You didn't go to grad school for this
You spend 8–15 hours writing every comprehensive evaluation report. Most of it is the same paragraph rewritten with different scaled scores.
Your testing is sharp. Your conceptualization is sharp. But the writing is eating your evenings, your weekends, and your fee-per-hour math.
DraftPsych takes your scores, observations, and recommendations and produces a complete first-draft report in your own clinical voice — so you can review and finalize in 30 minutes instead of writing from a blank page for half a day.
Built for the report types that actually pay your bills
We didn't build a generic “AI for psychologists.” We built templates for the six evaluations that make up 90% of private-practice testing work:
Coming next: pediatric neuropsych, adult ADHD, and forensic/IME templates.
Why this is different from pasting into ChatGPT
It sounds like you
Upload two or three of your past reports. DraftPsych learns your phrasing, your sentence rhythm, your favorite recommendations, your transition style. Every draft after that reads like you wrote it — because in every way that matters, you did.
It's grounded in your actual data
You enter scores in structured forms. The model interprets your numbers — it doesn't invent results, it doesn't hallucinate percentile ranks, and every claim is traceable back to the data you entered. No fabrication. No "the WISC-V suggests..." with no WISC-V in evidence.
You stay the author, always
Every draft is marked as a draft. You review, you edit, you sign. The recommendations bank is curated by licensed psychologists, but you choose which ones go in. Nothing gets sent or finalized without you.
Your data is yours
We've signed a Business Associate Agreement with our LLM provider. Your client data is encrypted at rest and in transit, never used for model training, and deletable on demand. We are not ChatGPT with a logo.
How a typical case flows
Open a new case
30 secondsPick a template. Enter the client's age, grade, referral question.
Drop in scores
5–10 minutesStructured forms for every common battery — WISC-V, WIAT-4, BASC-3, Conners-4, BRIEF-2, ADOS-2, Vineland-3, and more. Confidence intervals and percentiles auto-calculated.
Add your clinical observations
5 minutesFree-text box. Bullet points are fine — DraftPsych weaves them into prose.
Pick recommendations
3 minutesChoose 8–12 from a curated bank organized by domain. Or write your own.
Get your draft
90 secondsA complete, sectioned report in your voice. Reason for Referral → Background → Behavioral Observations → Cognitive → Academic → Attention/EF → Social-Emotional → Diagnostic Impressions → Recommendations.
Edit and finalize
~30 minutesRead, revise, export to .docx on your letterhead. Done.
Total time: ~75 minutes from intake folder to finished report.
Compare that to your current 8–12 hours.
What it costs
Built with practicing psychologists, not for them
DraftPsych is being designed in collaboration with a small advisory group of licensed clinical and school psychologists actively running testing practices. Every template, every prompt, and every recommendation in the bank has been reviewed by someone who writes these reports for a living.
We are not a generic AI company that woke up one morning and decided psychology looked profitable. We are building one tool, deeply, for one type of clinician.
Frequently asked, honestly answered
Who this is for
You're a fit for DraftPsych if you:
- ✓Are a licensed clinical, counseling, or school psychologist (or a postdoc nearing licensure)
- ✓Conduct standardized testing as a meaningful part of your practice
- ✓Write your own reports (i.e., you're not in a hospital with a template-generation system already)
- ✓Charge cash or a mix of cash and insurance for evaluations
- ✓Have ever finished a report at 11 p.m. on a Sunday and thought there has to be a better way
You're probably not a fit (yet) if you:
- ✕Work exclusively in a school district that uses Frontline, SEIS, or a district-mandated platform
- ✕Do only therapy, with no testing
- ✕Need integration with a specific EHR today (we'll get there in v2)
A small founder's note
I'm the founder. I'm not a psychologist — I'm a builder. I'm building DraftPsych in close collaboration with licensed clinicians because I watched someone close to me do testing work and realized the writing burden is the single biggest unfair tax on this profession.
If you're a testing psychologist with strong opinions about reports, please tell me what's wrong with this. I'd rather hear it from one of you in 20 minutes than learn it the hard way over six months.
— Manuja Verma
Join the Beta
Six months free. First 50 clinicians. We're looking for honest, opinionated testers who will tell us what's broken.
See a Sample Draft
Want to see what DraftPsych output looks like before you sign up? Drop your email and we'll send you a fully de-identified sample ADHD evaluation draft (PDF), generated from real-world score patterns.