Settlement fails, reconciliation breaks and corporate action errors are all measured at the moment they surface. That tells the firm how many there were. It does not tell anyone where they came from, which is the only thing that stops the next one.
1Why
Every firm can report its fail rate. Very few can say which upstream step produced it, because each platform in the chain reconciles to itself and the break lives in the space between them.
A daily fails report is a symptom log. It is accurate, it is watched closely, and it contains no information about the allocation, the standing instruction or the static data record that actually caused the entry.
Trade capture agrees with trade capture. The custodian agrees with the custodian. Every platform in the chain is internally consistent, which is exactly why a discrepancy between two of them can survive four stages before anyone notices.
Operations teams clear breaks every day and are measured on clearing them. Nobody is measured on where they originate, so the same root cause is re-cleared indefinitely rather than removed once.
Post-trade lifecycle
Fail rates are counted where they surface. Fixing them means knowing where they start, which no single post-trade system can tell you.
Illustrative shape of a common finding, not client data. The distance between origin and detection is measured from your own records.
2Where
Post-trade is where the evidence is densest and the cost of getting it wrong is most immediate. The same reconstruction applies across the client and control functions.
Where fails, breaks and capital cost are created.
Where regulatory standing and client experience are decided.
3How
No new instrumentation, no data warehouse programme and no curation layer to build first. Four steps, from access to a live view.
Status histories, workflow logs and audit trails in the systems you already run. If a record carries a case identifier, an activity and a timestamp, it is enough.
Records from separate systems are linked back into one case — a single instruction followed across everything that touched it, in the order it happened.
The process as it was intended, set against every path it really ran. Variants, rework loops, waiting time and team differences are counted rather than estimated.
The view refreshes as the data does, so drift shows up as it happens instead of surfacing in the next review cycle.
4What
Not a score or a maturity rating. The actual behaviour of your post-trade chain, in units your operations and risk teams already report on.
The step that produced the exception, rather than the step that reported it — so remediation lands on the cause.
How long an instruction carried a defect before anyone saw it, and what that delay costs in funding and buy-in exposure.
Which instructions were repaired by hand, how often, and by which teams — the work that never appears in the STP rate.
The same instrument type behaving differently by counterparty, market or custodian, instead of averaged into one firmwide number.
How many distinct paths one instruction type actually runs, and which of those paths concentrate the fails.
Which open instructions are drifting towards a fail while there is still time to intervene.
Pick the flow where fails cost you the most, in funding, penalties or client standing. We will tell you, before any commitment, whether the data you already hold can reconstruct it.
