How state3 stays true.
Most CMDBs you've used drifted out of date within a year. state3 doesn't, because we built the upkeep in. Automated where it's safe. Human where it matters. The two working together is what makes the answer trustworthy — and what makes the graph worth asking.
Three stages. One source of truth.
Every piece of data that enters state3 passes through the same pipeline — pulled in, interpreted, and confirmed before it earns a place in the graph.
Get it in.
Connectors run on a schedule. CSV upload covers the rest. Both paths land in the same staging layer — the pipeline doesn't care which one fed it.
Make sense of it.
The rules engine reads what arrived and proposes Add, Update, Delete or Merge actions — applied automatically if confidence is high, queued for review if it isn't.
Trust it.
Uncertain actions queue for review with field-level diffs. You approve or reject. Once a person has edited a field, automation runs around it — not over it.
Why most tools fall behind.
Auto-import everything from every system you've got, and the data goes stale — because nobody trusts it enough to fix it. Require manual entry for everything, and the platform never gets adopted at all. Most CMDB and EAM tools made one of those two choices and lived with the consequences. state3 didn't pick. We built a pipeline that knows the difference, and a UI that lets a human resolve the cases the machine isn't sure about.
How it actually works.
Three subsystems. Each one earns the trust the next one depends on. Read at whatever depth you need.
Get it in.
Two paths bring data into state3. Connectors — Azure AD, DevOps, Digicert, and others — run on a schedule and pull directly from the source system. The CSV import wizard covers everything without a connector: map CSV columns to state3 fields, mark anything you don't need as Skip, and upload. Both paths land in the same staging layer. The rest of the pipeline doesn't know or care which one fed it.
Every import is tenant-scoped, sanitised against CSV-injection prefixes, and capped at 100,000 rows and 200 columns, written in batches.
Column mapping — CSV import wizard
| CSV header | Maps to state3 field |
|---|---|
| Asset Tag | asset_tag |
| Make | manufacturer |
| Model | model_name |
| Serial | serial_number |
| Location | Skip |
Five columns from a source CSV mapped to their state3 equivalents, with a Skip option for anything the organisation doesn't track.
Make sense of it.
Once data lands in staging, the rules engine reads it and produces a list of proposed actions. Each action is one of four types — Add a new entity, Update an existing one, Delete something the source no longer reports, or Merge two records the engine recognises as the same thing. Each proposed action is tagged either Auto (high confidence — apply it) or Manual (queue it for human review). Confidence is shaped by alias lists, ignore lists, and transformation lists maintained per tenant.
Decision card — rules engine output
Update
Asset 1234 — location:
Hamilton → Tauranga
(alias match)
Merge
‘SQL Server’
‘Microsoft SQL Server’
(queued for review)
Two proposed actions: one auto-applied on an alias match, one queued for a human to resolve.
Trust it.
Once actions are proposed, three mechanisms make sure what ends up in state3 is what your team actually believes is true.
Review queue
Every uncertain action shows you the proposed change with a field-level diff — approve or reject, with a comment if you need one.
Diff card — action review
| Current | Proposed | |
|---|---|---|
| Edition | Standard | Enterprise |
| Version | 2016 | 2019 |
Confirmation
When you've looked at a record and you know it's correct, you confirm it. state3 records who confirmed it and when. You can confirm a single record or bulk-confirm an entire model in one operation.
Confirmation chip
Confirmed by basil.buwalda@state3.co.nz · 2026-04-29Manual edit protection
Once a person has edited a field, the rules engine never overwrites it. Your edits are sacred. Automation runs around them, not over them.
What this earns you.
This is what the upkeep is for. The five lenses on the home page rely on it. So do the MCP queries an AI assistant runs on your behalf. So does any number a CIO defends in a board meeting. A graph that lies to you isn't intelligence — it's just paperwork that breaks faster.