Evidence · first-party tested/Query Live Databases Using Plain English with AI

Showed the SQL used for each step instead of only presenting the final answer.

✓ Worked🧾 artifact-verifiedinput + output shownTest date not recordedAskYourDatabase →

What was measured

SQL Visibility

Can users inspect or copy the generated SQL?

capability

What was given, what came back

Test input: Order pipeline breakdown with paid-pending edge case and last-month comparison · text · group: ecommerce-nl2sql-benchmark

Input — what we sent

The exact prompt

How many orders do we have at each stage right now?

Follow-up 1: What percentage of our orders were successfully delivered vs cancelled?

Follow-up 2: Are there any orders that are pending but already paid?

Follow-up 3: Compare that to last month — same breakdown, I want to see if things have improved or got worse.

A deeper operational analysis of current order stages, delivery-versus-cancellation rates, pending-but-paid edge cases, and a month-over-month comparison of the same breakdown.

Why this input is hard

· Order pipeline analysis
· Percentage calculation
· Edge-case detection
· Payment/order status joins
· Multi-turn context retention
· Month-over-month comparison
· Ambiguity handling for 'same breakdown'

Output — unretouched

Also checked on this input — same tool, 12 other criteria

Ambiguity Handling✓ WorkedClarified 'last month' upfront as April 20, 2026 before running the comparison, rather than guessing what period the user meant.Business Insight✓ WorkedThe follow-up summaries explained what the raw counts meant operationally, including delivery/cancellation rates and the backlog reduction story.Business Insight✓ WorkedThe tool added operational context by flagging the two oldest pending-but-paid orders as needing manual intervention.Business Insight✓ WorkedTurned the comparison into a business judgment by stating that the backlog fell from 13 stuck orders to 2, an 85% reduction that 'improved significantly'.Chart / Visualization Support✗ FailedVisualization was not automatic; the report says the tool only produced a visual after an explicit follow-up prompt, so charts were not auto-generated on the initial answers.Chart / Visualization Support✗ FailedVisualization did not auto-generate for the order-pipeline flow either; the report says the user had to ask for it separately.Follow-Up Context◐ MixedThe tool retained the immediately preceding pending-paid edge-case context, but the report says it used that narrower metric for the follow-up comparison instead of the full original stage breakdown.Plain English Query Handling✓ WorkedThe tool correctly handled a multi-step operational conversation without requiring structured SQL from the user at each turn.Plain English Query Handling✓ WorkedAccepted a four-step operational analysis in plain English across the full follow-up chain without needing the user to write SQL.Result Readability✓ WorkedThe pipeline output was readable because it summarized counts, percentages, and the stuck-versus-resolved comparison in plain tables and short callouts.Result Readability✓ WorkedReturned a clean 7-row status table with 93 total orders and an easy-to-scan delivered-versus-cancelled percentage summary.SQL Generation✓ WorkedHandled the month-over-month follow-up with multiple SQL statements to compare the current snapshot, inspect paid-but-pending orders, and build the prior-month comparison.

Provenance

Observation: ea4e0c90-e39a-42db-9b13-d3561f30c240
Evidence run: db2bb5d5-0e0e-4cb3-8d76-3555c45c23cd
Study: Query Live Databases Using Plain English with AI
Research task: 86b9y6c99
Tested at: not recorded
Source: first-party
Evidence state: verified
Proof shown: input + output shown
Cost / latency: not captured
Repeat run: not captured
Tester: not captured

The last three rows are honest blanks, not placeholders — our capture has no field for them yet.

Query this

get_evidence({
  tool: "askyourdatabase",
  scenario: "ecommerce-nl2sql-benchmark"
})

MCP · mcp.aidemos.com/api/mcp

JSON · ai.aidemos.com/v1/observations/ea4e0c90-e39a-42db-9b13-d3561f30c240

Free with attribution.

Same input, same check — 1 other tool

measured on SQL Visibility

Querio✓ WorkedExposes the generated SQL directly in the chat, including the explicit filter for pending-but-paid orders.

This evidence is published in

Query Live Databases Using Plain English with AI →

From the same study (page rebuilt from a later run)

Best AI Tools to Query Live Databases Using Plain English →

Real inputs and real outputs, no retouching · every cell queryable via API & MCP · aidemos.com