twentii · a thesis · 2026
twentii turns decades of industry data into the decisions businesses make next.
01 · the problem
The dormant decades.
The most valuable dataset in most established industries is not on the internet. It sits in the businesses themselves — ledgers, ERPs, archives, filing cabinets — and in public records nobody has structured.
A company that has traded for 40 years holds 40 years of prices, orders, failures, and recoveries. Yet forecasting, pricing, and capital decisions are still made on instinct, because the history was never in a form a decision could use.
The data exists. It has never been asked.
- 1962
- 1966
- 1970
- 1974
- 1978
- 1982
- 1986
- 1990
- 1994
- 1998
- 2002
- 2006
- 2010
- 2014
- 2018
- 2022
- 2026
64 years in one example business. Read once, if ever.
02 · the method
Access. Structure. Intelligence.
Institutional memory and machine intelligence are worthless apart. Paired, they compound. The method is the pairing.
-
01 · access
We work inside the business.
History arrives in whatever form it was kept — paper ledgers, microfilm, ERP exports, public registries. We take it in that form. Nothing is required of you but the key to the cabinet.
- intake
- ledger scans · DMS exports · registration archives · auction sheets
- premise
- your records stay yours, on your terms
-
02 · structure
Records become one schema.
Entities matched, units reconciled, gaps declared. Every figure stays traceable to its source page — a table you can interrogate, and audit back to the paper.
- output
- entity-resolved tables, one schema per vertical
- provenance
- page-level; gaps declared, never silently filled
-
03 · intelligence
Structured history feeds the decision.
Forecasting and pricing models trained on decades of your market, not last month's. The output is not a dashboard. It is the next decision, with its reasoning attached.
- models
- demand forecast · price recommendation
- horizon
- quarters and cycles, not clicks
03 · first vertical
Auto, first.
Our first product is Revenue Engine — a profit-recovery system for car dealerships. It reads one month of your dealership's own records, finds where revenue quietly leaked, validates what is actually recoverable against your stock, leads and capacity, and turns that into an owned, dated action plan your managers run.
Last month's leakage is a diagnosis. This month's recoverable is a commitment — and at month end the engine reconciles the plan against realized recovery, line by line.
One capability, done properly, before any other claim.
- input
- one 5-sheet workbook from your DMS · five minutes, once a month
- output
- observed leakage → validated recoverable → owned action plan
- rule
- recoverable never exceeds observed — by construction
- briefing
- auto@twentii.ai — your numbers in 48 hours
six levers shown · illustrative — the engine surfaces what your records hold, and it is rarely just these
on the horizon — steel · real estate · trade · logistics · retail
04 · proof of craft
The assembly.
A month of loose dealership records. One schema. Two figures — the leak, and the recovery. You conducted it: the marks that scattered are the marks that decided.
05 · contact
Partner with twentii.
If your business holds decades of records — and decisions that deserve them — we want the conversation.
Write directly: auto@twentii.ai. We reply within two business days.