The Science of Management Cadence

Why evidence-based management practices compound when they operate in one system

It’s 9:04 on a Tuesday morning when the resignation lands. The engineer leaving is the strongest person on a team that has scored green on every dashboard for a year. Now the reconstruction begins: HR pulls exit-survey history from one tool, goal records from another, 1:1 notes from a manager’s private doc that turns out to be empty since March, a recognition feed nobody checked, and an ER case from last spring that was closed as “resolved.” Every one of those artifacts was a warning. Each lived in a different system, so nobody saw the pattern until the person was already gone.

There is a name for this way of running an organization: management by archaeology. You find out what happened to your people by excavating the fragments afterward. Most companies don’t choose it — they assemble it, one point tool at a time, until the signals that should converge are guaranteed never to meet.

Cadence exists to replace archaeology with cadence: a recurring operating rhythm where the practices management science already validates — structured conversations, clear goals, fair process, recognition, listening, role clarity, evidence-based people decisions — run in one connected plane. AI’s role in that plane is specific and bounded: AI develops managers, not replaces them.

This document lays out the research behind each practice, what Cadence does and does not claim, and why we believe connected practices beat isolated ones.

Key takeaways

1. Feedback improves performance on average, but it is not automatically good: over one-third of feedback interventions in the foundational meta-analysis reduced performance — design matters. [1]
2. Specific, challenging goals reliably outperform vague “do your best” direction — when commitment, ability, and feedback are present — and goal systems have documented failure modes when they aren’t. [2] [4]
3. Team engagement is consistently associated with productivity, profitability, retention, customer satisfaction, and safety across the largest business-unit meta-analyses ever conducted. [5] [6]
4. Managers are the single biggest lever: Gallup’s State of the American Manager found managers account for at least 70% of the variance in team engagement scores across business units. [7]
5. Fair process is measurable: procedural, interpersonal, and informational justice each independently predict satisfaction, commitment, citizenship, and withdrawal. [8]
6. People practices work as systems: meta-analytic evidence shows bundles of skill-, motivation-, and opportunity-enhancing HR practices relate to human capital, retention, and operational and financial outcomes. [22]
7. Cadence’s compounding argument — that connected management signals create context isolated tools cannot — is our reasoned synthesis of this evidence, clearly labeled as an integration thesis, not yet a Cadence customer-outcome study.

Who this is for

• HR and People Operations leaders evaluating whether the practices a management platform encodes have independent evidence behind them.
• Executives deciding whether to fund a system of management rather than another point tool.
• Investors and diligence teams testing whether Cadence’s claims are calibrated to what the research actually supports.
• Skeptics — especially the ones who suspect vendor “research-backed” pages of laundering marketing through citations. The strongest objection we know of is steelmanned below, in its own section.

What this document is — and is not

This is a research synthesis, not a customer outcome claim. It cites the science behind the management concepts Cadence operationalizes. It does not claim that Cadence customers have achieved measured improvements, because Cadence does not yet have customer outcome data.

The central compounding argument is Cadence’s reasoned point of view: when validated management practices are connected in one operating plane, leaders can see patterns and act earlier than they can with disconnected point tools. That synthesis claim remains labeled as Cadence’s argument until customer data or independent evaluation proves it.

What do HR buyers actually need to know?

The practical question from HR buyers is fair:

Is there independent research behind the practices Cadence operationalizes, and behind the argument that those practices become more useful when connected?

The answer is precise:

1. There is independent research behind the practices Cadence operationalizes.
2. Cadence does not yet claim customer-proven efficacy metrics.
3. Cadence’s distinctive argument is integration: the system connects practices that companies usually run in separate tools, review cycles, spreadsheets, and inboxes.
4. That integration creates a better management signal because it shows convergence: missed 1:1s plus declining survey sentiment plus a recognition deficit plus goal stagnation plus ER recurrence tells a different story than any one signal alone.

Hybrid work raises the stakes. Microsoft’s 2022 Work Trend Index documented the trust gap behind “productivity paranoia”: 85% of leaders said hybrid work made it harder to be confident people were productive, while 87% of employees reported they were productive. [13] Gallup’s hybrid-work research reaches the same operating conclusion: as work becomes more flexible, managers need more explicit communication about priorities, progress, and handoffs. [14] The management model that depended on proximity is gone; the replacement has to be built deliberately.

The evidence map

Do structured 1:1s actually improve performance?

The 1:1 is Cadence’s operating system because management quality shows up in repeated conversations: priorities, blockers, performance evidence, recognition, coaching, accountability, and development all pass through the manager-employee relationship.

The research base supports the concept, with an important caveat. Feedback is not automatically good. Kluger and DeNisi’s feedback-intervention meta-analysis found positive average effects, but also found that more than one-third of feedback interventions reduced performance. [1] The lesson is not “give more feedback.” The lesson is “make feedback task-relevant, timely, specific, and connected to action.”

Cadence’s product interpretation:

• Create a structured agenda before the conversation.
• Pull in goals, prior action items, role expectations, and recent telemetry.
• Separate private coaching from shared notes so both manager and employee can improve without turning every meeting into surveillance.
• Commit action items forward so the conversation has memory.

Hybrid work makes this more important. Microsoft reported that 81% of employees said it was important for managers to help prioritize workload, while less than one-third said managers had ever given clear one-on-one guidance. [13] Gallup likewise notes that hybrid and remote arrangements increase the need for managers to communicate explicitly about priorities, progress, and handoffs. [14]

A 1:1 calendar invite does not improve performance by itself. The evidence supports the underlying mechanism — clear, task-relevant feedback and regular manager communication — and Cadence operationalizes that mechanism by making the 1:1 structured, remembered, and connected to the rest of the people system. The AI in the room prepares the manager, remembers the commitments, and surfaces the patterns. It develops the manager. It does not replace them.

For the practitioner’s view, see How to Run Effective 1:1s and Why 1:1s Fail.

Does goal setting really work — and when does it backfire?

Goal-setting theory is one of the strongest foundations in management psychology. Locke and Latham summarized decades of evidence showing that specific, challenging goals lead to higher performance than vague goals, provided important moderators are present: goal commitment, ability, feedback, and manageable task complexity. [2] [3]

The boundary conditions deserve equal billing, because they are where real-world goal systems fail. The same research tradition is explicit that goal effects depend on those moderators — and critics have documented the systematic side effects of over-prescribed goal setting: narrowed focus, short-termism, risk-taking, and even unethical behavior when targets are rigid, rewards are high-stakes, and measurement is thin. [4] A goals module that ignores this becomes the problem it was meant to solve.

That maps directly to Cadence’s goals module:

• Goals are role-linked, not detached annual paperwork.
• Progress appears inside the 1:1, where the manager and employee can inspect evidence — the feedback moderator, built into the rhythm.
• The system tracks trend, not just a point-in-time rating.
• Missed goals become coaching and accountability prompts, not delayed surprises — and not automatic punishments.

The danger in many OKR implementations is ritual without operating value: teams write goals, then manage around them. Cadence’s point of view is that goals only become management infrastructure when they are tied to the recurring conversations where tradeoffs, priorities, support, and accountability happen.

The science supports specific, challenging goals paired with feedback and commitment — inside their boundary conditions. Cadence’s implementation keeps goals alive inside the management rhythm, rather than stranded in a quarterly planning artifact.

Related: Connecting Goals, Feedback, and Recognition.

Is the 9-box grid evidence-based, or corporate astrology?

The 9-box is a widely used talent-review framework: current performance on one axis, future potential on the other. It is useful because it separates “doing the current job well” from “ready for larger or different responsibility.” AIHR describes the 9-box as a common tool for talent management and succession planning. [15]

The risk is equally important. 9-box processes can harden subjective labels, reproduce bias, and create false precision if the input data is weak or the criteria are unclear. Research on data combination in selection contexts supports structured use of evidence over purely holistic judgment, while also underscoring that this does not make automated labels inherently fair or correct. [9] And because calibration outputs feed employment decisions — promotion, development investment, performance intervention — they sit inside the reach of adverse-impact scrutiny: under the EEOC’s Uniform Guidelines framework, selection procedures that disproportionately screen out protected groups require validation and monitoring, and a talent-review process is not exempt from that discipline just because it happens in a workshop instead of a hiring funnel. [10]

That is why Cadence does not position 9-box as an algorithmic truth machine. It is a structured human calibration process with better evidence and better checks:

• L1 and L2 rate concurrently and blind to reduce anchoring.
• Prior ratings are visible for trend, not destiny.
• Gaps between L1 and L2 become discussion prompts and potential bias signals.
• Calibration outputs trigger different management actions: development, retention, mobility, recasting, or structured performance intervention — each owned by a human.

The defensible claim is not that every 9-box label is objectively true. It is that talent decisions become more disciplined when organizations use shared criteria, multiple perspectives, trend evidence, and calibration process instead of unstructured manager opinion.

9-box talent calibration in Cadence is roadmap (Coming Q3), and this page will keep saying so until the customer-facing workflow is live. For the practice itself, see the 9-Box Talent Calibration Guide.

What makes an employee-relations process fair — and defensible?

ER is where management quality meets fairness, trust, and legal defensibility. The relevant science is organizational justice: whether employees perceive outcomes, processes, explanations, and interpersonal treatment as fair.

Colquitt and colleagues’ meta-analytic review found that distributive, procedural, interpersonal, and informational justice are related but distinct, and each contributes to workplace outcomes such as job satisfaction, organizational commitment, citizenship behavior, withdrawal, and performance. [8]

Cadence’s product interpretation:

• Standardize the case lifecycle.
• Document investigation framework, employee statement, action level, follow-through, and resolution.
• Track outcomes, including recurrence and whether an intervention recovered performance or simply moved a person toward exit.
• Keep confidentiality boundaries clear.

This is especially important in hybrid organizations because informal context is thinner. If the manager’s process is inconsistent or poorly documented, the organization has less ambient evidence to fall back on.

Cadence operationalizes the fairness conditions that ER research says matter: consistent process, clear explanation, respectful treatment, and auditable follow-through.

Does employee recognition actually matter, or is it a perk?

Recognition is not a perk layer. It is a management signal. Gallup’s workplace research emphasizes that effective recognition is honest, authentic, and individualized [11], and Gallup’s Q12 framework treats recognition and development as core employee needs tied to engagement. [12]

Cadence is deliberately careful with the “5:1 rule.” The 5:1 ratio is a useful product heuristic for manager coaching: if the only documented touchpoints are negative, the relationship is probably unhealthy or under-managed. But no universal scientific law says five recognitions neutralize one negative action in every workplace, and Cadence will not pretend otherwise.

Cadence’s product interpretation:

• Capture peer, manager, and formal recognition.
• Make recognition visible as part of the employee’s performance context.
• Detect recognition deficits and uneven patterns across managers.
• Use the ratio as a coaching signal, not a compensation formula.

Recognition research supports timely, meaningful acknowledgment as part of engagement. Cadence’s 5:1 ratio is a management-health heuristic that helps leaders notice whether documented interactions are balanced — not a claim of universal causal math.

Do engagement surveys predict business outcomes?

Surveys work when they are part of a listening-and-action loop. They fail when they become annual theater: ask employees for feedback, publish a score, and do little with the signal.

The evidence base here is among the largest in applied psychology. The foundational peer-reviewed study — Harter, Schmidt, and Hayes’ meta-analysis of 7,939 business units across 36 companies — found generalizable relationships between employee satisfaction/engagement and business-unit outcomes including customer satisfaction, productivity, profit, turnover, and accidents. [5] Gallup has continued that research program: the Q12 meta-analysis is now in its 11th edition, the largest study of its kind, tying team engagement to eleven performance outcomes across many tens of thousands of business units. [6]

Two honest qualifications belong next to those numbers. First, managers matter enormously to the signal: Gallup’s State of the American Manager found that managers account for at least 70% of the variance in team engagement scores across business units — which is why Cadence treats engagement data as a manager-development input, not a leaderboard. [7] Second, these are meta-analytic associations: strong, replicated, and consistent in direction, but engagement research cannot randomize managers to teams, and causality plausibly runs in both directions — engaged teams perform, and performing teams engage. Gallup’s own meta-analytic work addresses causal ordering with longitudinal designs, and the association survives; still, Cadence’s position is that engagement scores are a signal for management attention, not a lever that mechanically converts survey points into profit.

Cadence’s product interpretation:

• Protect individual anonymity while surfacing aggregate team and department patterns.
• Connect survey decline with other manager-area signals: 1:1 quality, ER cases, recognition patterns, turnover risk, and goal stagnation.
• Give L2 and People leaders specific coaching and intervention prompts.

Engagement measurement is useful when it drives action. Cadence’s survey module connects aggregate sentiment to manager coaching and organizational intervention, rather than merely reporting a score.

Related: Measuring Culture Without Survey Fatigue.

Why does job architecture matter for managers?

Management gets weaker when role expectations are vague. Employees cannot self-correct against an invisible job, and managers cannot coach consistently against a moving target. Role-clarity research distinguishes goal and process clarity as measurable parts of role ambiguity and links clarity to workplace antecedents and consequences. [16]

The relevant foundation is job analysis and competency modeling. O*NET describes job content in terms of tasks, work activities, skills, knowledge, abilities, and other worker characteristics. [17] SIOP’s personnel-selection principles treat job/work analysis as a foundation for defensible employment decisions. [18] Campion and colleagues define competency models as collections of behaviors, knowledge, skills, abilities, and other characteristics necessary for effective job performance, and describe how they support development, selection, promotion, and career planning. [19]

Cadence’s product interpretation:

• Keep role expectations visible during 1:1s.
• Tie goals and calibration to the role, not to generic impressions.
• Support role recasting when a person is misfit but the work still matters.
• Support role creation and mobility when a top performer has outgrown the current box.

Cadence uses job architecture as the shared reference layer for management: what the role requires, what growth looks like, and what evidence should inform coaching and talent decisions.

Related: Job Architecture From Scratch.

Does connecting people data actually help — or just make more dashboards?

This is where Cadence’s argument becomes distinctive. The external research supports the individual mechanisms. Cadence’s synthesis is that the mechanisms become more powerful when they are connected.

People analytics is not valuable because it produces more charts. CIPD defines people analytics as using people data, HR-system data, and business information to solve business problems and improve evidence-based decision-making. [20] Evidence-based management similarly emphasizes translating the best available evidence into organizational practice. [21]

Strategic HR research also supports thinking in systems rather than isolated practices. Jiang, Lepak, Hu, and Baer’s meta-analysis examined HR systems through skill-enhancing, motivation-enhancing, and opportunity-enhancing practices and found relationships with human capital, motivation, turnover, operational outcomes, and financial outcomes. [22] That does not prove Cadence’s product efficacy. It does support the broader idea that people practices often work as systems.

Cadence’s synthesis:

• A missed 1:1 is a weak signal.
• A missed 1:1 plus stagnant goals is a stronger signal.
• A missed 1:1 plus stagnant goals plus declining engagement plus low recognition plus ER recurrence is a management pattern.
• A management pattern is actionable — by a human manager, with AI holding the wide view. AI develops managers, not replaces them.

That is the compounding effect: not “eight modules create eight times the value,” but “connected practices create context that isolated practices cannot.”

Management intelligence compounds when signals are connected across the actual places management happens: conversations, goals, calibration, recognition, surveys, ER, and role expectations. This is an integration thesis grounded in evidence-based management and HR-systems research — not yet a Cadence-specific outcome study, and we will keep saying so until the outcome study exists.

The strongest objection: doesn’t connecting weak signals just compound noise?

Steelman the skeptic’s case, because it is a good one:

Each of these signals is noisy. A missed 1:1 might mean a manager crisis — or a public holiday. A dip in pulse scores might be disengagement — or a hard quarter everyone already understands. If you wire noisy signals together and paint them on one dashboard, you don’t get insight; you get correlated noise with an authority it hasn’t earned. Worse, managers may start performing the metrics — holding hollow 1:1s and spraying recognition — to keep their tiles green. Goodhart’s law does not stop applying because the dashboard is well-designed.

Three honest responses:

1. Convergence is the point, not any single signal. The failure mode of point tools is acting on one noisy signal — or ignoring all of them. Requiring multiple independent signals to converge before flagging a pattern is a noise-reduction strategy, the same logic that makes structured evidence combination outperform single-cue judgment in selection research. [9] One flag is trivia; five aligned flags from five different mechanisms is a pattern with a low false-positive rate.
2. The system routes to humans, not to actions. Cadence surfaces patterns as prompts for a manager or People leader to investigate — it does not auto-label people or auto-trigger consequences. A false positive costs a conversation, not a career. That asymmetry is deliberate, and it is the practical meaning of “AI develops managers, not replaces them.”
3. Gaming is real, and observable. When managers perform rituals to keep metrics green, the cross-module view is precisely where the performance shows: recognition volume without specificity, 1:1s logged with empty agendas, goals updated in bursts before review cycles. A single-metric tool cannot see its own gaming; a connected plane at least has the evidence to catch it. We do not claim immunity to Goodhart’s law — we claim better instrumentation for noticing it.

What would change our mind: if deployment data showed convergence patterns firing mostly on noise, or manager behavior distorting to the metrics faster than coaching corrects it, the thesis fails its own test and we would say so. That is the standard an evidence-based company has to hold itself to.

What Cadence should not claim yet

Do not claim:

• “Cadence improves productivity by X%.”
• “Cadence reduces turnover by Y%.”
• “Cadence’s 5:1 ratio is scientifically proven to optimize management.”
• “Cadence’s AI detects bias objectively.”
• “Cadence’s 9-box ratings are more accurate than human judgment.”
• “Cadence’s ER module reduces legal risk by a quantified amount.”

Safer alternatives:

• “Cadence operationalizes practices with strong evidence behind them.”
• “Cadence connects signals that are usually fragmented across point tools.”
• “Cadence is designed to help leaders act earlier and coach more consistently.”
• “The compounding effect is Cadence’s synthesis: better context from connected management data.”
• “Customer outcome studies are a future validation priority.”

The argument in one paragraph

Management effectiveness is not one behavior. It is a system of repeated behaviors: setting expectations, checking progress, recognizing contribution, coaching gaps, handling conflict fairly, listening to employees, and making talent decisions with evidence. The research behind those practices is not new. What is new is making them operate together in a hybrid-first management operating plane — replacing management by archaeology with a weekly rhythm where AI develops managers instead of replacing them.

Go deeper: the People Science pillar

This document is the hub of Cadence’s People Science pillar. Each spoke below takes one body of evidence to research depth — same citation discipline, same honesty rules:

• The Manager Effect — what the science actually says about why managers make or break teams.
• Continuous Performance Management vs. the Annual Review — the evidence on both sides of the ratings debate.
• Organizational Justice: The People Science of Fair Process — why ER is a trust system, not a compliance chore.
• Feedback That Doesn’t Backfire — why a third of feedback interventions make performance worse.
• The Real Science of Employee Recognition — including the truth about the 5:1 ratio.
• Goal-Setting Theory in Practice — when specific, challenging goals help, and when they backfire.
• People Analytics Without the Dashboard Theater — turning people data into decisions.
• Job Architecture and Role Clarity — the underrated foundation of everything else.
• Management by Proximity Is Dead — the people science of leading hybrid and distributed teams.
• Can AI Be Fair? — the people science of algorithmic talent decisions.
• The Compounding Organization — why people practices work as systems, not point tools.
• The People Science Glossary — an evidence index of every core concept, each pinned to its canonical source.

How to cite this document

Suggested citation: Cadence, “The Science of Management Cadence” (2026). https://cadencehr.ai/resources/people-science

Methodology and provenance. This synthesis was drafted in June 2026 and revised in July 2026 by Cadence. Sources were selected with a preference for peer-reviewed meta-analyses (feedback interventions, goal setting, engagement, organizational justice, mechanical vs. clinical judgment, HR systems), supplemented by primary research programs (Gallup Q12, Microsoft Work Trend Index) and established practitioner frameworks (O*NET, SIOP principles, CIPD, AIHR) where the relevant knowledge is methodological or definitional rather than experimental. Every citation below was verified to resolve to the named source as of 2026-07-27. Claims about Cadence’s product are labeled with current availability (live / preview / roadmap); claims that are Cadence’s own synthesis rather than external research findings are identified as such in the text.

References

1. Kluger, A. N., & DeNisi, A. (1996). “The Effects of Feedback Interventions on Performance: A Historical Review, a Meta-Analysis, and a Preliminary Feedback Intervention Theory.” Psychological Bulletin, 119(2), 254–284. doi:10.1037/0033-2909.119.2.254
2. Locke, E. A., & Latham, G. P. (2002). “Building a Practically Useful Theory of Goal Setting and Task Motivation: A 35-Year Odyssey.” American Psychologist, 57(9), 705–717. doi:10.1037/0003-066X.57.9.705
3. Locke, E. A., & Latham, G. P. (2006). “New Directions in Goal-Setting Theory.” Current Directions in Psychological Science, 15(5), 265–268. doi:10.1111/j.1467-8721.2006.00449.x
4. Ordóñez, L. D., Schweitzer, M. E., Galinsky, A. D., & Bazerman, M. H. (2009). “Goals Gone Wild: The Systematic Side Effects of Overprescribing Goal Setting.” Academy of Management Perspectives, 23(1), 6–16. doi:10.5465/amp.2009.37007999
5. Harter, J. K., Schmidt, F. L., & Hayes, T. L. (2002). “Business-Unit-Level Relationship Between Employee Satisfaction, Employee Engagement, and Business Outcomes: A Meta-Analysis.” Journal of Applied Psychology, 87(2), 268–279. doi:10.1037/0021-9010.87.2.268
6. Gallup, “The Relationship Between Engagement at Work and Organizational Outcomes,” Q12® Meta-Analysis, 11th edition. gallup.com/workplace/321725
7. Gallup (2015). State of the American Manager: Analytics and Advice for Leaders; summarized in “Managers Account for 70% of Variance in Employee Engagement,” Gallup Business Journal, April 2015. news.gallup.com/businessjournal/182792
8. Colquitt, J. A., Conlon, D. E., Wesson, M. J., Porter, C. O. L. H., & Ng, K. Y. (2001). “Justice at the Millennium: A Meta-Analytic Review of 25 Years of Organizational Justice Research.” Journal of Applied Psychology, 86(3), 425–445. doi:10.1037/0021-9010.86.3.425
9. Kuncel, N. R., Klieger, D. M., Connelly, B. S., & Ones, D. S. (2013). “Mechanical Versus Clinical Data Combination in Selection and Admissions Decisions: A Meta-Analysis.” Journal of Applied Psychology, 98(6), 1060–1072. doi:10.1037/a0034156
10. U.S. Equal Employment Opportunity Commission, “Employment Tests and Selection Procedures” (fact sheet on the Uniform Guidelines on Employee Selection Procedures, 29 C.F.R. Part 1607). eeoc.gov
11. Gallup, “The Importance of Employee Recognition: Low Cost, High Impact.” gallup.com/workplace/236441
12. Gallup, “Gallup’s Q12 Employee Engagement Survey.” gallup.com/workplace/356063
13. Microsoft WorkLab, “Hybrid Work Is Just Work. Are We Doing It Wrong?” 2022 Work Trend Index Pulse Report. microsoft.com/worklab
14. Gallup, “The Future of Hybrid Work: 5 Key Questions Answered With Data.” gallup.com/workplace/390632
15. AIHR, “9 Box Grid: A Practitioner’s Guide.” aihr.com/blog/9-box-grid
16. Sawyer, J. E. (1992). “Goal and Process Clarity: Specification of Multiple Constructs of Role Ambiguity and a Structural Equation Model of Their Antecedents and Consequences.” Journal of Applied Psychology, 77(2), 130–142. doi:10.1037/0021-9010.77.2.130
17. ONET Resource Center, “The ONET Content Model.” onetcenter.org/content.html
18. Society for Industrial and Organizational Psychology, Principles for the Validation and Use of Personnel Selection Procedures, 5th ed., 2018. APA-hosted PDF
19. Campion, M. A., Fink, A. A., Ruggeberg, B. J., Carr, L., Phillips, G. M., & Odman, R. B. (2011). “Doing Competencies Well: Best Practices in Competency Modeling.” Personnel Psychology, 64, 225–262. doi:10.1111/j.1744-6570.2010.01207.x
20. CIPD, “People Analytics” factsheet. cipd.org
21. Rousseau, D. M. (2006). “Is There Such a Thing as Evidence-Based Management?” Academy of Management Review, 31(2), 256–269. doi:10.5465/amr.2006.20208679
22. Jiang, K., Lepak, D. P., Hu, J., & Baer, J. C. (2012). “How Does Human Resource Management Influence Organizational Outcomes? A Meta-Analytic Investigation of Mediating Mechanisms.” Academy of Management Journal, 55(6), 1264–1294. doi:10.5465/amj.2011.0088

This is a research synthesis, not a Cadence customer-outcome claim. Module availability is labeled because this page covers both live and roadmap concepts.

See how Cadence turns people science into operating rhythm at cadencehr.ai/product, or check plans at cadencehr.ai/pricing.