“Personalised” is one of those words that’s been marketing-speak for so long it barely means anything anymore. Every app claims to be personalised. Most of them just mean they remember your name.
So when AI meditation claims to offer personalised sessions, you’re right to be sceptical. What does that actually mean in practice?
Let me walk you through exactly how personalisation works — at least how it works at InTheMoment — and why it genuinely changes the meditation experience.
The check-in conversation
Every session starts with a brief check-in. This isn’t a form or a mood slider. It’s an actual conversation.
Think of it like meeting a friend for coffee. They ask how you’re doing. You share a bit about your day, what’s on your mind, how you’re feeling. The conversation flows naturally — usually about 5 exchanges.
You can go as light or as deep as you want. Mention that you’re sitting on a train heading to a difficult meeting. Or just say you’re feeling tired and want to relax. The more context you provide, the more tailored your session becomes.
Here’s the key insight: this conversation isn’t just collecting data. It’s building context. The AI understands not just what you said, but the situation you’re in.
What gets personalised
So what actually changes based on your check-in? Several things:
Teaching focus
If you mention you’re anxious about a presentation tomorrow, the session might focus on techniques for pre-performance nerves. If you share that you’re grieving a loss, the teaching might centre on processing difficult emotions.
This isn’t generic “calm down” advice — it’s teaching that directly addresses your situation.
Technique selection
InTheMoment draws from a range of established meditation techniques: breath awareness, body scanning, loving-kindness, visualisation, open awareness, and more. Different techniques work better for different situations.
Feeling physically tense? Body scan or progressive relaxation might be emphasised. Racing thoughts? Breath counting or open awareness might help. Emotional turbulence? Loving-kindness or acceptance practice might be appropriate.
The AI selects techniques likely to help with what you’re actually experiencing.
Environmental awareness
Say you’re on a crowded train. The session won’t tell you to “listen to the silence around you.” Instead, it might guide you to notice the rhythm of the train, the sensation of your feet on the floor — working with your environment rather than against it.
Walking? The session adapts to movement. Lying in bed? Eyes closed, longer pauses, sleep-oriented. Sitting at your desk? Shorter, eyes-open friendly.
Pacing and language
Some people prefer more silence. Others find too much silence uncomfortable. Some want direct instruction. Others prefer gentler invitations.
The system learns these preferences through feedback (more on that shortly) and adjusts how sessions are delivered.
The feedback loop
Personalisation doesn’t stop when the session ends.
After each session, you can provide feedback — again, through a conversational format. What worked? What didn’t? Was anything confusing or unhelpful?
This feedback directly influences future sessions. If you mention that breath counting helps you but body scans feel uncomfortable, subsequent sessions will lean toward breath-based techniques.
It usually takes only one or two feedback sessions before the AI starts consistently matching your preferences.
You can also adjust technique preferences directly from your profile. Enable or disable specific methods. By default, all techniques are available, and the AI chooses what fits. But you have control.
Session continuity
Each session exists in context of your previous ones.
The AI knows what teachings you’ve explored, which techniques were used, and how you responded. This creates continuity — the sense that you’re on a journey rather than starting from scratch each time.
Yesterday’s session introduced breath awareness. Today’s might build on that foundation. Last week you explored loving-kindness. A month later, the teaching might reference back to it.
This addresses one of the biggest frustrations with generic meditation apps: the lack of progression when doing freeform practice.
Structured learning through playlists
For those who want explicit structure, InTheMoment offers playlists — curated curriculums with defined learning objectives.
Each playlist covers a topic: Foundations of Mindfulness, Stress Management, Introduction to Hypnosis, and so on. Within each playlist, lessons have specific content goals.
But here’s where personalisation still applies: the teaching content is defined, but the delivery is personalised. Two people following the same playlist learn the same concepts, but hear sessions adapted to their individual situations and environments.
It’s like having a tutor who follows a curriculum but explains things in a way that makes sense for you specifically.
Why this matters
So why does personalisation actually matter for meditation?
Relevance drives practice. When sessions feel directly connected to your life, they feel more worthwhile. And when practice feels worthwhile, you do it more consistently. Consistency is what generates real benefits.
Technique matching improves effectiveness. Not every technique works equally for everyone. Finding what resonates with you specifically — through experimentation and feedback — means better outcomes from each session.
Environment matching removes friction. If you can only meditate on your commute, you need sessions that work there. If instructions assume you’re in a quiet room with closed eyes, you’ll struggle and eventually stop trying.
Continuity creates growth. When each session builds on the last, you’re developing a practice rather than just having isolated experiences. That’s the difference between learning and just listening.
What personalisation doesn’t do
Let me be clear about the limits:
It doesn’t replace expertise. The underlying teachings and techniques come from established meditation traditions. The AI isn’t inventing methods — it’s selecting and applying proven ones.
It requires your input. Vague check-ins produce generic sessions. The more you share, the better the personalisation. This isn’t passive — you’re an active participant.
It can’t read your mind. The AI only knows what you tell it. If something’s bothering you but you don’t mention it, the session won’t address it.
It won’t fix everything. Personalised meditation is still meditation. It’s powerful for many things but isn’t therapy or medical treatment.
The technology behind it
You might be curious how this actually works technically.
The check-in conversation is processed using natural language understanding. The AI extracts context: your physical situation, emotional state, what’s on your mind, what you’re hoping for.
This context is then used to guide session generation. The system selects from a library of established techniques and creates a script that applies them to your specific situation.
The voice that reads the script is AI-generated — modern text-to-speech technology that’s remarkably natural. At InTheMoment, you can choose from six voices (Rain, Ember, Oak, Skye, Archer, Rowan), all currently available to all users.
Feedback is incorporated through a learning system that updates your preference profile. Over time, the system gets better at predicting what works for you.
All of this happens in seconds, creating a fresh session every time.
Try it yourself
The only way to really understand personalised meditation is to experience it.
Do a check-in. Share what’s actually happening in your life. Listen to the session that comes back. Notice how it addresses your situation specifically.
Then do it again tomorrow, in a different context. Compare how the session adapts.
For most people, the “aha” moment comes quickly: “Oh, this is actually about me.”
That’s what personalised meditation can feel like when it works. Not a generic voice talking about abstract concepts, but a guide who understands your moment and meets you there.
Curious how personalisation feels in practice? Get started with two free sessions per day — no credit card required.