Industry Benchmark Survey on Data Governance: From captured data to control-grade evidence
The data behind surveillance is a matter of evidence, not only technology. This survey of 200 respondents covers the data behind trade and communications surveillance and the archive.
· A-Team Insight, FinTech Global and Behavox
Are your surveillance and archive fed by control-grade data?
How 200 respondents say they govern the data behind surveillance, and what they say they could show a regulator.
A research report by A-Team Insight, FinTech Global and Behavox. 19 November 2026. The download form is at the end of this page.
Summary
Surveillance and record-keeping rules expect firms to keep records and produce them when asked. The report sets out where respondents say their evidence chain holds, and where they say it breaks.
- Most respondents do not expect to produce the evidence on demand: 40% expect days of manual assembly, and a further 25% say they could not readily give an integrated answer.
- Most respondents (72%) report reconciling trade feeds into surveillance daily or faster, but fewer than half (45%) describe the link between trade and communications data as governed and automated.
- Nearly all respondents (94%) say data quality breaks are typically detected automatically, but most (64%) report fixing them manually.
All results are self-reported and describe these respondents only.
Want to talk the findings through with your team? Request a Benchmark Review session.
Complete the form below to download the full report, with every chart, the seven practices and the self-test.
What leading respondents do
Leading respondents, the 71 who say they could answer the regulator's question with documented evidence, report each of seven practices more often than the other 129. Each difference also points the same way among bank and non-bank respondents separately. The practices are:
- Breaks detected and remediated automatically: 54% of leading respondents against 16% of others.
- Material breaks resolved within a day: 59% against 21%.
- Data quality metrics reviewed at a governance forum: 77% against 42% (55% of all respondents).
- Governed, automated identity linkage across trade and communications data: 75% against 29%.
- A dedicated data governance function as primary owner: 44% against 23%. The smallest difference: most leading respondents use another model.
- Lineage documented and maintained from source to archive: 76% against 25%. Overlaps with answering the regulator's question, so a larger gap is expected.
- An archive with automated chain-of-custody controls: 73% against 33%. Also overlaps with the regulator's question.
Run the regulator's question on your own programme
A firm can replace self-assessment with evidence by running the regulator's question on its own programme:
- Choose a recent date and request the communications and trade data ingested that day.
- Check that every expected channel and feed is present.
- Show how each record was transformed between capture, surveillance and the archive.
- Confirm that the surveillance decisions that followed were retained.
- Time the answer and note any step that needed manual reconstruction.
The same steps apply to a vendor that manages part of the pipeline. Each handoff the test exposes needs a clear owner, because a named owner is not the same as an owned control.