Build · Assess & Decide · Analysis & Finance
AIaaS productization decision register
A founder deciding which of their internal AI models to package and sell as AIaaS gets a defensible, rule-derived register that says Go / Go-with-notes / Stop per model and names the single best candidate to commercialize.
You receive: A JSON object {ideaName, models:[{name, input, output, dataVariation:int 1-3, logicVariation:int 1-3, interCustomerVariability:enum[low,medium,high], processMaturityBand:enum[low,medium,high], systemMaturityBand:enum[low,medium,high], technicalFeasibility:enum[notFeasible,lowFeasibility,mediumFeasibility,highFeasibility], businessPotentialScore:int 1-4, competitivenessScore:int 1-4, productizeVerdict:enum[go,goWithNotes,stop], rationale:string}], recommendedModel:string}
Part of Choose Business Model
What you receive
The shape of your delivered file. Your result is generated for your inputs.
Your deliverable is a structured document carrying these fields:
- Business Potential Score
- Competitiveness Score
- Data Variation
- Inter Customer Variability
- Logic Variation
- Models
- Name
- Process Maturity Band
- Productize Verdict
- Rationale
- Recommended Model
- System Maturity Band
- Technical Feasibility
Some fields carry set values, drawn from:
go, goWithNotes, stop
low, medium
low, medium, high
medium, high
mediumFeasibility, highFeasibility
notFeasible, lowFeasibility
notFeasible, lowFeasibility, mediumFeasibility, highFeasibility
Opens soon
STUD opens runs for everyone soon, on every play at once.
Scope of the check
What's verified: STUD verifies that the register is structurally complete, every score is in range, and every derived verdict (variability, feasibility, productize) correctly follows the book's lookup tables from the buyer-supplied inputs. STUD does NOT verify that the maturity/variability/business-potential inputs are factually accurate for the real models, nor that the recommended model will actually succeed commercially.
Get early access to STUD the day it goes live.