EVIDENCE FIRSTNO PROMISESDATA YOU CAN AUDIT

Before you start,see where you stand

Before you write the check, you need to know whether to spend the money, where, and what to ask.

A free positioning assessment powered by the Kaiso matching engine · 8 recent admission cycles + current-season policy. Every conclusion is auditable. We don't decide for you.

All free tools work without an email signupEvery claim links back to its data sourceThe applicant + parents stay in the driver's seat
EARLY DEADLINES
MIT · EA 11/01Caltech · REA 11/01Harvard University · REA 11/01Yale University · SCEA 11/01Princeton University · SCEA 11/01Stanford University · REA 11/01Ohio State University · EA 11/01University of Pennsylvania · ED1 11/01Columbia University · ED1 11/01Brown University · ED1 11/01Cornell University · ED1 11/01Dartmouth College · ED1 11/01University of Chicago · ED1 11/01Duke University · ED1 11/01Northwestern University · ED1 11/01Georgia Tech · EA1 10/15University of Michigan · EA 11/01UIUC · EA 11/01
SAMPLE REPORTS

What the report looks like — see for yourself

Two samples, open to read: generated from test data with pseudonymous applicants, through the school-selection chapter — enough to see how each conclusion is reached.

44
colleges on file
13
profile dimensions
8
admission cycles
Browse sample reports
PICK BY GRADE

What grade is the applicant in?

One recommended path per grade — no bundle-stuffing. Pick your grade to see the path and price.

Not sure? Start with the free 10-min assessment — see where the applicant sits across 44 schools, then decide whether to go deeper.

vs. generic LLMs · COMPARISON

Same question,different answers

Generic LLMs run on training-time data — they're prone to hallucinating numbers. Kaiso runs on 8 recent admission cycles + current-season policy data, with every judgment leaving an audit trail.

8 RECENT CYCLES · CURRENT-SEASON POLICY · TRACEABLE

Generic LLMsGPT / CLAUDE / GEMINI
VS
Kaiso enginePEIPAOLAB
Made-up numbers
"You have a 32–45% chance" — percentages out of thin air
Admit probability
Three-tier classification
Reach / match / safety — no fabricated admit rates
Training snapshot
Data from a year or two ago — policies have already shifted
Data freshness
Quarterly refresh
Public admit data + admissions policy refreshed each quarter
Generic POV
Defaults to US-domestic narrative — no specifics for Chinese-American families
Chinese-American context
First-hand Chinese context
7 Chinese-American hubs + ethnicity breakdown + Chinese-language scenarios
Black-box generation
Why this judgment? The model can't explain itself.
Traceable decisions
Auditable basis
13-axis scoring + comparison against similar admit cases
Data sources Public admit data · IPEDS · curated real admit cases
Evidence strength
EVIDENCE FIRST
Data driven · auditable decisions

Let the data
speak for theadmissions office

Start free assessmentBrowse all products →No signup · 2 minutes · always free