Knowledge base › Tools

Most self-improvement apps don't tell you what they're built on. You get a question in the morning, a chart in the evening, and you're expected to believe it means something. We'd rather write it down. This article shows how Mirrify is built: the loop it runs, where the questions come from, how a pattern becomes a goal, what the AI mentor actually does — and which established practices the modules draw on. At the end you'll also find what we don't know and don't claim.

1. The loop — this is the whole system

Mirrify rests on one idea: self-knowledge is worth something when it becomes a closed loop. Six steps, each working from the same data:

1. Entry. You write (or speak) what happened and what you felt. This is the raw material.
2. Pattern. Across entries it becomes visible what repeats — situation, thought, reaction.
3. Insight. The pattern gets a name. Not "I had a bad week" but "overloaded weeks always end the same way".
4. Goal. The insight becomes a concrete goal with a deadline — not a vague intention.
5. Daily habit. The goal becomes a today-sized step, doable even on a bad day.
6. Review. A weekly and monthly moment where the data returns to the top of the loop: what worked, what didn't.

Most tools stop at a single point of that loop. A journal takes you to step two or three. A habit tracker starts at step five and stays there. A chatbot gives advice but knows nothing about your habits. A system is a system because all six steps live in the same data — so step six genuinely knows what happened in step one.

2. Where do the questions come from?

This is the most common question, and it has two answers.

Guided question sets. The morning and evening routines, the schema journal, the wheel of life and Ikigai each have their own pre-written questions. These provide the frame: you aren't met by a blank page but by a concrete question that decides for you where to begin. The blank page is the number one killer of journaling — so there is no blank page here.

The AI mentor's questions. The deeper modes work from your own earlier entries. That's what makes the question personal: it doesn't ask "what was good today", it asks about what repeats for you. The morning block, for example, draws a one-sentence lesson from yesterday, names today's likely trap and the antidote to it, and once a week a "letter from your future self" closes the loop.

3. What are the modules built on?

Mirrify didn't invent new psychology. It draws on established self-knowledge and coaching traditions and binds them into one system. Module by module:

1. Journal and guided reflection — the tradition of written self-disclosure (expressive writing): putting a difficult experience into words organises the thinking and creates distance from it.
2. Morning pages — Julia Cameron's free-writing practice: in the morning, unfiltered, to empty the head.
3. Gratitude journal — the classic gratitude practice: not positive thinking but directing attention to what is already there.
4. Schema journal — the logic of cognitive and schema-based reframing: naming the recurring thought and behaviour loops, then placing a new response beside them. This is a self-knowledge exercise, not therapy.
5. Goals and habits — implementation intentions ("if situation X, then step Y") and the small-steps principle: a goal is worth as much as its smallest daily piece.
6. Wheel of life — coaching's classic assessment: eight life areas, self-scored, so you can see where the real gap is.
7. Ikigai — the four questions of the Japanese frame, used to sketch direction.
8. Decision journal — reasoning recorded at the moment of the decision, so the outcome can't overwrite the memory.
9. Mood tracking — not for its own sake but as a signal: it shows when to look at the journal.
10. Finances — spending as a behavioural trace: the amount is the symptom, the journal shows the state behind it.

To be precise: these traditions give the modules their shape, not a guarantee of effectiveness. A practice coming from a known methodology doesn't mean it will work in your case — which is exactly why the system has a review step.

4. How does a pattern become a goal?

This is what separates a system from a journal, so it's worth walking through concretely. Say that after three weeks of entries something emerges: your Sunday evenings are reliably anxious.

1. The pattern gets a name. Not "I feel bad on Sundays" but: the unreadability of the coming week triggers anxiety.
2. The wheel of life locates the area. It turns out this isn't a sleep or health question but work organisation.
3. The value supplies the reason. Why you want to change it — calm, predictability, being present with your family.
4. That becomes the goal. Concrete, with a deadline: for eight weeks, next week's focus is ready every Sunday.
5. The goal becomes a daily habit. Fifteen minutes of weekly closing on Friday afternoon — so nothing is left for Sunday.
6. The review closes the loop. Four weeks later you can see whether Sunday anxiety dropped. If not, the insight was wrong — not you.

That chain is what a standalone journal, a standalone habit tracker and a standalone chatbot can never produce — because none of them sees the other three steps.

5. What exactly does the AI mentor do?

The AI here isn't a stand-in for company but a synthesis layer. It does four things:

1. Connects. Your journal, mood, habits, goals and financial entries live in the same system, so it can surface connections that are invisible in separate apps.
2. Remembers for you. You don't have to hold in your head what you wrote six weeks ago — one of the biggest burdens that ends journaling.
3. Changes role. Separate modes look at your life, your work and career, your relationships, your finances and your Ikigai, and separate modes produce the weekly, monthly, quarterly and annual summaries. The same voice shouldn't tell you how to plan a quarter and how to close a day.
4. Quotes. The deeper analyses quote your own words — so you can check what they're based on and don't get generalities.

And what it does not do: it doesn't read through your account. Every analysis is a separate request carrying only the slice belonging to that task — currently at most 30 entries, each up to roughly 800 characters, and for conversations the last 20 turns at most. The exact technical details, including what we store and what we don't, are written up in a separate article.

6. The rhythm — why daily, weekly, monthly and yearly?

Each horizon has a different job, and if any is missing the system tilts:

1. Daily. One guided question, a short entry, three priorities. This is the source of the data — without it there's nothing to work from.
2. Weekly. Setting a focus forwards, and a short closing backwards. This keeps the daily level alive.
3. Monthly. An honest reckoning: what changed, what didn't, what you adjust. This is where direction becomes visible.
4. Quarterly and yearly. The level at which what's invisible week to week appears — for example that you've been talking about a goal for eight months and it hasn't moved.

7. What we don't claim

A methodology write-up is worth as much as the honesty of its limits. Here are ours:

1. Not therapy and not a healthcare service. We don't diagnose and we don't replace a professional. In a crisis get help immediately: your local emergency number (112 in the EU, 911 in the US) or a local crisis helpline.
2. Not clinical evidence. The modules draw on established practices, but we don't claim they treat any condition.
3. The data is self-reported. What you write is your perspective — valuable, but not objective measurement. The AI works from that, so its insights are of the same nature.
4. It doesn't work for everyone. For someone who just wants a beautiful memory journal it's dead weight. For someone using it instead of therapy it points the wrong way.
5. It takes time. A few days of data say nothing. The system starts being useful after three or four weeks — better to know that up front than to be disappointed.

Frequently asked questions

What is the core of the methodology?

A closed loop: entry → pattern → insight → goal → daily habit → review. Most tools work at one point of it; Mirrify is a system because it carries the whole loop from the same data.

Where do the daily questions come from?

Guided question sets (morning and evening routines, schema journal, wheel of life, Ikigai) and the AI mentor working from your own earlier entries.

What practices do the modules draw on?

Expressive writing, morning pages, gratitude journaling, cognitive and schema-based reframing, implementation intentions, small steps, the wheel of life, Ikigai and the decision journal. These give the shape — not a guarantee of effectiveness.

How does the AI recognise patterns?

It doesn't read through your account: each analysis is a separate request carrying at most 30 entries, each up to ~800 characters, and for chats the last 20 turns at most.

Why isn't journaling enough on its own?

Because it's the first step of the loop. Without becoming a goal and a daily act, next month you'll write the same thing again.

Is Mirrify a therapeutic method?

No. It's a self-knowledge tool, not a healthcare service, and it doesn't diagnose. In a crisis get help immediately: 112 in the EU, 911 in the US, or a local crisis helpline.

How long before it shows anything?

The first week is settling in. Meaningful patterns appear after three or four weeks; quarterly and annual reviews show what's invisible week to week.