The certification is the attestation. It is not the evidence.
Ask an institution to produce the evidence behind a certification and the most common answer is the certification. A signed record that says the distribution was correct, offered as proof that the distribution was correct.
That is circular and an auditor will say so. The attestation is a statement about the evidence, which means it cannot also be the evidence.
The standards are at 2 CFR 200.430(g). They moved there in the April 2024 revision, published at 89 FR 30136, and were previously at paragraph (i).
Records must be supported by a system of internal control giving reasonable assurance that charges are accurate, allowable and properly allocated. The word supporting all of it is system.
No certification form is required. The regulation does not mandate one and NIH has confirmed that no single method is prescribed, so the question is never whether you have the form.
The question is whether the control around the form produces reasonable assurance. That is what the evidence has to demonstrate.
These are categories rather than a template. How you hold them is your choice, and whether you hold them is not.
The payroll posted to each award for the period, taken from the accounting records rather than from the certification screen. This is the thing being attested to and it has to exist independently of the attestation.
Why the split was what it was: an appointment, a salary cap calculation, a cost share commitment, a documented change in assignment. Where charges were made on planned effort, the basis is the estimate and the regulation then requires an after-the-fact review, which becomes part of the file.
A record of why this person was in a position to know. The Uniform Guidance framing is suitable means of verification, which is why the principal investigator or a direct supervisor is usually the only defensible signatory.
When the certification was made relative to the period it covers, and every payroll adjustment posted to that award afterwards. A retroactive adjustment after certification means somebody attested to a distribution that no longer exists, and the file has to show how that was resolved.
Estimates may be used for interim accounting. Three conditions attach: the estimating system produces reasonable approximations, significant changes are promptly recorded, and periodic after-the-fact reviews are performed with adjustments made.
Budget estimates alone do not qualify as support for final charges. An institution charging on planned effort with no evidenced after-the-fact review has the first half of an allowed method and not the second.
That is a documentation finding rather than a misspending one, which is why it is easy to carry for years without noticing. It is also why the after-the-fact review, and not the certification, is usually the control actually being relied on.
This is what belongs in the file and why. It does not give you the evidence log, the sampling approach, the scoring, or the reconciliation between the certified distribution and the ledger.
Those are the instruments and they are what the paid review is. The same line the free health check holds.
2 CFR 200.430, eCFR. Paragraph (g) carries the documentation standards, including the conditions on using budget estimates.
NIH NOT-OD-18-108, standards for documentation of personnel expenses.
Testing effort certification in Workday is the companion to this page. Seven questions to ask about your own tenant.
The free health check costs nothing and takes about twenty minutes. The Sponsored Award Review is the scored version, where every check carries a second column asking whether you hold the evidence today.