Three surfaces onto one graph, each shaped for the desk that uses it. The same evidence base, the same versioned methodology, the same release cadence. The difference is the verb — search, forecast, query.
The analyst's desk. Search by entity, by HS code, by lane, by counterparty bank; pivot from a single hit to the full trade graph around it; export findings with the methodology footnotes intact.
For · compliance · trade finance · sourcing
Read on CrestLensLive lane-level disruption forecasts. Eight inputs, one decision-ready number per lane per week, with the contributing factors named explicitly. Calibration is published; the methodology is versioned and signed.
For · carriers · forwarders · insurers
Read on Signal RoutesDirect programmatic access to the graph. Four endpoints, stable contract, predictable schema evolution, OAuth 2.0 with hardware tokens for any write path. The graph at the speed of your own data layer.
For · in-house data teams · platforms
Read on CrestAPIThe three surfaces are configured into five workspace shapes — compliance, trade finance, sourcing, carrier and platform — because the verbs each desk performs against the graph are different, and the workspace ergonomics that best support those verbs are different too. The underlying graph, the evidence, the methodology and the release cadence are the same. The columns, the default watchlists, the alert rules and the report templates are not.
Every contracted client receives the workspace shape that best fits their desk; we do not charge extra for additional shapes if the engagement spans more than one function. A trade-finance team that also operates an in-house data layer receives both the trade-finance workspace and CrestAPI seats against the same engagement.
| Workspace | Primary verb | Default surface | Audit hook |
|---|---|---|---|
| Compliance | Resolve · screen · evidence | CrestLens | Per-query · per-user |
| Trade finance | Underwrite · monitor · escalate | CrestLens + Signal Routes | Per-deal · per-bank |
| Sourcing | Find · qualify · second-source | CrestLens | Per-supplier · per-PO |
| Carrier | Forecast · re-plan · price | Signal Routes | Per-lane · per-week |
| Platform | Ingest · join · expose | CrestAPI | Per-endpoint · per-token |
The methodological commitment behind the three surfaces is that no two of them disagree on a fact. If CrestLens shows that an entity sits one hop from a designated party on the bank graph, Signal Routes scores the lane that entity ships through with that hop in its inputs, and CrestAPI returns the same hop at the corresponding endpoint. The evidence is one. The methodology is one. Where any of the three appears to disagree with another, the disagreement is a defect, and it is logged, published in the Library, and resolved in the next release.
That commitment is the reason we did not build the surfaces on separate datasets, and the reason we do not licence those surfaces separately as if they were separate products. They are not separate. They are three views of one model, and we sell the model.
The most efficient way to find out which surface, or combination of surfaces, fits your desk is the briefing. Forty-five minutes against your live portfolio, no slideware, no qualification call, no obligation to continue. By the end of the session you will have a written report that maps your work to the three surfaces and tells you, plainly, whether and how the graph helps. If it does not, we will say so.