There are now several AI consultation practice tools aimed at GP registrars, and they are not all the same. We will be upfront: we build one of them, so we are not a neutral party. But the most useful thing we can tell you is not “choose ours”; it is how to judge any of them for yourself, so you end up with the one that actually fits how you learn.

Try more than one

The single best piece of advice is also the simplest: try a few before you commit. Most reputable tools offer a free case or trial. Use them. The experience of actually consulting through each platform tells you far more than any feature list, because the thing that matters most, the quality and usefulness of the feedback, is hard to judge from the outside.

You are not looking for the tool with the longest feature list. You are looking for the one whose feedback you find useful, in language you respond to, on the things you actually need to work on.

Judge the feedback, not the features

When you trial each tool, pay attention to the feedback specifically. A few questions worth asking as you go:

Does the feedback tell you why, not just what? “Your data gathering was weak” is a verdict; “you gathered thoroughly but did not check the patient’s understanding before moving on” is something you can act on. Is it specific to your consultation, or could it have been written about anyone? Does it point you to a clear next step? And does the style suit you — some registrars want warm and encouraging, others want blunt and direct, and the right tool is the one whose tone keeps you coming back.

Two kinds of feedback worth understanding

Tools tend to give feedback in different ways, and it is worth knowing the distinction so you can decide what helps you.

Some give domain-level feedback: a structured judgement across the consultation domains (data gathering and diagnosis, clinical management, relating to others), which maps neatly onto how the exam is marked and helps you see which broad area to focus on. Others quote your exact words back to you, showing you the specific phrase where you closed down a question too early, or where your safety-netting was vague. The first helps you see the shape of your performance; the second helps you catch the precise moments where it slipped.

Neither is better in the abstract. We happen to do both, because we think the domain view tells you where to look and the exact-words view tells you what to fix. The point for you is to notice which kind of feedback you personally learn most from, and weight your choice accordingly.

Other things worth weighing

Beyond feedback, a few practical factors separate the tools: whether you get multiple attempts at a case or are rationed by credits; whether it works hands-free on your phone or ties you to a desktop; whether it looks across your cases or only grades them individually; and how long your access lasts relative to the months your preparation will actually take.

Weigh those against how you intend to study. A registrar who wants to drill a handful of cases intensively has different needs from one building broad coverage across the curriculum.

The honest bottom line

Whatever you choose, make sure it works for you — not for the registrar who recommended it, whose weak points and learning style may be nothing like yours. Try a few, read the feedback critically, and back the one that moves your consultations forward. That is the right way to choose, and we are confident enough in what we have built to recommend judging it the same way as everything else.