Executive summary
High-achievers are trained to read declining output as the early warning signal. In a clinical review of 287 high-achieving clients seen at CEREVITY between January 2025 and June 2026, the opposite sequence was the more common one: 67 percent reported that their strongest periods of professional output coincided with their highest self-rated internal strain, cognitive fog, or emotional exhaustion. The metrics held. The person producing them did not.
Founders, chief executives and senior professionals operate under sustained load in roles where composure is read as fitness, and where the reward system keeps paying for output long after recovery capacity has been exceeded.
Standard burnout screens look for visible decline, so they miss the presentation in which performance stays intact and only the internal experience deteriorates, which is the presentation this population most often has.
Treat sustained high output under high internal strain as a clinical signal in its own right, and route the person to confidential, role-informed care before the decoupling becomes the operating baseline.
Recognition moves earlier in the cycle, the disclosure barrier is removed rather than argued with, and judgment quality is protected while the person is still in the seat rather than repaired after a rupture.
The problemDeclining output is the wrong early warning signal§
The usual model of burnout assumes performance falls before or alongside distress. In high-achieving populations the sequence is frequently reversed, and the external evidence points the same way. Gallup's 2026 global workplace data show leaders scoring higher than the people they lead on engagement and life evaluation while simultaneously reporting more daily stress, anger, sadness and loneliness, gaps of 7, 12, 11 and 10 points respectively.07, 08 A laboratory finding puts the same dissociation under a different instrument: in a controlled comparison of 28 people with occupational burnout and 26 without, the two groups performed equally on cognitive tasks while the burnout group showed altered neural processing consistent with compensatory recruitment, more resource spent to hold the same output.12
Founders continue to close rounds and ship product. Executives continue to hit board targets. Equity partners continue to originate work and bill hours. Internally they describe persistent exhaustion, narrowed cognitive flexibility, emotional flatness, and a growing sense of running the machine without being inside it. Because the external scores remain acceptable, the pattern is read as temporary intensity rather than a clinical signal. Three structural features make it more likely in this group. Identity is fused with the role or the enterprise, so admitting strain can feel like conceding that the venture is failing. Scrutiny is high and peer parity is low, so candour carries a visible cost. And the reward system keeps reinforcing output after recovery capacity has already been spent. The consequence is not only private suffering. Judgment quality softens while decision velocity stays high, and the organization keeps receiving fast answers from a person whose internal signal-to-noise ratio has already shifted.
The dashboard stayed green. The person reading it did not. CEREVITY clinical review, n=287, January 2025 to June 2026
The evidenceWhat the research shows§
Two bodies of data are used here and they are kept separate throughout. The first is a clinical review of 287 consecutive high-achieving clients seen at CEREVITY between January 2025 and June 2026. It is a help-seeking sample, not a population survey, and every figure drawn from it is labelled with its n and date range. The second is published survey and peer-reviewed research on founders, chief executives and senior professionals, which supplies the external benchmarks the clinical figures are placed against.
67%
of high-achieving clients reported peak professional output overlapping peak internal strain
CEREVITY clinical review, n=287, 2025 to 2026
61%
concealed the strain for four months or longer while metrics held steady or improved
CEREVITY clinical review, n=287, 2025 to 2026
54%
said the first place they fully disclosed the cost was a clinical setting
CEREVITY clinical review, n=287, 2025 to 2026
19 mo
median delay from clear recognition of the pattern to a first session
CEREVITY clinical review, n=287, 2025 to 2026
Read together, the internal and external figures describe one pattern rather than two: high prevalence, prolonged concealment, and delayed help-seeking among the people whose decisions carry the most leverage. Among founders, 72 percent report an impact on their mental health, 81 percent say they are not really open about their stress, fears and challenges, and 77 percent do not get professional help.03 Sifted's February 2025 survey of 138 founders found 54 percent had experienced burnout in the previous 12 months, 75 percent anxiety, and 46 percent rated their mental health as bad or very bad.02 CEREVITY's own 2025 review of tech founders documented the same concealment behaviour in a separate sample, and is summarised in our 2025 review of tech founder burnout.
The executive picture is consistent. In the Wall Street Journal and Vistage CEO Confidence Index surveyed 5 to 13 May 2025, 494 chief executives of businesses with revenues between 1 million and 20 million dollars reported burnout at a combined 71 percent: 39 percent occasionally, 25 percent frequently and 7 percent nearly every day.05 Vistage's own second-quarter 2025 reading put 24 percent at daily or frequent burnout with a further 44 percent feeling the pressure occasionally.17 Businessolver's 2024 study recorded 55 percent of chief executives saying they had experienced a mental health issue, a 24 point rise year over year.06 We have written elsewhere about what actually helps once a chief executive names burnout.
| Indicator | CEREVITY sample (n=287) | External benchmark | Source |
|---|---|---|---|
| Peak output overlapping peak internal strain | 67% | No direct published equivalent; Gallup records leaders higher on engagement and on daily strain at once | CEREVITY01; Gallup07 |
| Concealment of four months or longer while metrics held | 61% | 81% of founders are not really open about stress, fears and challenges | CEREVITY01; Startup Snapshot03 |
| First full disclosure occurred in a clinical setting | 54% | 77% of founders do not get professional help | CEREVITY01; Startup Snapshot03 |
| Median delay, clear recognition to first session | 19 months | 6 to 8 years from onset to first treatment contact for mood disorders, general population | CEREVITY01; Wang et al.14 |
| Reported decline in decision quality during high-strain periods | 58% | Acute stress impairs working memory and cognitive flexibility; effects on inhibition are more nuanced | CEREVITY01; Shields et al.13 |
| Burnout in the prior 12 months | Not separately scored | 54% of founders; 71% of small and midsize company chief executives | Sifted02; WSJ / Vistage05 |
| Lifetime mental health condition | Not separately scored | 49% of entrepreneurs (n=242) against 23.1% of US adults | Freeman et al.04; NIMH10 |
01, 05 CEREVITY clinical review, n=287, January 2025 to June 2026. Clinical, help-seeking sample; not a population prevalence estimate.
Wall Street Journal and Vistage CEO Confidence Index, n=494, fielded 5 to 13 May 2025.
11 Lehigh University, Nasdaq Entrepreneurial Center and TU Dortmund University, n=308 entrepreneurs, reported September 2025.
The frameworkA model you can name and own§
Four observable phases recur in the clinical sample. The model is descriptive rather than a formal diagnostic instrument, and its value is that each phase names something a clinician, a co-founder, a board member or the person themselves can actually notice. Naming the phase is what makes the pattern discussable before it is undeniable.
CEREVITY model
The Performance Paradox Cycle
A four-phase description of how strain and output come apart in people whose competence conceals the cost. External performance is the last thing to move, which is precisely why it is the wrong thing to watch.
Compensation
Output holds or rises. Hours, intensity and control all increase. Sleep and recovery are treated as discretionary. Strain is present and is framed as temporary.
Concealment
The person becomes deliberate about what boards, co-founders, investors and family are shown. Metrics still look strong. Disclosure is calculated as riskier than continued self-management.
Decoupling
Internal experience and external performance diverge further. Cognitive flexibility narrows. Decision quality softens while decision speed holds. People describe still performing but no longer feeling the wins.
Rupture point
A near miss, a health event, a family confrontation or a sudden loss of capacity forces the pattern into view. The alternative is that the person reaches a clinician first, which in this sample took a median of 19 months from clear recognition.
The clinical task is to interrupt the cycle during Compensation or early Concealment, before Decoupling becomes the operating baseline. That is a different objective from waiting for a performance decline to justify intervention, and it is the objective that executive burnout therapy is structured around for this population.
SCHEMATIC Schematic, not measured data.
Descriptive model derived from patterns observed in the CEREVITY clinical review, n=287, January 2025 to June 2026. The curves are illustrative and carry no units.
By professionHow it presents across roles§
The paradox is one pattern, but the pressure that sustains it is role-specific, and so is the calculation each group makes about disclosure. The three segments below are the ones represented in the CEREVITY sample.
Startup founders and tech operators
Identity and enterprise are fused here more completely than in any other segment. Fundraising and board updates reward the performance of certainty, so the incentive is not merely to feel confident but to be seen to be confident on a schedule. In the CEREVITY sample this group showed the highest rate of active concealment from investors, and the external data align closely: 54 percent of founders reported burnout in the prior 12 months, 75 percent anxiety, and 46 percent rated their mental health as bad or very bad.02 Broader founder research finds 72 percent reporting an impact on mental health while 77 percent do not get professional help, and 81 percent describing themselves as not really open about their stress.03 The gap between prevalence and disclosure is the segment's defining feature, not the prevalence itself. It is also modifiable. A 2025 study of 308 entrepreneurs run by Lehigh University with the Nasdaq Entrepreneurial Center and TU Dortmund found that founders who held clear work-life boundaries reported high burnout at 23 percent against 67 percent among those who struggled to hold them, and low burnout at 45 percent against 6 percent.11 Boundaries are not a wellness slogan in this population; they are the single strongest protective variable in the published data. Clinically, the founder presentation tends to combine sleep compression, decision volume that exceeds recovery, and a private conviction that stepping back for one quarter would be read by the cap table as a permanent verdict.
Individual therapy for people who cannot look uncertain to their cap table
Organizational founder mental health as portfolio risk management
C-suite and senior executives
The chief executive seat carries asymmetric visibility and almost no peer parity. Boards and teams read composure as evidence of fitness, which makes candour expensive in a way it is not one level down. The published rates are high and rising: 71 percent of chief executives at businesses with 1 million to 20 million dollars in revenue reported burnout over 12 months, split 39 percent occasionally, 25 percent frequently and 7 percent nearly every day,05 and 55 percent of chief executives reported having experienced a mental health issue, up 24 points year over year.06 Deloitte and Workplace Intelligence, surveying more than 2,100 employees and C-level executives, found roughly one in three executives always or often feeling exhausted, stressed, overwhelmed, lonely or depressed, 73 percent saying the job does not let them disconnect, and 69 percent seriously considering leaving for a role that better supports their wellbeing.09 Gallup's 2026 data locate the paradox precisely: leaders report higher engagement and higher life evaluation than the people they lead, and more daily stress, anger, sadness and loneliness at the same time.07, 08 In this segment the Performance Paradox presents as continued delivery against board targets while private recovery capacity collapses, and the loneliness is not incidental to it. There is often nobody in the person's professional life who can receive the information without acting on it.
Individual confidential therapy for chief executives
Organizational executive mental health benefit for companies
High-earning professionals, including equity partners
Compensation and status create a barrier the other two segments feel less acutely: the sense that distress is illegitimate given the external rewards. Clients in this group frequently open by disqualifying their own experience before describing it. Performance anxiety is common, as is the belief that a single visible failure will erase years of accumulated credibility, which makes the concealment calculation rational rather than avoidant. Concealment here is often maintained with spouses and close colleagues as well as with the firm, so the person has no setting at all in which the full picture has been said out loud. That is the mechanism behind the finding that 54 percent of the CEREVITY sample first disclosed the cost in a clinical setting rather than to a partner, peer, coach or board.01 The clinical presentation frequently includes decision fatigue, sleep disruption and a narrowing of identity outside work, and the underlying epidemiology is not reassuring: 49 percent of entrepreneurs in a matched comparison reported a lifetime mental health condition against 23.1 percent of US adults.04, 10 Recovery research in owner-operator populations points the same way, finding fewer recovery experiences across every dimension measured than in employed comparison groups, with psychological detachment scoring lowest of all.15
Individual clinical care for high-pressure professionals
Organizational confidential therapy for equity partners
The stakesThe cost of inaction§
The first cost of the Performance Paradox is not turnover. It is degraded judgment while the person is still in the seat and still being trusted with the decisions. In the CEREVITY sample, 58 percent of clients meeting Performance Paradox criteria reported a noticeable decline in decision quality during high-strain periods, describing more impulsive choices, reduced creativity, or higher regret after the fact.01 That self-report has a mechanism behind it. A meta-analysis of acute stress across the three core executive functions found working memory and cognitive flexibility both impaired, with more nuanced effects on inhibition.13
Judgment, while still in the seat
Cognitive flexibility narrows before anything visible breaks, so the organization keeps receiving decisions at full speed and reduced quality. This is the cost that is never attributed correctly, because by the time the consequences arrive the strain has usually been reframed as a bad quarter.13
Recovery capacity and health
Sleep compression and the loss of psychological detachment reduce the cognitive reserve the next decision draws on. In owner-operator populations, recovery experiences are lower across every measured dimension than in employed comparison groups, and detachment is the lowest of them.15 Relationships often register the change before the person does.01
Organizational and macro cost
Teams absorb a leader's narrowed bandwidth, and the wider bill is well documented: the World Health Organization attributes an estimated 12 billion lost working days and roughly 1 trillion dollars a year in lost productivity to depression and anxiety.16 Succession and replacement risk arrives only at the end, once rupture has already occurred.
01, 14 CEREVITY clinical review, n=287, January 2025 to June 2026.
Wang, P. S. et al. (2005). National Comorbidity Survey Replication, n=9,282. The context bar measures a different population and a different interval, onset to first treatment contact, and is shown for scale only.
The solutionWhat effective care looks like§
For this population the barriers are structural at least as much as motivational: fear of professional consequence, identity fused with the role, and the absence of any peer who can be told the whole thing. Effective care therefore has to remove the disclosure risk rather than argue the person out of it, has to be delivered by clinicians who already understand the pressures of the founder seat, the chief executive seat or the partnership track, and has to run in session lengths that match how dense the material is. It also has to name the Performance Paradox explicitly, so the person can identify the decoupling instead of treating it as a character problem to be out-worked.
CEREVITY is a nationwide network of independent licensed clinicians, matched to the person and delivered by secure video on a fully private-pay basis, which keeps the work off insurance records and employer systems. Sessions run in three formats and no others. The 50-minute weekly session carries continuity between weeks, and the 90-minute extended format leaves room for the work an hour keeps interrupting, which in this population is usually the material that took months to decide to raise at all.
Where the material is dense enough that weekly work would fragment it, the 3-hour intensive format does it in a single block instead. How the clinical model is structured is set out separately, including how clinicians are matched to the pressures of a role rather than to a diagnosis alone, which is the part that determines whether the first session is useful to someone who has spent months deciding not to have it.
ImplementationHow to put it into practice§
Four steps, written to be usable by an individual, by a board or family office reducing the lag for someone it is responsible for, or by an investor who would rather find this out before a portfolio company does.
- 01
Name the pattern rather than the mood
Ask directly whether the periods of strongest output have also been the periods of highest internal strain. That question separates the Performance Paradox from ordinary intensity faster than any symptom checklist, because it asks about the relationship between two things the person already tracks.
- 02
Separate care from the cap table and the employment record
Private-pay structure removes the most commonly cited disclosure barrier, since there is no diagnosis code travelling through an insurer and no shared record inside the employer's systems. In a population where 81 percent describe themselves as not really open about their stress,03 removing the mechanism of exposure does more than reassurance does.
- 03
Match clinician experience to the role, not only to the symptom
Generic counselling rarely reaches the specific dynamics of an investor update, a board meeting or a partnership vote. The clinical content of the founder seat is different from the clinical content of the chief executive seat, and both are different again from the partnership track.
- 04
Intervene during Compensation or early Concealment
Waiting for visible performance decline means waiting for the later phases of the cycle. The 19-month median between clear recognition and a first session in this sample is long enough to produce measurable judgment costs before anyone acts.01
RecommendationsWhere to start§
Clinical
Read sustained output under strain as a signal
Treat high output maintained under high internal strain as a clinical finding rather than a character strength. The controlled evidence that equal task performance can sit on top of measurably altered processing12 is reason enough not to accept intact metrics as reassurance.
Clinical
Screen for cognitive load, not just mood
Where the presenting complaint is decision volume, narrowing flexibility and post-decision regret, the useful frame is cognitive load rather than mood, which is what care for cognitive load in decision-heavy roles is built to address.
Structural
Measure the delay, then shorten it
Track the interval from recognition to first session as an operational number. It was a median of 19 months in this sample, and it is the one variable in the cycle that an individual, a board or an investor can actually move. Our earlier 2026 High-Achiever Burnout Index tracks the same lag in a separate cohort.
Structural
For boards and investors, normalise it before the rupture
Raise recovery capacity as a standing governance question rather than an emergency one. Organization-level change is more effective than individual apps where the work itself is the driver,16 but the individual still needs a private place to think, and the two are not substitutes.
FAQCommon questions§
Where does the 67% figure come from?
Is the Performance Paradox the same as burnout?
How is this different from normal high intensity?
How does private-pay billing work?
How is my privacy protected?
MethodologyHow this paper was built§
Methodology
This Index draws on two separate bodies of evidence, and they are labelled distinctly throughout the paper. The first is a clinical review of consecutive high-achieving clients seen through CEREVITY between 1 January 2025 and 30 June 2026. After inclusion criteria were applied, senior decision-making responsibility, an income or equity profile consistent with the high-achiever segment, and sufficient clinical contact for the relevant variables to be assessed, 287 clients remained. The Performance Paradox rate and every other internal figure is derived from that sample and carries its n and date range wherever it appears. Internal variables were recorded from structured intake and from clinician-documented review, including the temporal relationship between self-rated output and self-rated strain, the duration of concealment and the audiences it was maintained with, the setting of first full disclosure, the interval between clear self-recognition and a first session, and self-reported change in decision quality during high-strain periods. The second body of evidence is published survey and peer-reviewed research on founders, chief executives and senior professionals. Sources were identified through PubMed, PsycINFO and Google Scholar for the peer-reviewed material, and through direct retrieval from the publishing organisations for the survey material, covering publications from 2005 to 2026 with priority given to the most recent available reading of any recurring survey. Included survey sources carry a stated sample size and a stated field period: Sifted (n=138, February 2025), the Wall Street Journal and Vistage CEO Confidence Index (n=494, fielded 5 to 13 May 2025), Startup Snapshot (400 plus founders), Deloitte with Workplace Intelligence (2,100 plus employees and C-level executives), and the Lehigh University, Nasdaq Entrepreneurial Center and TU Dortmund study (n=308). Peer-reviewed sources include Freeman and colleagues (n=242 entrepreneurs against 93 comparison participants), Wang and colleagues (n=9,282), Pihlaja and colleagues (n=54) and Le Moal and colleagues (n=1,043). Limitations are material and should be read alongside every internal figure. The CEREVITY sample is clinical and help-seeking, so it over-represents people who eventually reached care and cannot support a population prevalence claim. Self-report of the historical relationship between output and strain is subject to recall and attribution bias, and the output side of that relationship is self-rated rather than independently measured. The external surveys use varying definitions of burnout and of mental health impact, several are self-selected samples, and the founder and chief executive surveys differ in geography and company size, so they benchmark rather than replicate one another. Comparisons across sources, such as entrepreneur prevalence against the United States adult baseline, combine instruments that were not designed to be compared and are presented as context rather than as a matched contrast. The Index is therefore best read as a synthesis of clinical observation placed against published epidemiology, not as a single definitive prevalence estimate.
References
- 01CEREVITY. (2026). Performance Paradox Index: clinical review of 287 consecutive high-achieving clients, January 2025 to June 2026. Internal clinical data, not publicly posted.
- 02Sifted. (2025, February 26). More than half of founders experienced burnout last year. Founder mental health survey, n=138. sifted.eu
- 03Startup Snapshot. (2023). The Untold Toll: the impact of stress on the well-being of startup founders and CEOs. 400 plus founders surveyed. startupsnapshot.com
- 04Freeman, M. A., Staudenmaier, P. J., Zisser, M. R., and Andresen, L. A. (2019). The prevalence and co-occurrence of psychiatric conditions among entrepreneurs and their families. Small Business Economics, 53(2). n=242 entrepreneurs, 93 comparison participants. link.springer.com
- 05Wall Street Journal and Vistage. (2025). CEO Confidence Index, fielded 5 to 13 May 2025, n=494 CEOs of businesses with 1 million to 20 million dollars in revenue. Reported in Becker's Hospital Review. beckershospitalreview.com
- 06Businessolver. (2024). 2024 State of Workplace Empathy Study: 55 percent of CEOs say they have experienced a mental health issue, up 24 points. businessolver.com
- 07Gallup. (2026). State of the Global Workplace Report. gallup.com
- 08Ratanjee, V. (2026, April 8). Gallup's 2026 workplace report: leaders are more engaged, they are also more stressed. Forbes. forbes.com
- 09Deloitte and Workplace Intelligence. (2022). The C-suite's role in well-being. Survey of more than 2,100 employees and C-level executives. workplaceintelligence.com
- 10National Institute of Mental Health. (2022 data). Mental illness: any mental illness among US adults, 23.1 percent. nimh.nih.gov
- 11Lehigh University, Nasdaq Entrepreneurial Center and TU Dortmund University. (2025, September 12). We studied entrepreneurs' well-being. Surveys, interviews and focus groups with 308 entrepreneurs. Fortune. fortune.com
- 12Pihlaja, M., Perakyla, J., Erkkila, E.-H., Tapio, E., Vertanen, M., and Hartikainen, K. M. (2023). Altered neural processes underlying executive function in occupational burnout: basis for a novel EEG biomarker. Frontiers in Human Neuroscience, 17. n=54. frontiersin.org
- 13Shields, G. S., Sazma, M. A., and Yonelinas, A. P. (2016). The effects of acute stress on core executive functions: a meta-analysis and comparison with cortisol. Neuroscience and Biobehavioral Reviews, 68, 651 to 668. escholarship.org
- 14Wang, P. S., Berglund, P., Olfson, M., Pincus, H. A., Wells, K. B., and Kessler, R. C. (2005). Failure and delay in initial treatment contact after first onset of mental disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry, 62(6), 603 to 613. n=9,282. pubmed.ncbi.nlm.nih.gov
- 15Le Moal, M., Thurik, R., Torres, O., and Soenen, G. (2026). Mental health of entrepreneurs and daily recovery experiences. Small Business Economics. n=1,043. link.springer.com
- 16World Health Organization. (2022). Guidelines on mental health at work. who.int
- 17Vistage. (2025, August 26). CEO burnout and resilience strategies for uncertain times. Vistage CEO Confidence Index, Q2 2025. vistage.com
Licensed Clinical Social Worker
Martha Fernandez, LCSW is Co-Founder of CEREVITY and a Licensed Clinical Social Worker licensed in California, seeing clients by telehealth nationwide through CEREVITY's network of independent licensed clinicians. USC-trained and bilingual in English and Spanish, she works with founders, executives, attorneys and pilots on burnout, anxiety and depression in high performers, on trauma, grief and high-stakes transitions, and on couples and relationship strain under pressure. Her clinical work draws on cognitive behavioral therapy, acceptance and commitment therapy, behavioral activation, and narrative and solution-focused approaches. She is the author of Wired to Burn.
Talk to someone who gets it.
If this paper describes something you recognize, a confidential conversation is the next step. CEREVITY matches you to an independent licensed clinician who works with people in your position.
Schedule consultationPrivate-pay, telehealth, nationwide. Questions: (562) 295-6650