ADP and ACP
Compare average deferral rates (ADP) and match plus after-tax rates (ACP) for HCEs against non-HCEs.
WORKFLOW LIBRARY · ANNUAL COMPLIANCE
A nondiscrimination testing workflow for 401(k) plans.
How a TPA team moves from census prep through ADP/ACP, 410(b) coverage, top-heavy, and 402(g) and 415(c) limit monitoring to corrections and a reviewed testing report, and what to look for in nondiscrimination testing software for 401(k) plans.
WHY THIS STAGE MATTERS
Testing is where a year of data gets judged.
Annual nondiscrimination testing checks that a 401(k) plan does not favor highly compensated employees (HCEs) over everyone else. A failed test is not a disaster if it is caught and corrected on time. A failed test discovered late, or a test run on a census nobody scrubbed, can mean excise taxes, plan corrections, and an unhappy sponsor.
The results are only as good as the inputs. Most testing problems trace back to the census: missing hours, wrong hire dates, compensation that uses the wrong definition, or owners and family members who were never flagged. That is why a good testing workflow spends as much time on data preparation and review as it does on the calculations themselves.
Coverage and top-heavy
Check 410(b) coverage and whether key employees hold more than 60% of plan account balances.
Individual limits
Monitor 402(g) elective deferral limits and 415(c) annual additions limits for each participant.
STEP-BY-STEP PROCESS
Seven steps from census to signed-off results.
Request and scrub the census
Send the year-end census request early and check what comes back: dates of birth, hire, termination, and rehire; hours; plan-year compensation; deferrals; and ownership. Reconcile deferral totals to what was posted from payroll during the year. The census validation tool catches common gaps before they reach testing.
Confirm eligibility and classify employees
Determine who was eligible during the plan year using the plan's eligibility and entry rules, then identify HCEs and key employees under current IRS definitions, including ownership attribution. The eligibility entry date checker helps with entry date questions.
Run ADP and ACP tests
The basic limit: the HCE average may not exceed the greater of 1.25 times the non-HCE average, or the lesser of 2 times the non-HCE average and the non-HCE average plus 2 percentage points. For example, if non-HCEs average 4%, the HCE limit is the greater of 5% and the lesser of 8% and 6%, so 6%. Safe harbor plans that meet the requirements are treated as passing ADP, and ACP for qualifying matches. Try scenarios with the ADP/ACP test estimator.
Run coverage and top-heavy tests
Test 410(b) coverage to confirm the plan benefits enough non-HCEs relative to HCEs. Test top-heavy status: a plan is top-heavy when key employees hold more than 60% of plan account balances, which can trigger minimum contribution and vesting requirements under the plan document.
Check 402(g) and 415(c) limits
Flag participants whose elective deferrals exceed the 402(g) limit and whose total annual additions exceed the 415(c) limit. The 2026 415(c) dollar limit used as the default in our calculator is $72,000; confirm the current IRS figure. Check individuals with the 415(c) annual additions limit checker.
Decide and process corrections
For ADP/ACP failures, corrective distributions made within 2½ months after plan year end avoid the 10% employer excise tax, and correction is generally due within 12 months. A QNEC to non-HCEs is a common alternative where the plan allows it. 402(g) excess deferrals are generally distributed by April 15 of the following year. Model the cost of each option before recommending one to the sponsor.
Peer review and deliver the report
Have a qualified reviewer check inputs, classifications, and results before anything goes to the sponsor. Deliver a testing report that shows the results, the margin on each test, and any required actions with their deadlines, then record those deadlines on the compliance calendar.
COMMON EXCEPTIONS
Issues that derail testing season.
| Issue | Why it matters | How to handle it |
|---|---|---|
| Census deferrals do not match posted contributions | Testing on the wrong amounts produces the wrong result | Reconcile to the ledger and resolve differences with the sponsor before testing. |
| Owners or family members not flagged | HCE and key employee status may be wrong | Ask for an ownership list every year and apply attribution rules. |
| ADP failure near the deadline | Late corrective distributions trigger the excise tax | Run preliminary tests mid-year so failures surface early. |
| 402(g) excess across two employers | The plan may not see the other employer's deferrals | Ask participants to report excesses; distribute by April 15 when notified. |
| Match formula changed mid-year | ACP results and corrections shift | Confirm the formula and effective dates from the plan document before re-running. |
Run a preliminary test before plan year end. A projected failure in the fall gives the sponsor time to adjust deferrals or plan design; the same failure found in March leaves only correction.
CONTROLS AND EVIDENCE
Workpapers that support the conclusion.
Controls
- Census reconciled to posted contributions before testing
- HCE and key employee lists reviewed each year
- Plan provisions confirmed against the current document
- Peer review by a qualified reviewer before delivery
- Correction deadlines entered on the calendar the day results are final
Evidence to retain
- The census used, with the date received and any changes
- Test results with inputs, margins, and assumptions
- Correction calculations and sponsor approval
- Proof of corrective distributions or QNEC deposits
- The final testing report as delivered
IN THE PILOT BUILD
How the KinetiPlan pilot build supports this.
The Compliance Testing area in the pilot build runs annual nondiscrimination and limit testing on each plan's census. It uses synthetic demo data and is not a certified testing engine; results are for qualified staff review.
- Testing by plan year across all plans, with ADP, ACP, 410(b) coverage, top-heavy, 402(g), and 415(c) results per plan. Safe harbor plans show ADP/ACP as deemed to pass.
- Results show margins such as ADP and ACP margin in percentage points, the 410(b) ratio, and the top-heavy percentage.
- Plans that need attention are marked Action Required with a plain description, for example 402(g) excess for named participant counts with a distribute-by-April-15 note.
- Summary tiles for ADP/ACP pass rate, top-heavy monitoring, 415(c) monitoring against a $72,000 default cap, Form 5500 readiness, and estimated corrective distributions.
- Change the match formula and re-run to see corrective distributions and a QNEC fix move. Each plan has a testing report and a link to its testing task in the annual administration project.
See the screens on the pilot build tour.
REGULATORY TASK
Mid-year ADP/ACP testing projection
The cheapest testing failure is the one found in July instead of the following March. Mid-year ADP/ACP testing projection software exists to give a TPA that early warning while deferral elections and match formulas can still change.
- In the pilot build today: the testing suite runs ADP, ACP, 410(b) coverage, top-heavy and 402(g) checks on a plan's current census at any point in the year. You can change the match formula and re-run to see how corrective distributions and a QNEC fix would move. That makes a mid-year screen on year-to-date data a normal step, not a special project.
- Not automated today: annualizing year-to-date compensation and deferrals into a projected year-end result. Staff apply their own projection assumptions, and results are for review by qualified staff. It is not a certified testing engine.
For a quick averages-only screen before you open the platform, use the ADP/ACP test estimator.
FAQ
Common questions
What is the ADP test limit?
The average deferral percentage of HCEs may not exceed the greater of 1.25 times the non-HCE average, or the lesser of 2 times the non-HCE average and the non-HCE average plus 2 percentage points. The ACP test uses the same structure for matching and after-tax contributions. Confirm current IRS rules and your plan's testing method.
What is the deadline for correcting a failed ADP or ACP test?
Corrective distributions made within 2½ months after the end of the plan year avoid the 10% employer excise tax, and correction is generally required within 12 months after plan year end. Confirm current IRS guidance and plan terms.
When is a 401(k) plan top-heavy?
A plan is generally top-heavy when key employees hold more than 60% of plan account balances. Top-heavy status can require minimum contributions and faster vesting for non-key employees under the plan document.
When must 402(g) excess deferrals be distributed?
Excess deferrals are generally distributed by April 15 of the year following the year of the excess. Confirm current IRS guidance.
Does the KinetiPlan pilot build replace review by qualified staff?
No. The pilot build runs on synthetic demo data and is not a certified testing engine. Testing results are inputs for review and sign-off by qualified staff.
This guide is general educational information about retirement plan operations, not legal, tax, or investment advice. Confirm current IRS and DOL rules and the terms of each plan document before acting.
SEE IT ON YOUR OWN WORKFLOW
Walk through a testing run in the pilot build.
We will run ADP/ACP and limit testing on a demo census, change the match formula, and show how corrections and action items update.