Time-to-Consequence: The Metric Your Portfolio Tools Don't Have

The question that matters most about a position is not "what is the metric this quarter." It is "how long until that metric matters in a way we can no longer reverse." That is time-to-consequence.

What the metric measures

Time-to-consequence is the projected runway between the current state of a position and the moment a structural break becomes inevitable. For a credit position, the structural break is usually a covenant breach or a payment default. For a PE position, it is the moment the IC thesis is no longer defensible — when the conditions the team relied on are gone and the case for holding the asset has dissolved.

The metric is computed by taking the live operator data, propagating it forward under a base case (and a small set of stress cases), and asking: how many quarters until we hit the structural break? It is a number, in time units, on every position. The team can sort the entire book by it.

Why every senior PM measures it informally

Talk to a senior credit PM and ask which positions are keeping them up at night. The answer is never "the ones with the worst KPI this quarter." The answer is "the ones where the runway to a covenant break is shortest." They are computing time-to-consequence in their head, every day, on every position. They are also doing it from memory, on incomplete data, without a consistent definition.

That is the gap. Senior people do this calculation reflexively. The monitoring tools do not surface it. So the institutional knowledge about which positions are running out of room sits in the head of one PM — and when that PM leaves, the institutional knowledge leaves with them.

With $3.6 trillion of unrealised value sitting on GP balance sheets and extended exit windows, the cost of finding out late is structurally higher than it has ever been.

Why it does not exist in the dominant tools

Time-to-consequence requires three things the dominant monitoring stack does not have. First, a structured representation of the conditions a position depends on — covenants, thesis assumptions, structural triggers. Second, a forward projection model that can run the live data through those conditions. Third, a unified ranking surface across the entire book.

iLEVEL ingests the KPI. It does not project the KPI forward against a covenant clause. Allvue rolls up the fund. It does not surface a time-to-consequence ranking on the underlying positions. Chronograph and Cobalt focus on benchmarking and LP reporting. None of them was designed around the forward projection layer.

Three things change when the metric is live

The IC review reorders. Positions get ranked by remaining runway, not by colour code. The team spends its attention on the things where the time-to-act is shortest — not the things that happen to have a red dot this quarter.

The escalation logic gets cleaner.When time-to-consequence on a position drops below a threshold, the escalation is automatic. The principal does not have to be the one watching — the structured record is.

The institutional memory is durable.The senior PM who is reflexively doing this in their head no longer has to. The structured record holds the calculation, and a new analyst joining the team inherits the same view of the book the senior PM has.

How Capital Refinery computes it

Capital Refinery parses the credit agreement (for credit positions) and the IC memo (for equity positions) into a structured set of conditions. The live operator data — accounting feeds, KPIs, narrative — is bound to the condition it tests. A forward projection runs continuously, propagating the latest data forward under the base case and a configurable shock pack. Each position gets a live time-to-consequence score in quarters, and the book gets a unified ranking surface.

It is the same calculation the senior PM is doing in their head — except consistent, durable, and present on every position the firm holds.