What the research actually says.

Eight mechanisms, two phases

The product is built on eight mechanisms drawn from established therapeutic traditions. Each is mapped to the phase of the product that engages it.

See How It Works for how each phase is implemented.

What the evidence supports

Across 400+ peer-reviewed studies, multiple meta-analyses, and clinical applications across diverse modalities, writing interventions reliably produce measurable benefits — visible both in how people report feeling and in objective health measures.

Outcome evidence

What makes it work — and how the product is built around it

The same literature is clear about the conditions under which these benefits appear. Each one shaped the product directly:

• Emotional engagement, not facts alone (Pennebaker 1986). The benefit is strongest when writing connects to feeling — which is why the assessment asks how much each experience affected you, not merely whether it occurred.
• Structure and pacing (SAMHSA TIP 57). Structured, paced, user-controlled formats are safer and more effective than unstructured writing — which is why the assessment is paginated, auto-saved, and entirely self-paced.
• Privacy raises disclosure. Self-report questionnaires elicit greater disclosure than face-to-face interviews — which is why the whole process is private by architecture.
• Safety supports are a precondition. Engaging traumatic material without them can do harm — which is why screening runs before the assessment and crisis resources stay within reach throughout.

The session-equivalence figure, with all three qualifiers

You will see the figure 3.2 to 4.5 therapy sessions per letter in writing about narrative therapy. We use this benchmark in our framing, and we surface every qualifier:

1. It comes from clinicians asking the people best placed to judge — their own clients. Epston asked clients in his practice "how many sessions do you consider a letter worth?" The answers averaged 4.5. Nylund & Thomas (1994) asked the same question in a Kaiser Permanente clinic and found 3.2. These are practitioner findings rather than controlled experiments — and they measure exactly the thing that matters: what a letter was worth to the person who received it.
2. It describes therapist-written letters within ongoing therapeutic relationships. The letters in those samples were written by therapists who already knew their clients, summarising sessions, externalising problems, highlighting unique outcomes.
3. Our pipeline has not yet been measured against that benchmark. No study yet compares a letter assembled from a structured self-assessment with a therapist-written or self-written one. The mechanisms it draws on are well-established; the specific pipeline is new.

We use this figure as an anchor for thinking about value, never as a direct equivalence claim.

Sources: Freeman, Epston & Lobovits (1997), Playful Approaches to Serious Problems, Norton; Nylund & Thomas (1994), "The economics of narrative," Family Therapy Networker, 18(6), 38–39.

The 200 questions: where they come from

The questionnaire is original work, but every question is grounded in established literature on childhood abuse, neglect, and dysfunctional family dynamics.

Two integrated frameworks:

• Abuse & Neglect — 69 questions spanning emotional abuse, physical abuse, sexual abuse, emotional neglect, and physical neglect. Source domains include the CTQ-SF construct map (Pearson; we did not adopt it), the ACE study domains (Felitti & Anda 1998; public domain), and the broader trauma literature.
• Dysfunctional Dynamics — 131 questions covering parentification, enmeshment, invalidation, role reversal, golden child / scapegoat patterns, chronic emotional unavailability, conditional regard, and other relational dysfunction patterns.

Why we wrote our own. Two reasons, and they define the product.

First, every question is phrased as an experience, not an event. The established instruments in this space — the ACE study's ten yes/no items, the CTQ-SF's frequency ratings — record events as external facts: did this occur, how often. That is the right design for screening, but it leaves the person out. Our questionnaire takes the subjects the literature knows well — abuse, neglect, dysfunctional family dynamics — and rephrases every one of them as lived experience: what it was like from the inside.

Second, the questionnaire measures impact, not occurrence. You do not just confirm that something happened — you rate how much it affected you. Whether an event took place and what it did to a person's psyche are different questions: something can have happened and cost you little, and something that looks small on paper can have shaped a whole life. It is the second question a letter needs answered.

Underneath both sits the reason the framework had to be built from the ground up: no existing instrument takes in the whole of a life. Each established questionnaire examines one slice from one narrow angle. There was no single, continuous questionnaire covering every aspect of a person's life and how adverse experiences touched it — and that is also what letter generation needs: not categories and severity scores, but specific, attributed, personally weighted experiences.

Each question carries:

• A first-person description of the experience.
• An attribution of source framework (which tradition the construct came from).
• A row hash for change detection across questionnaire versions.

See How It Works for what you do with each question during the assessment.

The screening instruments

Two screeners run before the main assessment, and both were designed from the ground up — for one reason. The standard instruments in this space assume a clinician in the room or a clinic behind the screen. Nothing we found was built for the situation this product actually operates in: a person alone with a browser, about to engage difficult personal material, with no professional on hand. So the screening layer was built for exactly that — practical, structured safety and wellbeing checks of a kind online tools have largely gone without.

Both are described in detail under How It Works → Before the assessment. Their academic basis is here.

SRS-8 (Safety Risk Screener)

A structured screener for suicidal ideation and self-harm risk, designed for unsupervised digital self-administration. Adaptive branching keeps the experience brief for low-risk respondents (2 questions) and progressively deepens inquiry only where disclosure warrants it. Embedded micro-interventions (protective-factor articulation, help-seeking readiness) function as brief therapeutic contact points within the screening itself.

The C-SSRS (Columbia) is the institutional gold standard for clinician-administered suicide screening. It is purpose-built for trained clinicians in care settings. The SRS-8 is built for the gap C-SSRS does not fill: scenarios where no clinician is present and no clinical infrastructure exists.

PWA-9 (Personal Wellbeing Assessment)

A nine-question self-report measure of current wellbeing across seven dimensions (emotional stability, cognitive clarity, energy, social connection, agency, sleep quality, engagement) plus two standalone signals: perceived trajectory (PGIC-style) and distress tolerance.

The WHO-5 (CC BY-NC-SA 3.0 IGO) is the unidimensional benchmark for brief wellbeing screening. PWA-9 is dimensional rather than composite — a person scoring low can immediately see which dimensions are compromised, without requiring a second instrument.

Status

Both are original instruments, constructed on established psychometric principles and face-valid by design, built for the specific context they run in. Peer-reviewed validation has not yet been performed — it is on the roadmap, and we say so plainly. What they deliver today is something this space has mostly lacked: a practical safety and wellbeing layer built into the product rather than bolted on. They are triage gates by design — they decide what the product does next — and they hand off to professional care rather than stand in for it.

Where the evidence stops

Honest framing means marking the boundary as clearly as the support. Four claims you will not find us making:

1. "This product heals childhood trauma." No single intervention heals trauma — not therapy alone, not writing alone. What writing interventions reliably deliver is meaningful, documented improvement; that is what we claim, and nothing more.
2. "AI-generated letters are as effective as therapist-written or self-written letters." That head-to-head comparison has simply not been studied yet. The mechanisms the letter draws on are well-established; the direct evidence is still to come.
3. "This replaces therapy." Writing interventions work alongside professional therapeutic work, not instead of it. The product's standalone value — structured self-discovery, a tangible document, months of articulation compressed — is real, and it is its own thing.
4. "Your letter is worth X therapy sessions." The 3.2–4.5 session figure was measured on therapist-written letters within ongoing therapy. We use it as an anchor for thinking about value, not as a claim about our output.

Full citations

Active processing — Pennebaker paradigm

• Pennebaker, J. W., & Beall, S. K. (1986). Confronting a traumatic event: Toward an understanding of inhibition and disease. Journal of Abnormal Psychology, 95(3), 274–281.
• Pennebaker, J. W., Kiecolt-Glaser, J. K., & Glaser, R. (1988). Disclosure of traumas and immune function: Health implications for psychotherapy. Journal of Consulting and Clinical Psychology, 56(2), 239–245.
• Pennebaker, J. W. (2018). Expressive writing in psychological science. Perspectives on Psychological Science, 13(2), 226–229.
• Pennebaker, J. W. (2016). Opening up by writing it down: How expressive writing improves health and eases emotional pain (3rd ed.). Guilford Press.

Meta-analyses

• Smyth, J. M. (1998). Written emotional expression: Effect sizes, outcome types, and moderating variables. Journal of Consulting and Clinical Psychology, 66(1), 174–184.
• Frisina, P. G., Borod, J. C., & Lepore, S. J. (2004). A meta-analysis of the effects of written emotional disclosure on the health outcomes of clinical populations. Journal of Nervous & Mental Disease, 192(9), 629–634.
• Frattaroli, J. (2006). Experimental disclosure and its moderators: A meta-analysis. Psychological Bulletin, 132(6), 823–865.
• Pavlacic, J. M., et al. (2019). A meta-analysis of expressive writing on posttraumatic stress, posttraumatic growth, and quality of life. Review of General Psychology, 23(2), 230–250.
• Sohal, M., Singh, P., Dhillon, B. S., & Gill, H. S. (2022). Efficacy of journaling in the management of mental illness: A systematic review and meta-analysis. Family Medicine and Community Health, 10(1), e001154.
• Kirca, A., Malouff, J. M., & Meynadier, J. (2023). The effect of expressing gratitude on subjective well-being: A meta-analysis. International Journal of Applied Positive Psychology, 8, 347–373.

Mirror function — Narrative therapy letters

• White, M., & Epston, D. (1990). Narrative means to therapeutic ends. Norton.
• Freeman, J., Epston, D., & Lobovits, D. (1997). Playful approaches to serious problems: Narrative therapy with children and their families. Norton.
• Nylund, D., & Thomas, J. (1994). The economics of narrative. Family Therapy Networker, 18(6), 38–39.
• Bjoroy, A., Madigan, S., & Nylund, D. (2015). The practice of therapeutic letter writing in narrative therapy. In The handbook of counselling psychology (4th ed., pp. 332–348). Sage.

Structured exposure — CPT and trauma writing

• Resick, P. A., & Schnicke, M. K. (1992). Cognitive processing therapy for sexual assault victims. Journal of Consulting and Clinical Psychology, 60(5), 748–756.
• Resick, P. A., et al. (2012). A randomized clinical trial of CPT-C, CPT, and written accounts. Journal of Consulting and Clinical Psychology, 80(2), 201–210.
• Substance Abuse and Mental Health Services Administration (SAMHSA). TIP 57: Trauma-Informed Care in Behavioral Health Services.

Recognition & validation — ACE and disclosure

• Felitti, V. J., Anda, R. F., et al. (1998). Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The Adverse Childhood Experiences (ACE) study. American Journal of Preventive Medicine, 14(4), 245–258.

Closure — Confrontation letters

• Forward, S. (1989). Toxic parents: Overcoming their hurtful legacy and reclaiming your life. Bantam.

Self-compassion

• Neff, K. D., & Germer, C. K. (2013). A pilot study and randomized controlled trial of the mindful self-compassion program. Journal of Clinical Psychology, 69(1), 28–44.
• Neff, K. D. (2023). Self-compassion: Theory, method, research, and intervention. Annual Review of Psychology, 74, 193–218.

Narrative reauthoring (additional)

• "Dear John: Letter writing as narrative therapy intervention" (2024). Trends in Psychology, Springer.

Related practitioner frameworks

• Walker, P. (2013). Complex PTSD: From surviving to thriving. Azure Coyote.
• Bradshaw, J. (1990). Homecoming: Reclaiming and championing your inner child. Bantam.
• Young, J. E., Klosko, J. S., & Weishaar, M. E. (2003). Schema therapy: A practitioner's guide. Guilford.
• Prasko, J., et al. (2012). Therapeutic letters and emotional schemas. Activitas Nervosa Superior Rediviva.
• Paivio, S. C., & Greenberg, L. S. (1995). Resolving "unfinished business": Efficacy of experiential therapy using empty-chair dialogue. Journal of Consulting and Clinical Psychology, 63(3), 419–425.
• Wade, N. G., et al. (2014). Efficacy of psychotherapeutic interventions to promote forgiveness: A meta-analysis. Journal of Consulting and Clinical Psychology, 82(1), 154–170.
• Seligman, M. E. P., et al. (2005). Positive psychology progress: Empirical validation of interventions. American Psychologist, 60(5), 410–421.