vin
Vehicle identifier
Domains are labels. Mathematics is the code.
XENTHK reads any dataset by its shape, its measures and its relations. It chooses the analysis, tests the evidence and tells you what the data shows, and what it cannot.
Your rows stay on your computer. XENTHK computes there.
Check your data
558,837 used-vehicle auction sales, 16 fields
ID
state
US state, two-letter code
mmr
Market value, looked up: Manheim Market Report
sellingprice
Price in US dollars
Which state has the strongest demand for used cars?
The file holds completed sales, not demand, so this answer uses sales activity. Florida has the most: 82,945 sales, then California with 73,148.
From raw data to a decision
Five steps, each one visible. Wherever the data does not settle a question, you decide.
Raw data
year, make, trim, body, vin, state, odometer, mmr, sellingprice
2015, Kia, LX, SUV, 5xyktca69fg566472, ca, 16639, 20500, 21500
26 rows sit one column to the right: a comma inside a trim value put a vehicle number in the state field.
Understanding
vinstatemmrsellingpriceVehicle identifier, US state, market value looked up, price in US dollars. One row is one sale, confirmed by you before anything is computed.
Analysis
Asked for: demand.
Not in the data: no bids, inventory or time to sale.
Measure used: count of sales by state. Second check: total selling price by state.
Evidence
Sales by state, top five
26 shifted rows left out at your request.
Decision
Florida has the highest observed sales activity.
Allowed what the data records
Refused “strongest demand”, or any cause
Data and mathematics are like DNA and RNA: they always move together. A domain is a class people defined. The foundation of this system has to be built on data and mathematics.
1,728 KPIs. 23 mathematical forms.
The KPIs of dozens of industries look different until you write them down. Then 1,706 of them fall into 23 forms, and a single form, the rate, covers 384. XENTHK is built on the forms, not on the labels.
79 analytical methods compute them on your computer.
See the KPI libraryNot every question deserves the same analysis.
XENTHK picks the route from the question and the shape of your data, and tells you which route it took.
Which product line brings the most revenue?
Total by group, ranked
How did sales move month by month?
Total by month, in order
How spread out are order values?
Quartiles, density and the values beyond the fences
Do higher mileages go with lower prices?
Correlation: shows association, never a cause
Selling price against market value, by state?
Ratio of two totals per group
What is our gross margin by branch?
The KPI’s formula, bound to your fields
It adapts to your data. Your data does not adapt to it.
Unfamiliar names, mixed spellings and broken rows are found when the file is read, and you decide before anything is analysed.
An unfamiliar term
“mmr” is looked up, confirmed with you and read as the Manheim Market Report value: a market price, not a count.
Broken rows
26 of 558,837 rows carry a vehicle number in the state field. XENTHK shows them and asks whether to leave them out.
One value, many spellings
“Access Cab” and “access cab” are the same trim. XENTHK asks once whether to count them as one.
Your rows never leave your computer.
The data engine and the analyst model run in your browser. XENTHK delivers them encrypted after you sign in. Your files, your query results and your charts stay on your device.
On your computer
- Your files and the records read from your sources
- The data engine, which computes every table
- The analyst model, which reads fields and writes answers
- Every result, chart and export
At XENTHK
- Your account and sign-in
- Encrypted delivery of the model and the engine after you sign in
- An encrypted relay to the web addresses you connect; it carries bytes it cannot read
- The record that you accepted the usage notice
Your data already contains the answer. XENTHK finds the evidence.
Bring a file or connect a source. Ask in your own words.