essentially level through eight reports, slip 15% at eight to twelve, and fall 54% where managers carry twelve or more. The 2024 data shows the same cliff.
Feedback from colleagues helps too. Where 15–30% of employees gave feedback in 2024, 93% of reviews were completed, against 85% where no one gave feedback. But participation is thin: 55% of organizations used continuous feedback in 2025, and in the typical one only 4.3% of employees gave any.
Asked for the single biggest reason reviews are not completed on time, 45% of surveyed organizations say managers forget or deprioritize them, far ahead of too much manual work (14%) or employees not seeing the value (13%). 28% say completion is not a problem for them, and those organizations report 89% completion, against 75% where managers are the blocker.
For 2027: flag any manager above ten direct reports, and give every manager a reminder before the cycle closes. Forgetting, not resistance, is what stalls reviews.
07 · AI
AI went from experiment to default in one year.
When an organization is collecting reviews, 1:1 notes, and survey answers all year, someone has to make sense of it all. PerformYard's AI Review Insights tool helps with that job, and it went from experiment to default in a single year.
14.5% of organizations used this tool regularly in 2025. By September 2026, 56% of organizations were using the tool regularly, including 91% of organizations with 750 or more employees.
The largest organizations, with the most to read, adopted AI first, and new customers now switch it on as part of setup (66% of first-year organizations use it AI Review Insights). The organizations are also using AI-Assisted Reviews to write longer review answers, not shorter ones. The question for 2027 is not whether AI belongs in reviews but where: it's used to help managers prepare, is correlated with higher review completion, and can help flag issues that needs attention before the review cycle ends.
The numbers 14.5% of organizations used AI Insights in 2025. By September 2026, 56% had, including 91% of organizations with 750 or more employees.
When an organization is collecting reviews, 1:1 notes, and survey answers all year, someone has make sense of it all. That is the job PerformYard AI helps accomplish.
Go deeper: the detail behind this chapter
Adoption rises with size: 38% of organizations under 50 employees, 84% of those above 250, and 91% of those above 750. It is also highest among the newest customers, with 66% of organizations in their first year using it against 44% of those in their second. AI is now part of how performance programs are set up rather than something added later.
What does AI use go with? Organizations using AI Insights in 2025 completed 94% of reviews against 88% for non-users, their goal completion rates were 30% higher, and their written answers ran 367 characters versus 314. Some of that gap is selection: larger, more engaged organizations adopted first. Following the same organizations over time gives a more modest picture: those that adopted AI Insights in 2026 grew their typical answer length by 9.6% from 2025, compared with 7.5% for organizations that have not adopted.
For 2027: the question is no longer whether to use AI in reviews but where. Summaries that help managers prepare are already mainstream; the next step is AI that flags what needs attention before the cycle ends.
08 · 1:1 meetings
The weekly 1:1 with an agenda is the meeting that produces action.
The weekly 1:1 is the most efficient tool for learning how employees are doing. The agenda is the tool inside the meeting that makes the meeting productive: meetings with two or more agenda items generate 5 times as many action items as meetings with 1 or 0 agenda items.
Regular 1:1s carry a running record into the formal review, so the review becomes a summary rather than a reconstruction of the year. Organizations that run 1:1s finish their reviews 36% faster than those that don't.
The numbers Meetings with two or more agenda items produce 5× as many action items as meetings with almost none.
Organizations that run 1:1s on PerformYard finish their reviews 36% faster than those that don't. When the conversation happens every week, the formal review has less to catch up on.
Go deeper: the detail behind this chapter
The typical cadence is weekly: the median gap between meetings is 6.8 days, and half of organizations fall between 3 and 12 days. The strongest predictor of whether a meeting produces follow-up is whether it had an agenda. Meetings with two or more agenda items generate a median 1.45 action items; meetings with almost no agenda generate 0.27. Most meetings still travel light, with a median 1.1 agenda items and 1.0 action item each.
Regular 1:1s go with the rest of the performance process. Organizations that ran 1:1s in 2025 finished reviews 36% faster than those that did not, with higher completion (94% vs 90%), and in 2026 73% of them use continuous feedback against 33% of everyone else.
For 2027: make the weekly 1:1 the unit of performance management, and give every meeting an agenda. Two items are enough to turn a check-in into commitments.
09 · Size and industry
Bigger organizations complete more reviews, not fewer.
Small organizations are often assumed to be nimbler at performance management. The data says otherwise. Organizations under 50 employees completed 84% of their reviews in 2024; those with 251–749 employees completed just under 95%, and those with 750 or more completed just over 95%. Larger organizations do have more HR capacity and more refined processes, and they also run more review cycles.
As organizations grow, they gather more performance information, more regularly, and it shows in their follow-through. Small organizations would be wise to borrow this structure even if they don't have the headcount.
The numbers Organizations under 50 employees complete 84% of their reviews. Organizations with 251–749 employees complete just under 95%, and those with 750 or more complete just over 95%.
Small organizations are often assumed to be nimbler at this. The data says larger organizations, with more structure and more review cycles, follow through on reviews more reliably.
Go deeper: the detail behind this chapter
Larger organizations run more cycles and lean on more structure, while small ones build richer individual reviews (2.0 forms per review under 50 employees vs 1.8 at 750 or more).
Industry benchmarks
Industries with at least 20 organizations, 2025 medians (AI adoption is 2026).
10 · Outside the platform
Most organizations still run the once-a-year, spreadsheet review, and grade themselves a C.
Step outside the PerformYard platform and the picture is about a decade in the past. Among organizations surveyed that don’t use PerformYard, 51% run reviews on spreadsheets, PDFs, or paper; 52% give formal feedback once a year; and 43% gather only one kind of performance signal at all.
A third of these organizations spend a week or more of HR and manager time on each cycle, and 87% grade their own process a C or D. Spreadsheets have no reminders and no visibility, so reviews happen once a year and depend on memory, and one annual form is about all a spreadsheet can handle.
Your software tools shape how much your organization can learn about its people. Half of spreadsheet users run one review per year and they're only completing about 74% of that annual review. The gap between the market and the customers measured in this report is a process gap, which means it can be closed.
The numbers Among organizations surveyed that don't use PerformYard, 51% run reviews on spreadsheets, PDFs, or paper; 52% give formal feedback once a year; 34% spend a week or more of HR and manager time on each review cycle; and 87% grade their own process a C or D.
The tool shapes how much an organization can learn about its people. Organizations on performance software review employees more often, complain far less about manual work, and finish more of their reviews.
Go deeper: the detail behind this chapter
The organizations that took PerformYard's online Performance Management Grader are a window on performance management beyond any one platform. 96% of them manage reviews without PerformYard, and just over half do it on spreadsheets, PDFs, or paper. Their answers are self-reported, and they chose to grade themselves, so they describe organizations curious about their own process; even so, the picture is consistent.
The tool shapes how much gets collected. 43% of all surveyed organizations use just one kind of review or feedback. Among spreadsheet users it is 50%; on an HRIS, 31%; on performance software, 36%. 41% of performance-software users gather three or more kinds, against 30% of spreadsheet users.
The tool shapes the process. Organizations on dedicated performance software report 88% review completion; HRIS modules, 82%; spreadsheets and paper, 74%. Spreadsheet-based organizations are far more likely to review once a year (64% vs 31% on performance software), twice as likely to use nothing but an annual review (31% vs 16%), and more than twice as likely to name manual work as their biggest problem (20% vs 9%). The few PerformYard customers who took the grader are excluded from these comparisons.
Time spent on reviews pays off. A third of organizations spend a week or more of HR and manager time on every cycle, and they report higher completion than organizations spending less than a day. The customer data shows what that time buys when the process is frequent: organizations running three or more cycles finish each review 28% faster, and organizations running 1:1s finish 36% faster. Outside the platform, the week is being spent on the annual form.
11 · Received wisdom
Six things people believe about performance reviews that the data does not support
- "A thorough review is a long review." When review forms have ten or fewer questions, 91% of reviews get completed; with 26 or more, 85%. The organizations with the highest goal completion rates use 15-question forms and get 351-character answers; those with the lowest use 17 questions and get 286.
- "Fewer goals means more focus." Organizations whose people set five to ten goals a year complete 2.1 times the share of their goals that organizations averaging under one goal per person do. The organizations with the highest goal completion rates set 3.5 goals per person; those with the lowest, 0.7.
- "Small companies are nimbler at this." Organizations under 50 employees completed 84% of their 2024 reviews; organizations of 251–749 completed 95% and ran more than twice as many review cycles.
- "The annual review is dead." Among PerformYard customers, 17% of organizations that ran a cycle in 2025 ran exactly one. Outside the platform it is very much alive: 52% of organizations surveyed give formal feedback once a year, and 22% use nothing but an annual review. Those annual-only organizations report the lowest completion in the survey (67%).
- "Continuous feedback has arrived." 55% of organizations use it, but in the typical one only 4.3% of employees gave feedback in 2025. Where 15–30% of employees give feedback, 93% of reviews get completed, against 85% where none do.
- "AI will hollow out the written review." Organizations using AI Insights write longer answers, not shorter: 367 characters versus 314 in 2025, and the typical answer across the platform has grown every year since 2024.
12 · What to do
Four moves for HR leaders shaping 2027
- Cut the form, keep the open-ended questions. Ten to fifteen questions with three to five written prompts. When forms have 26 or more questions, 85% of reviews get completed; with ten or fewer, 91%.
- Add a second review cycle before you change anything else. Going from one cycle to two raises the share of reviews completed from 84% to 90%; three or more cycles finish each review 28% faster. Organizations recording 8 or more touchpoints per employee complete 94% of reviews versus 76% for one or fewer.
- Set a feedback target, and remind the manager. Getting 15–30% of employees to give feedback goes with 93% review completion versus 85% with none. And 45% of organizations surveyed say a manager forgetting is the main reason reviews stall.
- Decide where AI fits, because it is already here. 56% of organizations used AI summaries this year. Put it where managers prepare, then let it surface what needs attention.
About the data
Two sources, kept separate
What organizations actually do. Anonymized, aggregated PerformYard product data exported September 9, 2026: 8,191 company-years across 2,553 active customer organizations, 2014–2026. Industry comes from account records and is known for 98% of organizations; 34 industries have at least 20 organizations and appear in the benchmark table. Figures are medians across organizations unless stated.
What the wider market says it does. 300 organizations completed PerformYard's online Performance Management Grader between late 2024 and September 2026, answering nine questions about how they run reviews, goals, and feedback. 96% manage reviews on spreadsheets, an HRIS, or other software rather than PerformYard; 58% of those who gave a title work in HR or people roles. Answers are self-reported. Survey findings appear in amber panels and in chapter 10, and are never mixed with customer data.
Definitions, exclusions, and notes
Review completion. Share of an organization's reviews started in the year that are marked complete, as of September 9, 2026. Because reviews continue to be completed after a year ends, 2024 is the primary year for this measure and recent years are not compared with earlier ones. Means are used for review completion because most organizations sit near 100%.
Time to finish a review. Median days from review start to completion for reviews started in the year, shown only as percentage differences between groups.
Goal completion rate. Share of goals started in the year that were completed, as of September 9, 2026. Shown as relative comparisons or indexes. Values above 100% (numeric goals that exceeded target) are capped at 100%; values that are impossible by definition are excluded.
Top and bottom tiers. We ranked every organization that ran reviews and set goals in 2024 by its goal completion rate. The top tier is the quarter with the highest goal completion rates that also completed at least 95% of their reviews. The bottom tier is the quarter with the lowest goal completion rates.
Goals per person, reviews per person, feedback participation. Annual counts divided by active users. Feedback participation is the share of users who gave at least one piece of feedback in the year.
Touchpoints per employee. Reviews per person + goals per person + feedback participation rate + 1:1 meetings per person, recorded in the platform that year.
Questions and open-ended questions per form; answer length. Based on the form versions used and the answers submitted in the year; written answers are measured in characters.
Cycles. Review cycles started in the year. The cadence chart is based on organizations that started at least one cycle.
AI Insights. Uses of the AI Insights feature, which summarizes results from reviews, meetings, and surveys. 2026 figures cover January 1–September 9.
Year with PerformYard. Calendar year minus the first year the organization appears in the data, plus one.
Organization size. Active users: under 50, 51–100, 101–250, 251–749, 750 and over.
Practitioner survey. PerformYard's online Performance Management Grader, 300 responses exported from HubSpot on September 10, 2026 (the grader has been live since late 2024; submission dates are not exported). Nine multiple-choice questions on current tool, feedback cadence, review types used, goals per person, who sets goals, form length, completion rate, biggest blocker, and HR/manager time per cycle. Completion is the respondent's exact slider value, recovered from the grader result. All answers are self-reported and respondents are self-selected. Survey groups are reported at 20 or more respondents. The 11 respondents who manage reviews in PerformYard are excluded from tool comparisons because the grader assigns them a fixed grade and the group is too small. Employee counts and industry come from HubSpot company records. No names, emails, or identifiers were used in the analysis.
Sample sizes and exclusions. Every group reported has at least 30 organizations, except industry cuts (minimum 20) and survey groups (minimum 20). Company-years with fewer than five users are excluded, as are direct-report ratios above 30, goals-per-person values above 50, and negative or multi-year durations. Outcome measures use 2024, the most recent year with enough time to settle; design, cadence, and adoption measures use 2026 (January 1–September 9). Patterns described are associations, not causal effects.
PerformYard 2027 State of Performance Management. Customer data: anonymized, aggregated product data as of September 9, 2026. Survey: PerformYard Performance Management Grader, 300 responses, self-reported. Figures are medians unless stated. Patterns described are associations, not causal effects.