Try engines with sample data
Sample data is a pre-loaded example result that lets you see what an engine produces without uploading a document. Five engines have one: Bid Leveling, Contract Review, Bid Scope Compliance, Change Order Review and Pay App Audit. A sample runs in one click from canned output, so it does not call the AI or spend a run.
Updated
Why use sample data?
New users often stall at the "what do I upload?" step. Sample data lets you skip that — five engines show a Try with sample data chip, and clicking it renders a fully populated result for a realistic scenario. You see the output, the callouts and the layout before committing to your own data.
Sample runs are rendered from canned output — they don't run a real analysis and don't consume your runs quota. They use the same result screens as a real analysis, filled with example data, and every sample result carries a Sample Data banner. Use them to orient yourself before spending a real run.
The sample scenario
Four of the five samples use one fictional project, Bridgeport Datacenter, and one fictional electrical subcontractor, Tancredi Electric Co., so you can follow the same vendor through a contract review, a scope check, a change order and a pay application. The Bid Leveling sample is a separate comparison of three HVAC vendor quotes.
The Platform Atlas tells a day-by-day story of the same fictional project. Its figures are its own and differ from the in-app samples.
| Engine | What the sample shows |
|---|---|
| Bid Leveling | Three HVAC vendor quotes (Arctic HVAC Systems, ComfortAir Solutions and ThermalPro Inc) across 12 line items, with quantity mismatches and missing items. |
| Contract Review | An AIA A401 electrical subcontract in Connecticut: risk score 58, seven clauses listed (three marked risk, including the $8,500-a-day liquidated damages rate and a broad pay-if-paid clause) and three missing standard clauses. |
| Bid Scope Compliance | Tancredi Electric Co.'s bid against the Bridgeport Datacenter Division 26 electrical spec: 78% coverage, two critical exclusions (the UPS, about $420K, and temporary power, about $340K) and four clarifications to request. |
| Change Order Review | CO #3 from Tancredi Electric Co., an $847K power-density upgrade: $548K recommended, $299K disputed, and three manipulation flags. |
| Pay App Audit | Tancredi Electric Co.'s Pay App #3 for March 2026, a $2.2M request with three red flags, including a retainage shortfall of about $182K and a $340K stored-materials claim with no backup. |
How do I try an engine with sample data?
To try an engine with sample data, open one of the five engines, click the Try with sample data chip, and look at the rendered output. A sample is never saved to your project, so there is nothing to clear when you are ready to go back to your own work.
Open one of the five engines
Go to Engines in the app and open Bid Leveling, Contract Review, Bid Scope Compliance, Change Order Review or Pay App Audit.
Click the 'Try with sample data' chip
A small chip near the top of the input panel offers the sample. It goes away once you run an analysis.
See the rendered output
The engine skips straight to results, populated with the example data and laid out the way a real result is: summary, tables and risk flags.
Return to your own work
Leave the results screen, or attach your own documents and run the engine. The sample is not saved to your project.
Every sample result shows a Sample Data banner: it is a demo result generated from a built-in sample document, not a real analysis of anything you uploaded, and the vendor names, prices and figures are fictional.
Engines without a sample
The other engines and tools have no sample. The tracker engines (COI Tracker, RFI Tracker, Owner Report, Field Daily Report, Punch List, Submittal Tracker, Procurement Tracker) produce tracker rows rather than standalone reports, so a sample would need seeded tracker state. RFQ Generator, Contract Summary and the remaining engines have none either; run them on your own documents.