About one hour of auditor attention for stage-one testing
CLA worked with Clark on sampling, contributions, and eligibility testing. Five benchmark plans across three recordkeepers established the timing; 261 repeated runs carried the method through the season. Auditors retained the parameters, exceptions, review, and sign-off.
CLA (CliftonLarsonAllen) · Employee benefit plan audit case study
Engagement at a glance
- Benchmark plans
- 5
- Recordkeepers
- 3
- Auditor attention
- 1.06 hrs
- Effective time
- 3 hrs 5 min
- Repeated runs
- 261
- Documents incl. repeats
- 9,166
The benchmark used a working review, not full detail review to sign-off standard. Effective time is per plan with two plans running concurrently. Usage totals include repeated runs, not unique engagements.
A focused first stage
Clark and Ava build the population, select the sample, test items against evidence, and document the result. CLA auditors helped develop the method across Fidelity, Empower, and ADP formats. This stage covers sampling, contributions, and eligibility; it does not represent a complete audit.
What the benchmark showed
The five plans averaged 1.06 hours of active auditor attention and 3 hours, 5 minutes of effective time per plan with two running concurrently. Effective time was about 85% below an allocated industry-average baseline of 21 hours, 2 minutes. That comparison is not a measured before-and-after reduction at CLA, and a complete detail review adds auditor time.
A method used through the season
From June 12 to September 10, 2026, CLA's account recorded 261 repeated runs, processing 9,166 documents and 1.58 million population lines, including repeats. The study models additional capacity from broader adoption and future coverage; those projections are not realized outcomes. The next proposed step is to document the method, teach it more widely, and validate work through full review and sign-off.
See Clark in action
The peer review guarantee, in practice
Watch Ava prepare a workpaper, then trace every figure back to the source document, the change, and the approval.