How to choose a therapy provider for your organisation: the questions to ask before you sign

If you are the person who has been asked to "sort out some counselling" for your team, or for customers who have been through something, you are probably about to have four or five conversations with providers who all sound broadly the same. They all say their therapists are qualified. They all say it is confidential. They all have a page about wellbeing with a photograph of someone looking calm.

The differences between them are real and they are large, but almost none of them are visible on a website. This guide is the set of questions that surface them. It is written from the buying side, and several of the questions are ones that we, as a provider, find harder to answer than a marketing page would suggest. That is deliberate. A checklist that only asks questions your favourite provider answers well is not a checklist, it is a brochure.

Work out what you are actually buying

Before you compare anyone, be clear about which of three quite different things you need. Organisations routinely start a procurement for one and discover halfway through that they needed another.

• An ongoing benefit. Therapy funded as a standing thing, used at whatever rate your people use it. Slow burn, budgeted annually.
• A response to something that happened to your staff. A death, a serious incident on site, a redundancy round, or a customer-facing team that has spent six weeks absorbing other people's anger. Bounded, urgent, and shaped around a specific group.
• A response to something that happened to your customers. A recall, a safety failure, a public incident. Urgent in hours rather than weeks, and with an extra requirement most buyers do not anticipate: the support has to be visibly independent of you.

These need different contracts, different lead times and sometimes different providers. We have written up how each one is put together on our pages for incident and change support for staff, crisis response for customers, and employer-funded therapy as an ongoing programme. If you are still working out which shape fits, the overview for organisations is the shorter read.

Which register is each practitioner on?

This is the first question to ask and the one that separates providers fastest, because there is a gap in UK law that most buyers do not know about.

The titles "counsellor" and "psychotherapist" are not legally protected in the United Kingdom. Anyone may use them. There is no statutory register, no legal minimum training, and no offence committed by someone who calls themselves a psychotherapist after a weekend course. Titles such as "practitioner psychologist", "clinical psychologist" and "counselling psychologist" are protected, and restricted to people on the Health and Care Professions Council register.

What follows from that is simple. The sentence "our therapists are fully qualified" carries no information at all. The sentence that carries information names a register and offers a number you can check.

Ask, in these words: which professional body is each practitioner registered with, and can I have registration numbers to verify? The answers you are looking for are BACP, UKCP, NCPS, HCPC or BPS. All of those bodies publish a searchable register. A provider who cannot give you numbers, or who answers with a training qualification instead of a registration, has told you something important.

We have set out what each body signifies, and how we verify them, on our clinical governance page. It is written to be forwarded to a legal or compliance team without editing, which is the only sensible format for this kind of information.

Insurance, vetting and who actually checked

Two follow-up questions that get skipped.

Is professional indemnity insurance current, and who verified it? Registration and insurance are separate things. A therapist can be registered and uninsured. Ask whether the provider checks cover at onboarding, and whether they check again when it lapses.

Did a human being speak to each practitioner before they saw a client? There is a meaningful difference between a platform that collects an uploaded certificate through a form and one that puts every therapist on a video call to verify identity, registration and insurance before they are allowed near anybody. Ask which of the two you are buying, because both are described as "vetted".

Who can hear a session, and what comes back to you

Ask two questions here and insist on specific answers.

What exactly will our organisation receive? The right answer is anonymised aggregate reporting and nothing else: sessions offered, sessions taken up, sessions delivered. If a provider offers you attendance data, named lists, department-level breakdowns small enough to identify a person, or any form of "insight" into what people are discussing, that is not a feature. It is the thing that will stop your staff using the service the moment one of them notices.

Who other than the therapist can hear or access a session? Ask it plainly, ask about supervision arrangements, about quality monitoring, about training, and about whether anyone from the client organisation has ever been given access to live or recorded sessions. Ask whether the provider's own organisational accreditation has ever been suspended or investigated. These are not hypothetical questions in the UK market, and a provider with nothing to hide will answer them without flinching.

Your own staff will judge the programme almost entirely on whether they believe this part. It is worth being able to repeat the answer accurately when you announce it.

Where the health data sits

Bookings and session records are special category health data under the UK GDPR, a category the Information Commissioner's Office treats as needing particular care precisely because of the harm its disclosure can do.

Some providers will offer to run the service inside your own app or platform, and it can be presented as a nice piece of integration work. Think carefully before accepting. If those records sit in your systems, your organisation becomes the holder of health data about your own staff, or about customers who may be unhappy with you and may later be in dispute with you.

There is a second reason, and for a customer-facing programme it is the decisive one. Support that is visibly independent of the organisation paying for it is the only kind people trust enough to use honestly. Route it through your systems and take-up falls, whatever your privacy notice says. Independence is not a compliance nicety in this situation. It is the mechanism by which the thing works at all.

How many sessions does a person actually get?

This is the question that decides whether the programme does anything measurable, and it is almost always settled by dividing a budget by a headcount rather than by looking at the evidence.

The outcome research is clearer than most people expect. A 2020 systematic review of twenty-six studies of psychological therapy delivered in real services found optimal doses ranging between four and twenty-six sessions, with four to six for low-intensity guided self-help. The classic dose-effect study, across more than 2,400 patients, found roughly half were measurably improved by session eight and roughly three quarters by session twenty-six.

Set against that, a programme offering everybody one or two sessions is buying something that sits below the bottom of every range in the literature. It will produce a good number for "people offered support" and very little else.

The honest complication is worth knowing too, because a provider who only tells you the convenient half of the evidence is telling you something about themselves. One large UK primary care study found that the rate of reliable improvement did not rise with the number of sessions attended, running from 88 per cent among people who came once down to 62 per cent among those who came twelve times. That is not evidence that less therapy works better. It is evidence that people stop coming when they have got what they came for, which is exactly why the ending should be a clinical decision rather than a budget one.

We have set all of this out with the sources attached on how many therapy sessions actually help. If you read one thing before setting a budget, read that.

Does the person choose their therapist, or get allocated one?

Ask how a person ends up with a particular therapist. There is a real difference between being allocated whoever is next available and choosing after a short, free conversation.

It matters commercially as well as clinically. A poor match burns the first two sessions of any course, and in a funded programme those are your sessions. Anyone who has been through it knows the feeling of spending two appointments deciding whether to continue rather than doing any work. Our blog post on red flags and green flags in therapy covers what a good match actually looks like from the client's side.

Ask whether people can change therapist without a fuss, and what happens to the funded allowance if they do.

How fast can they start, and how fast do they answer?

Two separate questions, and the second one predicts the first.

Ask what the wait is between someone deciding they want a session and actually having one. Ask it about the specific week you would be starting, not in general. Then notice how long the provider took to reply to your own enquiry, because that is the same organisation, on a good day, when they are trying to win your business.

If the situation is live, this stops being a nice-to-have. An organisation dealing with an incident at four in the afternoon is choosing a provider by six, and the deciding factor is usually just who picked up. We publish our own commitment on every page in our business section: a reply from a person within two hours, eight in the morning to eight in the evening, seven days a week, with WhatsApp as the fastest route. Ask every provider you are considering for their number, and then hold them to it during the sales process, when it should be at its best.

What happens to the money when nobody uses it?

Providers charge in two broadly different ways, and the difference shows up in year two.

A traditional employee assistance programme is typically a per-employee annual licence. You pay for coverage across the whole headcount whether or not anybody uses it, which makes budgeting easy and makes low usage invisible and free for the provider.

The alternative is paying for the sessions that were actually delivered. Budgeting is slightly less predictable, though a monthly cap solves most of that, and the incentives point the right way: a provider only earns when someone is actually being seen.

You will find a lot of confident numbers quoted about EAP utilisation rates. Ask any provider who quotes one where it comes from, and treat an unsourced figure as marketing. We do not publish a comparison table for the same reason, which is that we cannot independently source the market-wide numbers, and a table built on figures we cannot stand behind would be worth less than nothing. What we can tell you is how our own billing, capping and invoicing works, in detail, before you commit to anything.

What happens when the funding runs out?

This one is almost never asked and it produces the worst experiences in a funded programme.

Someone has six funded sessions. They have found a therapist they trust. On session six the funding ends. What happens next?

The answers you want: the person was told the number at the start rather than discovering it at the end, and they can continue with the same therapist privately if they want to. The answer you do not want is a handoff to a different service at exactly the point they were getting somewhere, which is a reliable way to teach someone that opening up is not worth the trouble.

Ask about the cost of continuing privately, too. If your programme ends and the private rate is out of reach for the people who were using it, you have funded an introduction to something they cannot have. Our guide to affording private therapy in the UK covers what people are realistically looking at, and it is worth knowing that number before you set an allowance.

Answers that should worry you

• "All our counsellors are fully qualified." Not an answer. Ask which register and for numbers.
• "We can give you full reporting on usage." Ask what "full" means. If it can identify a person, it will suppress take-up.
• "We can host it inside your platform." Ask who becomes the data controller, and whether the people using it will believe it is independent.
• "Studies show our approach reduces absence by X per cent." Ask for the study. Then read it.
• "This will reduce your liability." It will not, and a provider willing to say so is willing to say other things that are not true.
• Case studies with no client named and no outcome measured. Either the client agreed to be named or they did not. A nameless success story is a story.
• A quote that arrives without anyone asking how many people are affected. The shape of the programme determines the price. A provider who quotes before asking is selling a product, not designing one.

One thing we will say about ourselves

MatchyMatch is early in the organisational market. We have a vetted UK panel, a platform that already runs bookings and payments, and we answer fast. We do not have corporate case studies, and we are not going to invent any, because inventing social proof would destroy the only thing worth selling to someone making a decision about vulnerable people, which is that we can be trusted to tell the truth about what we can do.

If you want to test that, ask us the hardest question on this list and see how the answer compares to the others you get. The business section lays out the three programme shapes, the clinical governance page answers the compliance questions in full, and the first conversation is free and takes about fifteen minutes.

And if what you actually need is help for one person rather than a programme, they can find a therapist and speak to them free for fifteen minutes without anybody at work being involved at all.