Can Coaching Be Harmful? A Practical Guide for Coaches

Yes. Can coaching be harmful? The peer-reviewed answer is that it can, and more often than most training programs admit. Researchers who asked coaches to evaluate 123 engagements found an average of two unwanted side effects for the client per coaching, usually mild, occasionally serious. The causes are mostly ones you control: taking on a client who needs a different kind of help, letting the work drift into therapy territory, and pushing people with pressure or promises. This guide gives you a checklist, the research behind each item, scripts for pausing or referring, and a fictional case so you can see the whole thing play out. Nothing here is legal, medical or mental-health advice, and the rules on what counts as therapy or counselling vary by state and country, so check with a licensed professional or attorney where you practise.

We make Coachful, a coaching platform, and it shows up here only where it helps with this question: a client record that keeps intake answers, notes and check-ins in one place, and flags for clients who go quiet, make it easier to spot the early signs that coaching is going wrong. The rest of the article works with any setup, including a notebook and a calendar.

What is on a safe-coaching checklist?

Run through these nine checks before you take on a client and again whenever an engagement feels off. Each one maps to a cause of harm that the research or official guidance points to, and each is explained further down.

None of this is exotic. Most of it is what therapists, physiotherapists and personal trainers already do as routine. Coaching is younger and largely unregulated, so the routine is yours to build. If you are still deciding whether your background is enough to start, the guides on life coach qualifications and whether you need a license cover the legal side; this article is about doing no harm once you are working.

What does peer-reviewed research say about harm from coaching?

The short version: negative effects are common, mostly low in intensity, and linked to things the coach can influence. For years coaching research only measured whether coaching worked. A group of researchers led by Carsten Schermuly in Germany started measuring the other side in the early 2010s, borrowing methods from psychotherapy research, where side effects have been studied for decades.

The first systematic study of side effects for clients

In 2014, Schermuly and colleagues published the first systematic study of negative effects of coaching in the Zeitschrift für Arbeits- und Organisationspsychologie. After a qualitative pre-study to identify possible negative effects, they had coaches evaluate 123 coaching engagements online. On average, each coaching produced two negative effects for the client, mostly of low intensity. The causes coaches named were mainly client factors, but also how the coach handled a client's mental illness and lack of time. One more finding matters for anyone building packages: the more topics a coaching covered, the more negative effects it produced (Schermuly, Schermuly-Haupt, Schölmerich and Rauterberg, 2014).

Read that last point twice. A sprawling "life audit" package that touches career, relationships, money, health and purpose in twelve weeks is not just less focused. In this data, breadth went together with more harm.

Coaches are affected too

The same year, Schermuly surveyed 104 German business coaches about negative effects on themselves, published in the International Coaching Psychology Review. In more than 90 per cent of their most recent coachings, at least one negative effect for the coach was present, with an average of 5.9 per coaching. Over their careers, 99 per cent had faced at least one. The effects were frequent and varied but of rather low intensity, and they were associated with lower wellbeing and lower psychological empowerment in the coaches (Schermuly, 2014, International Coaching Psychology Review 9(2)).

What predicts harm

A time-lagged study of 275 coaches from an international sample, by Carolin Graßmann and Schermuly, asked coaches to evaluate their last completed coaching and then measured their health eight weeks later. Coaches experienced more negative effects when the relationship quality with their client was low. When coaches saw low goal attainment and many negative effects for their clients, they felt less competent and experienced more negative effects themselves. And coaches with more negative effects at the start reported more stress and impaired sleep eight weeks later (Graßmann and Schermuly, 2018).

A randomised field experiment by the same pair, run with a student sample, gave half the novice coaches supervision during the coaching and half only afterwards. From the coaches' perspective there was a strong link between harm to clients and harm to coaches. That link was stronger when the coach scored high on neuroticism, but only when they did not have supervision during the coaching (Graßmann and Schermuly, 2017).

The literature review that pulls it together

Schermuly and Graßmann's review in Coaching: An International Journal of Theory, Research and Practice synthesised nine studies with qualitative, cross-sectional, time-lagged and experimental designs. Their conclusions: negative effects occurred frequently, only a few were severe and most were low in intensity; a higher-quality relationship between client and coach went with fewer negative effects; supervision appeared to help; and negative effects for clients and for coaches were interrelated. They argue that coaching should treat side effects the way other helping professions do, as a natural occurrence to be managed rather than a taboo (Schermuly and Graßmann, 2019).

Two honest caveats. Most of this evidence comes from coaches rating their own work, and much of it from German business coaching. Clients may report different things, and life or health coaching may look different. The direction of the findings is still consistent across designs, which is why it is worth acting on.

Why does poor client fit cause harm?

Because coaching assumes a client who is well enough to set goals and act on them, and some people who book a coach are not in that place. Anthony Grant, a coaching psychologist at the University of Sydney, noted in his chapter on a languishing-flourishing model of goal striving (reprinted in the collection Coaching Researched) that research suggests some coaching clients have high levels of depression, anxiety or stress, and that people who voluntarily seek life coaching tend to show higher levels of psychopathology than people coached through a workplace program (Grant).

That matters most for independent coaches in their first years, whose clients find them through social media and self-refer. Some of those clients choose coaching precisely because it feels less stigmatising than therapy, or because therapy has a waiting list. Coaching them anyway, with goal-setting exercises and accountability, can turn into one more thing they are failing at.

Poor fit is not only about mental health. Other mismatches that produce harm:

• The goal is someone else's. A manager sends an employee to coaching to "fix" them, or a partner books sessions for the other partner. Low ownership drags down the relationship quality the research ties to fewer negative effects.
• The client needs expertise you do not have. Debt that needs a financial counsellor, a medical condition that needs a doctor's clearance before any exercise or diet change, an immigration problem that needs a lawyer.
• Timing. A client in the first weeks after a bereavement, a diagnosis or a job loss may need support, not a stretch goal.
• Style. Some clients want direct challenge, some fold under it. A coach who only has one gear will hurt one of the two.

If you are a therapist adding coaching to your practice, fit is also about which hat you are wearing; the article on whether a therapist can be a life coach covers keeping the two separate.

When does coaching drift into therapy, and why is that risky?

Scope drift happens gradually, usually when a client trusts you and starts bringing the hardest material in their life to sessions. It is risky because you are not trained or licensed to treat it, you have no clinical framework for what you are hearing, and in some places presenting yourself as providing counselling or therapy without a license is unlawful. The 2014 study's finding that coaches named their handling of a client's mental illness as a cause of harm is the research version of this problem.

Drift rarely announces itself. These are the patterns to watch for:

• Sessions increasingly spent on the past (childhood, a past relationship, trauma) rather than on present choices and future actions.
• The client stops setting or reviewing goals and uses the hour to unload.
• You find yourself offering explanations for their behaviour that sound like diagnoses ("that sounds like an anxiety disorder", "you might have ADHD").
• The client contacts you between sessions in distress, and you feel responsible for managing it.
• You leave sessions drained, or think about the client at night. The 2018 study links this to stress and sleep problems in coaches.

Some coaching niches sit closer to the line than others. ADHD, anxiety, relationships, divorce and health coaching all attract clients with conditions that a clinician also treats. The guides on becoming an ADHD coach and coaching when you have anxiety yourself go further into those niches. This table is a starting point for where coaching ends.

This table is not medical or legal advice, and services differ by country. Build your own version with the actual local names and numbers before you need it.

Can sales pressure and big promises harm clients?

Yes, and it is the cause most within your control. Harm from pressure looks like a client who buys a package they cannot afford because a call felt urgent, who feels ashamed when they do not get the promised result, or who stays in a program that is not working because they were told the dip is always part of the process.

The US Federal Trade Commission treats some of this as consumer protection. Its consumer advice on coaching scams, as listed in November 2026, says that if an offer "promises guaranteed income, large returns, or a 'proven system,' it's likely a scam" (FTC consumer advice). In October 2021 the FTC sent a Notice of Penalty Offenses about money-making claims to more than 1,100 businesses, including investment and business coaching, warning of civil penalties for false or misleading earnings claims; being on that list did not itself suggest wrongdoing (FTC press release).

Testimonials are covered too. The FTC's Endorsement Guides say endorsements must be honest and not misleading, and that testimonials showing exceptional results need proof that they are what people generally achieve, or the ad must make the generally expected results clear (FTC Endorsement Guides Q&A). The Consumer Reviews and Testimonials Rule, in effect since October 21, 2024, prohibits fake reviews and testimonials and conditioning incentives on a particular sentiment (FTC rule Q&A).

Outside the US, other consumer-protection rules apply; ask a local attorney if your marketing makes outcome claims. The ethical point holds everywhere: a promise you cannot keep sets the client up to blame themselves when the result does not come.

Pressure also happens inside the engagement. Holding someone to a goal they have outgrown, shaming missed tasks, or pushing a "breakthrough" the client did not ask for all count. A minimum commitment can be healthy when it is explained honestly; the article on structuring a multi-month commitment through the three-month dip shows how to do that without trapping anyone.

What warning signs should you watch for in your own clients?

Watch for change, not for any single bad week. A client who was engaged and goes quiet, whose mood scores slide for three weeks, or whose language shifts from "I want to" to "I can't" is telling you something. The table below pairs common signs with what they may mean and a first response.

Spotting silence is easier if you are not relying on memory. Any system that shows when each client last checked in will do. In Coachful, the clients overview shows days-silent flags and a needs-attention list across your whole roster, so the client who quietly dropped off three weeks ago does not wait until their renewal date to be noticed.

What do you say when you need to pause or refer a client?

Say what you have noticed, say what coaching can and cannot do, and offer a concrete next step. Clients rarely hear a well-delivered referral as rejection; they hear it as being taken seriously. Adapt these scripts to your own voice.

Script 1: naming scope drift

"I've noticed that over the last few sessions we've spent most of our time on what happened with your father, and much less on the goals you came in with. That makes complete sense, it's clearly important. It's also the kind of work a therapist is trained for and I'm not. Would you be open to working with someone on that side of things? We can keep coaching on your career move alongside it, or pause until you feel ready. Which would feel better to you?"

Script 2: pausing for health

"You've mentioned sleeping three or four hours a night for a few weeks now. Before we push on with the plan, I'd really like you to see your doctor about that. I'm happy to pause our sessions for a couple of weeks at no cost to you, so you're not using up your package while you get that checked."

Script 3: the between-session check-in after silence

"Hi Sam, no tasks, no catching up needed. I just noticed it's been a little while and wanted to say I'm here. If things have got busy or heavy, that's okay. Want to keep Thursday, move it, or take a week off?"

Script 4: when safety is at risk

"Thank you for telling me. I'm glad you did. Right now the most important thing is that you're safe, and that's beyond what I can help with as a coach. Can we call [crisis line] together now, or is there someone who can be with you?"

Fill in the crisis line for each country you serve before you need it. If you believe someone is in immediate danger, contact emergency services. This is not clinical training; if you work in a niche where distress is common, a recognised mental health first aid course is worth taking.

Script 5: ending an engagement that is not working

"I want to be honest with you. I don't think our sessions are giving you what you need right now, and I don't want you to keep paying for something that isn't helping. I'd like to refund the remaining sessions and suggest two people who might be a better fit. How does that land with you?"

What should your coaching agreement say about harm and referral?

It should say what coaching is, what it is not, what you will do if you think a client needs other help, and how either side can end the engagement. Signed before the first session, these clauses make the later conversation easier because you are pointing back to something you already agreed. The policy text below is a starting point, not legal advice; have a local attorney review your agreement, because consumer and cancellation law varies by state and country.

Nature of coaching. Coaching is a partnership focused on your present goals and future actions. It is not therapy, counselling, medical care, legal, financial or tax advice, and it does not diagnose or treat any mental or physical condition.

Your health. If you are currently receiving treatment for a mental or physical health condition, please tell me before we begin. I may ask you to check with your clinician that coaching is appropriate alongside your treatment.

Referral. If at any point I believe another professional would serve you better, I will tell you and suggest options. We may pause or end coaching in that case, and any unused prepaid sessions will be [refunded / held for up to X months].

Safety. If you tell me something that suggests you or someone else is at immediate risk of harm, I may contact emergency services. Outside that, what you share stays confidential, except where the law requires otherwise.

Ending early. Either of us may end the coaching with [X days] written notice. [State your refund terms clearly.]

No guaranteed outcome. Results depend on many factors, including your own actions and circumstances. I do not guarantee any specific result, income or timeline.

Pair the agreement with an intake form that asks, gently and directly, about current treatment, recent major life events and what support the client has. Example questions:

1. Are you currently working with a therapist, counsellor or doctor on anything related to what you want from coaching?
2. Has anything significant happened in the last six months (a loss, a diagnosis, a move, a job change)?
3. On a scale of 1 to 10, how is your energy most days at the moment?
4. Who else supports you right now?
5. What would tell you, three months from now, that coaching was worth it?
6. Is there anything you would rather we did not work on?

Answers that mention active treatment or a recent major loss are not automatic reasons to decline. They are reasons for a conversation before you take payment. You can see how intake questions and agreements fit together on the coaching forms and questionnaires and coaching contracts pages.

How do you protect yourself as the coach?

With supervision, limits and insurance. The research on coaches is clear that harm runs both ways: coaches who experience more negative effects report more stress and worse sleep weeks later, and supervision during an engagement weakened the link between client harm and coach harm for the more anxious coaches in the experiment. A coach who is exhausted or anxious coaches worse, which brings the risk back to the client.

Practical steps for your first years:

• Find supervision or a peer group. A paid coach supervisor, a mentor coach, or a small group of peers who meet monthly to discuss anonymised cases. The point is to say out loud the cases that keep you up at night.
• Cap your hard cases. If two clients in your current roster are in heavy territory, think twice before taking a third.
• Set contact boundaries. State in your agreement when you reply to messages, and stick to it. Being reachable at 11pm teaches a distressed client that you are their crisis service.
• Write session notes the same day. Brief, factual, focused on what was agreed. They help you notice patterns and they matter if a dispute arises.
• Get insured. The US Small Business Administration describes professional liability insurance as protecting businesses that provide services "against financial loss as a result of malpractice, errors, and negligence" (SBA). Check with a broker what a coaching policy covers in your country.

If you are coaching part time around a job, protect your energy even more carefully; the guide to being a part-time life coach covers how to set that up.

What does this look like with a real client? A fictional example

The following example is fictional. The people and details are invented to illustrate the steps above.

Ines is a career-change coach in her second year of practice. Her sales page used to say "Land a job you love in 90 days." Daniel, a 41-year-old accountant who wants to leave his firm, signs up for a 12-week package.

Weeks 1 to 3. Daniel's intake form mentions a divorce finalised four months earlier, but he says he is fine. Ines notes it. They agree two goals: a decision on whether to retrain or move sideways, and three informational conversations by week six. His weekly check-ins sit at 6 or 7 out of 10 for energy.

Weeks 4 to 6. Check-ins slide to 4, then 3. He misses one task, then all of them, and apologises at length in each session. Sessions drift toward the divorce and how he "ruined everything". Ines notices she is dreading Thursdays.

Week 6. Ines takes the case to her monthly peer group. They point out that she is offering interpretations ("it sounds like you blame yourself because...") and that his sleep comments have come up twice. In the next session she uses the scope script, mentions the sleep, and suggests he see his doctor and consider a therapist. Daniel is relieved. He had been embarrassed to raise it.

Week 7 onward. They pause for three weeks, as her agreement allows, without using his sessions. When they restart, Daniel is seeing a counsellor. Ines narrows coaching to the one decision about retraining. He makes it in week 11.

What Ines changed afterwards. She rewrote her sales page to describe the process and typical timelines instead of promising a job in 90 days. She added "Has anything significant happened in the last six months?" to her intake with a follow-up conversation for any yes. And she now asks every client in session three, "How is this working for you, honestly?"

Nothing dramatic happened in this story, which is the point. The harm the research describes is mostly low intensity and slow: a client quietly feeling worse, a coach quietly burning out. Catching it in week six instead of week twelve is the whole job.

How do you set this up in Coachful when you are starting out?

You can run every step above with paper, a calendar and email. What a coaching platform adds is that the warning signs sit in one place rather than across five tools. Coachful is a coaching platform for coaches who deliver programs, not just sessions: booking, payments, coaching agreements, programs with habits and check-ins, a community, a website and email, with one login for you and one branded app for your clients. It is not HIPAA-certified and is not built for clinical record-keeping, so if you also see therapy clients, keep their clinical records in a clinical system.

Here is how the safe-coaching checklist maps onto it:

• Agreement before the first call. Coaching agreements with e-signature go out before the first session and are stored on the client record, so your scope and referral clauses are signed, not just mentioned.
• Intake questions on booking. Add the screening questions above to your booking page, and the answers land on the client record where you will see them again.
• Check-ins between sessions. Morning and evening check-ins inside a program give you the weekly mood and energy trend that showed Ines something was wrong.
• Notes and a per-client briefing. Notes, goals, check-ins and intake answers sit on one client record. Michelle, the built-in AI assistant, can brief you before each session from that live data.
• Engagement status. Each client shows as thriving, active, quiet or at risk, which is the silence signal from the warning-signs table.

When a client is flagged, Client Lab shows a drafted check-in message for you to read and approve before it is sent. Treat the draft as a starting point and rewrite it in the tone of Script 3; never send anything to a struggling client that you have not read.

This short walkthrough shows the client area, where those records and flags live:

There is also an AI Coach option that selected clients can chat with between sessions. If your niche brings in clients with heavy material, think carefully before switching it on for them, for the same scope reasons a human coach must. The article on what an AI life coach can and cannot do covers those limits.

Why Coachful stands out: a free white-glove service. On every plan, you can describe what you want built (an intake form, a coaching agreement flow, a program with weekly check-ins, a website) and the Coachful team builds it inside your own account, usually within about a day. Nothing is invented: the team leaves gaps for your own credentials, prices and testimonials rather than making them up, which suits a coach trying to market honestly. Read more about the white-glove build service. Coachful runs on flat monthly plans with a free trial; for current pricing, Lite: $29/month for up to 5 clients. Coaches in their first year can start from the page for new coaches.

What mistakes make coaching more likely to cause harm?

Most of these come from good intentions: wanting to help, wanting to fill a calendar, wanting to be the person who finally gets the client unstuck.

1. Taking every client who books. Early on, every enquiry feels like income. Saying "I don't think I'm the right fit, here is who might be" is a skill, and it protects your reputation as much as the client.
2. Building packages that cover everything. The 2014 data linked more topics with more negative effects. Narrow promises are safer and easier to deliver.
3. Never asking how the relationship is going. Across the studies, higher relationship quality went with fewer negative effects. Ask, and make it safe to answer honestly.
4. Playing therapist because the client asked you to. "You're better than my therapist" feels flattering. It is a scope warning.
5. Diagnosing in passing. Labels like "trauma response" or "narcissist" can stick with a client for years. Describe behaviour, do not diagnose.
6. Using testimonials you cannot stand behind. One client's exceptional result presented as typical is both misleading under FTC guidance and a source of shame for the clients who do not match it.
7. Coaching alone for years. Supervision helped in the research. Peers catch blind spots you cannot see from inside the relationship.
8. No exit route in the agreement. Without a pause or refund clause, a coach who should stop has a financial reason not to.

How do you know coaching is helping and not harming?

Ask regularly, measure a few simple things, and treat a lack of progress as information, not failure. A coaching engagement that is working usually shows steady or rising check-ins, goals the client can describe in their own words, and sessions that end with a clear next step the client chose. One that is harming usually shows the opposite: falling engagement, vague goals, and sessions that end with you having done most of the talking or the feeling.

Three questions to ask at session three and every month after:

• "What has been most useful so far, and what has been least useful?"
• "Is there anything I do in our sessions that doesn't work for you?"
• "On a scale of 1 to 10, how confident are you that coaching is the right support for you right now?"

If the third answer is below six, have the referral conversation, even if everything else looks fine. If you are tracking hours toward an ICF credential, a written record of sessions and client feedback also helps, and the ICF hour tracking page explains how Coachful counts them.

Frequently asked questions

Can coaching be harmful to clients?

Yes. In a 2014 study of 123 coaching engagements rated by their coaches, there were on average two negative effects for the client per coaching, mostly low in intensity. A later review of nine studies found negative effects were common, few were severe, and they happened less when the client-coach relationship was strong.

What are the most common causes of harm in coaching?

Coaches in the research named client factors, their own handling of a client's mental illness and lack of time. Covering more topics in one engagement was also linked to more negative effects. In practice that maps to poor fit, scope drift and overloaded packages, plus pressure from unrealistic promises.

Can coaching make anxiety or depression worse?

It can if a client who needs treatment is given goals and accountability instead, and then feels they are failing at those too. Coaching is not a treatment for anxiety or depression. If a client shows signs of either, encourage them to see a doctor or licensed therapist and decide together whether coaching should continue alongside treatment, pause or stop.

Is coaching harmful to the coach?

It can be. In one survey, 99 per cent of business coaches had experienced at least one negative effect from their work over their careers, and a time-lagged study found coaches with more negative effects reported more stress and impaired sleep eight weeks later. Supervision and limits on hard cases help.

When should a coach refer a client to therapy?

When sessions keep returning to the past or to trauma, when the client shows signs of a mental health condition, when wellbeing keeps sliding over several weeks, or immediately if there is any risk of self-harm. A referral can sit alongside continued coaching on clearly separate goals, if the client and their clinician agree.

Do I need insurance as a coach?

It is strongly advisable. The SBA describes professional liability insurance as protecting service businesses against financial loss from malpractice, errors and negligence. Whether it is required, and what a policy covers, depends on where you work, so speak to a broker in your country.

Can I guarantee results in my coaching marketing?

Be very careful. The FTC warns consumers that offers promising guaranteed income or a proven system are likely scams, and its Endorsement Guides require testimonials with exceptional results to be typical or to state what people generally achieve. Describe your process and realistic timelines instead. Check local consumer law with an attorney outside the US.

What is Coachful's free white-glove service?

On every Coachful plan, you can ask the team to build things inside your account, such as an intake form, a program with weekly check-ins, a website or an email sequence. It usually takes about a day. The team does not invent credentials, testimonials or prices, and it does not include recording videos, logo design or ongoing marketing.

Can I use Coachful to store therapy notes?

No. Coachful is not HIPAA-certified and does not offer a business associate agreement, so it is not suited to clinical record-keeping. It is built for coaching records: goals, check-ins, coaching notes, sessions and payments. Keep any clinical notes in a clinical system.

Is this article legal or medical advice?

No. It summarises published research and official guidance for coaches. Rules on counselling, therapy and consumer protection vary by state and country, so check with an attorney, a licensed clinician or your professional body for your situation.