Schema & Insight Tools

How Data Explorer turns a NoSQL result set into multiple lenses so you can move from raw documents to schema patterns, quick metrics, and next-step MQL faster.

Raw JSON tree

Preserves the original document shape when you need to debug nested objects, arrays, or sparsely populated fields.

Auto-Flatten Grid

Turns common fields into columns so you can compare many documents at once instead of expanding each JSON node manually.

Schema Analysis

Focuses on collection structure itself: field coverage, type variance, and example values instead of just row-by-row reading.

Pick the right lens for the right question

"What does this document actually look like?"

Use Tree (JSON). This is the best mode for nested shapes, embedded objects, and fields that do not appear consistently.

"Which fields are the most common?"

Use Auto-Flatten Grid or Schema Analysis. The grid gives a faster scan; schema analysis gives a more structured explanation of field coverage.

"I want grouped metrics or patterns"

Use visualize/result insights for a quick read, or move into the Aggregation Builder if you need a deliberate, repeatable pipeline.

"I want AI to draft the next MQL step"

Use the NoSQL AI box after you have the correct collection selected. It is much stronger when it knows the active collection and what the previous result already suggests.

What Schema Analysis really tells you

Schema Analysis does not try to turn MongoDB into a rigid SQL-style schema. Instead, it inspects the current sample to answer practical questions: which fields are stable, which appear only in a small subset, where data types are mixed, and which values are worth noticing. The sample has explicit runtime limits, so the result is directional evidence rather than an absolute promise about the whole collection.

What it is good at

It is excellent for onboarding into a new collection, checking ingest quality, detecting type drift, or preparing a cleaner `$project` / `$group` pipeline.

What it should not be mistaken for

It is not a guarantee that every document in the collection is identical. It is an observational tool based on the current data sample and runtime limits, so use it for direction and then verify with your own queries when needed.

Visual insights: useful, but not a full BI layer

Fast insight from the current result set

The NoSQL visualize flow is designed for extracting grouped metrics, standout categories, or quick comparisons from the current result set. If the result set is capped, the insight reflects only the returned data. It helps you decide the next step, not replace a long-term dashboard.

AI is the next support layer

When schema analysis or visualize sparks the next question, the NoSQL AI box is where you turn that question into the next MQL draft instead of starting from scratch.

When the backend returns `truncated`, treat the insight as a quick read over a capped sample. If you need collection-wide numbers, build an intentional Aggregation Builder pipeline with `$match`, `$project`, `$group`, `$sort`, and `$limit`.

A very effective workflow is: open the collection -> run a small `find` -> inspect Tree/Grid to learn the shape -> use Schema Analysis to confirm field patterns -> move into the Aggregation Builder or AI for the next step.

PreviousAggregation Builder Next Monaco Editor

Ready to explore data?

Join the community on GitHub to contribute to the smartest local database IDE.

Launch App