AI Coaching Services for Corporate Use: Ten Providers and How to Choose

AI coaching services for corporate use come in four quite different shapes, and most buyers compare them as if they were one product. A tool that nudges thousands of first-line managers and a service that supports a CEO through a board dispute are not competing for the same job, so a ranked list of the best tells you very little. The better question is which kind of service matches the leaders you want to develop, and what the vendor will and will not let the AI decide.

I have spent 25 years on leadership architecture work, and I run an AI coaching product of my own. I am not against these tools. I am against buying them for the wrong reason, because the decision that matters most is what you hand to the machine and what stays with a person. This page lists ten services, and the first one is mine. I put my own service first so that the conflict of interest is visible, and the order is not a ranking. I have not ordered the other nine by quality, size or fit.

How I chose these AI coaching services

Every description below comes from the vendor's own website, and each entry links to the page it relies on. Where a vendor says something about itself, I attribute it to the vendor, because I cannot tell from a website whether a claim is true. I did not use reviews, rankings, price lists or third-party write-ups, so nothing here compares one vendor with another or tells you how well any of them works. Websites change, so read the current page before you rely on any entry, and treat the list as a starting point for your own questions rather than a verdict.

Ten AI coaching services for corporate use

The ten entries follow. Mine comes first, and the other nine carry no order of merit. Words in quotation marks are the vendor's own, copied from the page named at the end of each entry. Each entry also carries one question worth asking that provider, whatever the answer turns out to be.

1. Stuart Andrews: Executive coaching and CapabilityAI

My site presents two things that belong in this conversation, on two separate pages. The first is my executive work. The work with me page describes engagement options for CEOs, CHROs and executive teams, by application only. Its Private Architecture model runs for 12 to 24 months and includes unlimited 1:1 access to me.

The second is CapabilityAI, which its page describes as an AI leadership coach grounded in my 25 years of methodology. The page says it remembers your context, says it is built for founders, consultants, fractional executives and owner-operators who make every call alone, and says it is free to start. It also labels access as early access for a founding cohort, so check what is open on the day you read this. Every voice call on it is with an AI, not with me, and the page says the call announces this at the start. CapabilityAI is not a network of human coaches. I would not use it in place of a human coach for a high-stakes decision, and I do not position it that way. Because I run this service, discount my view and put the five criteria below to it as you would to anyone else.

2. BetterUp

BetterUp's homepage says its coaches "work with leaders and managers", and describes a second layer in these words: "Always-on AI coaching meets workers in the moment through the apps they use every day". So BetterUp presents live coaching and AI coaching as two parts of one platform. A fair question for any provider that offers both is how the two connect: what the AI knows about the live sessions, and what the coach sees of the AI conversations. Source: betterup.com.

3. CoachHub

CoachHub calls itself "The Digital Coaching Platform", and its homepage refers to "human, AI and hybrid coaching". Its product menu lists individual coaching, collective coaching, executive coaching, AIMY™ AI Coaching and hybrid coaching. When a vendor uses the word hybrid, ask what it means in practice: which parts of the experience are a person, which are software, and who decides when a leader moves from one to the other. Source: coachhub.com.

4. Ezra

Ezra's homepage headline reads "An enterprise coaching platform built to scale", and its page description promises "Scalable 1:1 coaching for every level" and "Measurable impact, powered by AI". The site names offers including EZRA Edge and EZRAx. The EZRAx page describes "Unlimited 1:1 human coaching tailored to the realities of senior leadership". If you are buying for a mixed population, ask which offer would cover which group of leaders, and what changes between them. Source: helloezra.com and its EZRAx page.

5. Sounding Board

Sounding Board describes itself as "Your strategic partner for leader transformation", offering "Leadership development through coaching solutions, mentoring management, and flexible program software". Its homepage carries the headline "BTS acquires Sounding Board to serve global companies end-to-end, executive to frontline coaching needs". As I read its homepage, it does not describe an AI coaching product; the AI items it shows are content pieces, including a download titled "Leading Digital Transformations in the Age of AI". I include it as a human-led reference point, because comparing AI offers is easier when you also see how a coaching and mentoring provider presents itself. Ask who owns the relationship with your leaders after the acquisition it announces. Source: soundingboardinc.com.

6. Torch

Torch's homepage describes its offer in these words: "Connect experienced human coaches, Spark AI and your business priorities." The page also says "Spark AI keeps the development moving between sessions". Before you choose, ask what Spark AI draws on, what it keeps, and whether a coach can see what a leader has discussed with it. Source: torch.io.

7. New Level Work, formerly BetterManager

This company was called BetterManager. Its own site carries a news post dated 23 January 2024 which reads "Leadership development platform BetterManager today announced it has rebranded as" New Level Work, and the old bettermanager.co address now redirects to newlevelwork.com. The homepage menu lists 1:1 leadership coaching, group coaching, group training and Leora AI, and presents Leora AI under the heading "Human-Led Coaching, Powered by AI". If you meet the old name in a supplier list, check that you are looking at the same company and the current offer. Source: newlevelwork.com and its rebrand post.

8. LEADx

LEADx's homepage describes a "micro-coaching platform" built around "nudges, micro-coaching, and micro-learning". Its menu lists "AI for Managers" alongside EQ assessments, EQ training and EQ certification. Short, frequent prompts suit habit building, so ask how the vendor shows that a prompt changed what a manager did, and not only that the prompt was opened. Source: leadx.org.

9. CoachVantage

CoachVantage calls itself "All-in-One Coaching Software" and carries an "AI-Enabled" label on its homepage. The page says "Everything you need to run and grow your coaching business", and its pitch is addressed to coaches: "Built For Coaches At Every Stage Of Their Coaching Journey". Check whether this is the right kind of product for the purchase you have in mind, because software for running a coaching practice and a coaching service for your employees answer different needs. Source: coachvantage.com.

10. Valence

Valence's homepage carries the title "AI Coaching for Every Manager, Tailored to Your Company", and introduces its coach, Nadia, as "An AI coach built by an AI company". Ask what Nadia knows about each manager, who in your organisation can see it, and what the provider does with those conversations. Source: valence.co.

How to choose AI coaching services without comparing the wrong things

Start with the leader, not the product. Write down who the service is for: new managers, experienced people leaders, senior executives, or an intact executive team. Then write down what you want to change in their behaviour in the next two quarters. Only then look at vendors. The usual trap is the opposite order, where the feature sheet arrives first, the dashboard demo is impressive, and the question of who actually benefits gets answered by whoever has the most persuasive deck.

Next, separate the technology from the coaching. In any service, some of the value comes from software (availability, pattern detection, record keeping, scale) and some comes from judgement (challenge, context, a read of the room). A good buyer asks the vendor to say, in plain language, which is which. If a vendor cannot split the two, you may end up paying coaching prices for software or software prices for coaching, and not know which.

Finally, decide how you will know it worked before you sign. Satisfaction scores are easy to collect and weak as evidence. The more useful measures are changes in decisions, in the quality of conversations, and in what the leader's own team says about them over time. If you cannot name your measure on day one, you will end up reporting usage, because usage is what a dashboard offers.

The four kinds of AI coaching service

I sort offers into four shapes. This is my own framework, not an industry standard, and a single vendor can straddle two of them. Knowing the shape tells you what to expect before you read a feature list, and it explains why two services with similar marketing can behave differently once your people start using them.

Standalone AI coaching assistants

These are software only. A leader talks or types to an AI, usually at any hour, and gets prompts, reflections and suggested practice. The strengths are availability, low cost per person and no scheduling. The limit is just as clear: there is no one in the loop who knows the leader's organisation, and nobody who will tell them their plan is weak and mean it.

This shape suits practice and rehearsal: preparing for a difficult conversation, thinking through a decision out loud, building a reflective habit. It is a poor fit where the stakes are high and the context is messy. Ask the vendor what the assistant does when a leader describes something that sounds like a legal, safety or wellbeing issue, because the answer shows how seriously the product has been designed.

Human coaching platforms with an AI layer

Here a network of human coaches is the product, and AI sits around the edges: matching leaders to coaches, scheduling, summarising sessions, tracking progress across a cohort. For a company that wants coaching to reach several hundred people, this can be the practical route, because the human relationship is preserved and the administration is automated.

The risk is that scale pushes a model toward coach supply rather than coach calibre. A large roster says very little about whether a particular coach has ever worked with someone in your position. If you choose this shape, ask how coaches are matched, what their backgrounds are, and what the AI is allowed to feed back to the organisation about individual leaders.

Practice and nudge tools for managers at scale

These products use short prompts, scenarios and reminders to build day-to-day management habits across many people at once, and some add human coaching. Their strength is reach and repetition. In my experience, development that arrives as a single event rarely changes behaviour, and rhythm does, so this shape suits the problem it targets.

The limit is depth. A nudge can remind a manager to give feedback this week. It cannot work out why that manager avoids feedback in the first place. Use this shape for foundational capability across a wide population, and pair it with something more personal for the people carrying the heaviest decisions.

Practitioner-led AI companions

The fourth shape is an AI companion built on one practitioner's method and knowledge, so the AI reflects a defined point of view instead of a generic model of coaching. It may be offered alongside access to the practitioner, or as an extension of their work. The advantage is coherence: the questions, the language and the framework stay consistent between sessions.

The risk is dependence on one person's thinking. If you do not share that practitioner's approach, the product will feel wrong however well it is built. Ask to see sample output, and judge whether the AI is applying a real method to your situation or producing fluent, generic encouragement.

Five criteria that decide whether an AI coaching service fits

Whichever shape you shortlist, the same five questions separate a serious provider from a persuasive one. I use them when a client asks me to sanity-check a vendor. They are deliberately about decisions and consequences, not features.

Who wins when the AI and the human disagree

Ask directly: when the AI's pattern-matching and the human coach's read of the situation point in different directions, which one prevails? In a service with no human, the answer is by definition the AI, which is acceptable for rehearsal and a risk for real decisions.

Where there is a human, I would want the answer to be that the human decides and the AI informs. A hesitant reply is worth noting. Real leadership decisions are contextual, not statistical, and a service designed as if they were statistical will mislead the people who trust it most.

What the coaches did before they were coaches

Where the service includes human coaches, the size of the roster is a weak measure. What matters is the experience behind each coach. A coach who has run a business unit, sat through a restructure or handled a board conflict coaches differently from one whose whole career has been coaching.

For a first-line manager, a well-trained generalist is often enough. For a CEO or an executive team, ask for specifics about similar work the coach has done, and talk to the person who will actually sit with your leader, not the person selling the contract.

What the AI actually draws on

Ask what the AI is built on and what it knows about your leader. Is it drawing on that leader's real behaviour, feedback and goals, or running them through a generic competency model and calling the output insight? Request sample output for a realistic scenario. Generic language is the tell: advice that could be handed to anyone is advice for no one.

Also ask how the product handles being wrong. Any AI will sometimes state something with confidence that is not true or not helpful. A serious provider can describe how errors are caught, how leaders are told to treat the output, and how feedback improves it.

How confidentiality is built, not just promised

A privacy line on a terms page is easy to write. What you need is the architecture. Ask what data the AI layer keeps, who inside your organisation can see it, whether it is shared with the leader's manager or HR, and whether your conversations are used to improve models that serve other clients.

There is a public reference point. The International Coaching Federation published its AI Coaching Framework and Standards in November 2024. The ICF describes bias and confidentiality as ethical risks of putting AI into coaching, and its standards cover six domains, including technical factors such as privacy and accessibility. It is a useful checklist to hand a vendor, and I make no claim here about which providers meet it.

This matters more than most buyers expect. Leaders will only use a coaching service honestly if they trust it is private. If they suspect their candid reflections feed a report to their boss, they will stop being candid, and you will have paid for a tool that records performance theatre. Ask for the data policy in writing and have your security and legal teams read it before the pilot, not after.

Whether the provider admits its own limits

I trust a provider more when it says plainly where the AI stops and the human starts. If a vendor told me its AI replaces senior coaching, I would read that as marketing, not method. Look for clear statements about what the product is not for, such as clinical mental health support, legal advice, or decisions with irreversible consequences.

Ask what the provider measures and what it will share if the results are poor. A willingness to be measured against changes in behaviour, rather than hours of usage, says a lot about how confident the provider is in the service.

Matching the service to the leader

Seniority changes the answer more than any other variable. For a population of new and early-career managers, the strongest case is usually a nudge and practice tool or a human platform with a strong AI layer, because they need frequent, low-risk practice and cannot all have a personal coach. Cost per person and ease of use matter most.

For experienced people leaders, the balance shifts toward human interaction with AI support between sessions. They already know the basics. What they lack is a challenger who will point at the pattern they cannot see, supported by a record of what happened between conversations so that time together is spent on the right topic.

For senior executives and intact executive teams, the human relationship is the product. AI can improve preparation and follow-through, and I think it should. But if you are paying for judgement on a board dispute, a succession decision or a failing leadership team, the person is what you are buying, and the AI is an aid. I have written more on this in how to choose an executive coach that fits.

Where AI helps and where it cannot replace a human coach

AI is useful for the high-volume, unglamorous parts of coaching. It notices patterns across dozens of interactions faster than any human can. It keeps a record of what actually happened between sessions instead of relying on a leader's selective memory of their own week. Used well, it makes human coaching time sharper, because you walk in already knowing what to talk about.

Where it stops dead is judgement under ambiguity. A leader deciding whether to part ways with a co-founder, restructure a team mid-crisis, or push back on a board that is wrong is not facing a pattern-matching problem. They need someone who has sat with consequence before, who can read the room that is not in any dataset, and who is willing to say something the client does not want to hear. No model does that, because that is not what models are for.

So my quotable position is this: AI coaching tools are a very good research assistant and a poor confidant. If a provider sold AI as a replacement for senior human judgement, I would see a research assistant wearing a coach's badge. For more on dividing decisions between people and machines, see the human-AI decision ladder and decision rights for AI-assisted decisions.

A sensible way to run the buying process

Corporate buyers get the best results by treating this as a pilot, not a procurement. A small, well-chosen group, a clear measure and an agreed end date will tell you more than any demo. These steps keep the process honest.

1. Define the population and the purpose: Name who the service is for and the specific behaviour change you want, before talking to vendors.
2. Shortlist by kind, not by brand: Choose the shape of service that fits the leaders, then compare two or three providers inside that shape.
3. Put the five criteria in writing: Ask each vendor to answer the five questions in writing and compare the answers side by side.
4. Run a small pilot with a clear measure: Use a group of leaders, agree the measure up front, and include a few sceptics who will tell you the truth.
5. Check data and exit terms: Have security and legal review data retention and sharing, and agree what happens to the data when the contract ends.
6. Decide with evidence: Judge the pilot on changes in decisions and conversations, not on logins and satisfaction scores.

Treat the pilot result as information, not a verdict on AI coaching in general. A service can fail for your organisation because it was the wrong shape for the leaders involved. If you are weighing whether coaching of any kind is the right intervention at all, read when executive coaching is the wrong intervention first, and see how my own CapabilityAI fits the fourth shape above.

If you want a second opinion on a shortlist, or help working out which kind of service your leaders actually need, you can talk to me directly. I will tell you honestly if the answer is not something I sell.

Related reading

• What is corporate coaching and what it covers
• 10 benefits of corporate coaching programmes
• Executive coaching ROI and whether it is worth the investment
• Why leadership development programmes fail