The Systems Impact Framework

The foundational methodology underpinning all Jungle products. The Systems Impact Framework (SIF) provides a structured, evidence-based approach to identifying the market constraints that affect the performance and strategy of companies. It does so by distinguishing binding bottlenecks from background noise using formal constraint tests, causal role classification, and structured scoring.

Evidence to Outcomes.

LEGALTRADEOPERATIONALCLIMATE RISKDERIVESRESOLVESSEQUENCESEXECUTESCALIBRATESEvidenceConstraintsDecisionActionsOutcomes

Line weight follows the direction of flow. Outcomes is drawn open because that layer fills as engagements complete.

01EvidencePrimary legal, trade, operational and climate risk sources, held as a versioned corpus rather than scraped per question. Each item carries the date it was checked.
02ConstraintsThe requirement set that reaches a given subject, structured so that severity and the order in which conditions bind are properties of the model rather than an analyst's opinion.
03DecisionA verdict that can be put in front of a board or a counsel, carrying the constraints that produced it and the article each one rests on.
04ActionsThe steps that follow the decision, sequenced against what binds first, so the order of work is a finding rather than a plan written after the fact.
05OutcomesWhat actually happened during execution, returned by delivery partners and fed back into the evidence layer. Designed and filling as engagements complete.
01

The method is open, so the answer is checkable.

The Systems Impact Framework is published rather than described. Anyone can read how objectives, constraints and interventions are modelled and then test whether an assessment we produced followed it.

Read the framework on SSRN →

02

Ask the same question twice and you get the same answer.

Assessments are generated from a held requirement set and a versioned corpus, not written from scratch each time. Re-run one and it reproduces line for line, which is what makes a finding auditable rather than merely persuasive.

03

We state what we did not check.

Every report carries a coverage statement: what was examined, what was not, and where the corpus ends. A bounded answer you can rely on is worth more than an unbounded one you cannot.

04

One engine, three receptors.

The same architecture runs against a company, a specific asset, or a portfolio of them. Applications differ in what they are pointed at, not in how they reason.

Market Entry →

Tested against regulation that had not happened yet.

The clearest test of a constraint model is whether it anticipates the constraint. The Chemical Screener applies the same structural reasoning to the REACH substance universe, detecting embeddedness in ECHA's hazard network to identify which substances regulators reach next, and it was validated against six independent historical tests rather than a single favourable window.

Read the methodology, feel free to reach out to collaborate.

Get in touch →