The Wrong Question Has a Price Tag
Imagine your CFO presents the annual financial statements. You ask: “Were these audited?” She says yes. You nod, sign off, and move on.
No auditor’s report. No itemized list of accounts examined. No methodology. No signature from a licensed firm taking accountability. Just the word: yes.
That scenario is absurd. In financial reporting, “we tested it” without evidence is not a test. It is a statement. The audit report is not bureaucracy — it is the only thing that makes the word “audited” mean anything at all.
And yet, every week, boards and C-suites sign off on data project deliveries with exactly that level of rigor. The vendor says yes. The project manager nods. The invoice gets approved.
The Real Question Is: Why Doesn't Anyone Ask?
The tools to demand evidence exist. The contractual right to require it is unambiguous. Data quality checks can be documented, timestamped, and handed over just like any other deliverable.
So why does the room stay quiet?
Because it feels like an accusation. Asking a vendor to prove their work implies they may have cut corners. In enterprise culture, where relationships get built over months of steering committees and status calls, that question feels confrontational. So it gets swallowed. Everyone smiles. The go-live email goes out.
Because the language belongs to the wrong room. “Validation suite.” “Automated test coverage.” “Pipeline observability.” These are engineering words. They never make it into the C-suite conversation, so the people with the authority to demand evidence never think to ask for it. By the time the topic reaches the board, it has been reframed as a technical problem — not a governance one.
Because accountability has no clear owner. The business assumes IT signed off. IT assumes the vendor validated. The vendor assumed the client defined the acceptance criteria. Nobody ran a comprehensive, documented check. And because nobody admits this out loud, nobody is responsible. Until something breaks.
Good Intentions Are Not a System
Some organizations have a motivated project manager who pushes for test documentation on this project. With this vendor. Under these particular circumstances.
And then the next project starts. Different vendor. Tighter deadline. New team. The habit doesn’t travel — because it was never a habit. It was one person asking the right question on a good day.
This is what corporate policy is for.
Organizations that consistently deliver reliable data do not rely on individuals asking the right questions at the right moment. They build the requirement into the contract, the delivery checklist, and the payment terms:
No delivery is accepted without documented test results.
Automated checks are a contractual deliverable, not a courtesy.
Data quality accountability is owned at C-suite level.
Without policy, you are gambling on whether the right person is in the room. With policy, the evidence is a condition of payment — just like the auditor’s signature is a condition of filing the accounts.
What the Evidence Actually Looks Like
You do not need to understand data engineering to hold a vendor accountable. You need to ask one question, and refuse to move forward until it is answered:
“Can you show me the test results?”
Not “did you test it.” Show me.
A proper delivery includes a documented record: which checks ran, against which data, at what point, and what the outcome was. Pass. Fail. Remediated. Signed off.
If that document does not exist, the vendor has delivered a system in an unknown state. It may work perfectly. It may contain errors that will not surface for six months. You have no way to know — and neither do they.
CAT Makes the Evidence the Default
CAT — a data quality observability tool — closes the gap between policy intention and operational reality. Every pipeline, every integration, every critical data feed runs against a defined set of checks. Results are logged, timestamped, and accessible on demand.
When a vendor delivers, you do not ask whether they tested it. You look at the results together.
When a new data source comes in, the new checks should be added “automatically” — not because someone remembered, but because that is how the system works.
When your auditors, regulators, or board ask about data integrity, the answer is not a verbal assurance. It is a report.
The standard already exists in every other discipline where quality matters. Financial statements require audited accounts. Buildings require certified inspections. Medical devices require documented clinical trials.
Data infrastructure is no different. The question is simply whether you are willing to require the same standard from the people you pay to build it.
The conversation starts with two words: SHOW ME.
About the Author
I help organizations build trust in their data by making automated data testing a natural part of every project. At CAT, I work closely with corporate data leaders and our partners to ensure strategies, platforms, and teams rely on reliable, validated data – whether in banking, insurance, e-commerce, AI initiatives or F1 team. With more than 15 years of experience in the data field, I combine technical expertise with business insight to guide clients, mentor teams, and contribute actively to the data quality community.
