Reconciliation Tests for Time Accounting

Reconciliation Tests for Time Accounting are the six checks a workforce function runs on its Agent Time Accounting Model to establish that the ledger closes, that every logged minute carries a state the dictionary knows, that no scheduled minute is unexplained, that no minute is claimed by two systems, that recorded in-work time is consistent with the units of work closed, and that no production is leaking into an available state. The tests carry ids T-01 to T-06, which the Source of Truth Register, the Data Lineage Card for Planning Models and The Data Definition Card cite. Each test names the ledger rows it compares, a pass condition with a locally set tolerance, a failure signature, and what the failure means for the capacity model. The tests are part of the Workforce Data Definitions Framework; the state-to-bucket mapping they rely on is on Agent State Taxonomy and Activity Codes, and the population-level reconciliation of who is counted at all, Headcount Reconciliation in Service Operations, is assumed done.

The tests are the data-quality dimensions of the data-management body of knowledge, completeness, consistency, uniqueness and validity, applied to one object with one identity to satisfy, and they follow the standards' principle that data quality is measured against a stated requirement rather than asserted. The practitioner texts describe the same arithmetic as a planning chain; the tests make it a control with a result, on the reasoning Activity-Code Consistency gives for its own check: a rule never run is indistinguishable from no rule.

Test design

Field	Rule
Rows compared	named ledger tiers or buckets, and the system-of-record rows from the Source of Truth Register they are checked against
Pass condition	an identity or an inequality on those rows
Tolerance	SET LOCALLY; published with the result
Failure signature	what the numbers look like when the test fails, including the sign of the discrepancy
Meaning for the capacity model	which register entry or ladder rung is wrong, and in which direction
Owner	the data and analytics lead runs the suite; the definitions owner adjudicates

The six tests

T-01 Sum-to-paid

Rows compared: paid time (tier 1, from the payroll or invoiced-hours row) against the sum of every leaf bucket: off-schedule paid, scheduled-not-logged, each non-productive state class, available-idle, and in-work per channel. Pass: residual = paid − Σ buckets, within tolerance and non-negative. Failure signature: a positive residual above tolerance means minutes no bucket claims; a negative residual means the buckets claim more minutes than were paid, which is a double count that test T-04 localizes. Meaning for the capacity model: the shrinkage denominator and the roll-up's shrinkage rungs are wrong by the residual's share; a positive residual understates shrinkage and overstates supply, a negative one the reverse. The test assumes shrinkage is computed as paid minus productive on a paid denominator, under which the residual is already counted as non-productive, as the model page's example does. Until the residual is explained, every ratio on the ledger is blocked from publication at [M] and carries [E], with the residual as its range.

T-02 State coverage

Rows compared: logged-in time (tier 3, from the platform session log) against the sum of state time by code (tier 4), where every code is in the dictionary or in the mapping table of inherited codes with a treatment. Pass: logged = Σ state time, and the minutes in codes outside both tables are within tolerance. Failure signature: logged minutes with no state event, or minutes in a code the dictionary does not know, typically a site-created auxiliary reason or a catch-all other. Meaning for the capacity model: the shrinkage class of those minutes is unknown, so shrinkage, occupancy's denominator and adherence are each uncertain by that amount; cross-node comparisons are published with the failing node excluded, as Activity-Code Consistency requires, or held.

T-03 Orphan time

Rows compared: scheduled time (tier 2, from the published roster) against logged-in time (tier 3) plus recorded exceptions: absence, late and early codes from the absence feed and the real-time log. Pass: every scheduled minute is either logged or carries an exception code; the uncovered share is within tolerance. Failure signature: scheduled minutes with neither a session nor an exception, concentrated at shift edges or at seats whose platform emits no session event. Meaning for the capacity model: unplanned shrinkage is understated and supply the plan assumed was never delivered; the in-schedule shrinkage rung is too low and the next cycle inherits the error. Orphan minutes are posted inside scheduled-not-logged flagged as orphan, never absorbed into an exception class, until the exception feed explains them.

T-04 Cross-system double count

Rows compared: for each agent and minute, the in-work claims of the telephony contact record, the chat session log and the work-item timer, against the single bucket the precedence rule on Agent State Taxonomy and Activity Codes allows. Pass: Σ in-work across channels ≤ productive time for every agent-interval, and the count of minutes with more than one claimant is within tolerance. Failure signature: in-work exceeding productive or logged time; occupancy above 100%; chat elapsed time counted as focus time; an email timer running through a voice contact. Meaning for the capacity model: workload is overstated on the channel that lost the precedence contest, concurrency is mis-parameterized, and the combined requirement in Multichannel Capacity Model Assembly is inflated; the losing channel's handle-time entry drops to [A].

T-05 Item-count against time

Rows compared: items closed on a channel (the work-item row, CON-01 units) multiplied by handle time per asynchronous item (AHT-02, RG-008, the asynchronous definition The Data Definition Card introduces; AHT-01 stays the session-channel definition), against in-work time recorded for that channel (tier 6). The handle time used is the plan or prior-cycle factor, or one from an independent instrument such as time-and-motion sampling; a same-period handle time derived from the same in-work row makes the ratio 1.0 by construction. Pass: the ratio of implied to recorded in-work time is within tolerance of 1.0. Failure signature: implied time well below recorded means agents sat in a work state without closing items (idle-in-state, batch closing, pended items held open); implied well above recorded means items closed outside any in-work state, which test T-06 isolates. Meaning for the capacity model: the handle time per item is wrong or the time is misallocated, the roll-up's reverse walk will not reproduce the ledger, and the asynchronous requirement is wrong in the direction of the ratio; AHT-02 for the channel is re-measured before the next cycle.

T-06 Availability-bleed

Rows compared: items closed, by agent and interval, against the minutes the same agent spent in a telephony available state with no contact in hand (the available-idle bucket). Pass: items closed during available-idle minutes are within tolerance of zero. Failure signature: a steady rate of items closed per hour of available time at blended seats, with low occupancy and near-zero recorded in-work time on the asynchronous channel. Meaning for the capacity model: available-idle is contaminated with production, so voice occupancy is understated, the voice model finds the pool overstaffed, and the asynchronous workload is invisible to the plan; Production Accounting for Blended Agents carries the separation methods, and until fixed the bleed is published beside occupancy.

Cadence

Test	Runs	Gate it precedes
T-01 Sum-to-paid	weekly and monthly on the closed period	the monthly Capacity Planning Cycle lock
T-02 State coverage	on every extract where the platform allows; otherwise daily	any cross-node report; the weekly forecast-against-actual review
T-03 Orphan time	daily on the previous day	the real-time lead's daily review of delivered supply
T-04 Cross-system double count	daily; on every extract at the automation layer	the intraday model's read of occupancy
T-05 Item-count against time	weekly, per asynchronous channel	the handle-time entry's re-measurement decision
T-06 Availability-bleed	weekly, per blended seat and group	the blended-pool requirement in the assembly

All six also run after any change the register's carry rule names (a platform migration, a channel launch, a dictionary version), because a number carried across such a change is asserted until re-measured, and the tests are how it is re-measured. The Time and Production Accounting Audit (generic process) runs the full suite once as the baseline.

What a failure means

A failure has four consequences, in order.

1. Disposition of the metric. Metrics that read the failing rows are published with the failure stated and their grade reduced per Human Gates and Number Grades, or held; none is published at [M].
2. A register row. The failure is filed against the source-of-truth row or definitions entry it implicates, with the test, period, discrepancy and sign. A repeated failure on one row is a defect in the instrument or definition, not the period, and goes to Definitions and Source-of-Truth Change Control (generic process).
3. The capacity model's correction. The implicated rung is re-measured or re-graded and the roll-up re-run; the plan that used it is annotated, never silently replaced.
4. The planner gate. A number that fails entry conditions does not enter the model; Data Quality Gates for Planning Inputs holds the conditions and the quarantine.

Two failures outrank the rest: a negative residual on test T-01, which makes every published efficiency figure suspect, and a persistent failure on test T-06, which means one channel has been planned on the time of another and the correction moves both requirements at once.

A headcount residual from Headcount Reconciliation in Service Operations and a time residual from test T-01 are different findings and are never netted; the general data-quality framework is on WFM Data Governance and Quality.

Maturity Model Position

Four scales on this wiki use the word level; the launch page states which is which. On the WFM Labs Maturity Model™ a Level 2 function runs tests T-01 and T-03 by hand at month end and publishes the residual. Level 3 runs all six on the stated cadence with tolerances and files failures as register rows, the state the automation layer requires before any rule reads the ledger. At Level 4 tests T-02 and T-04 run on every extract, and the release of any rule that reads occupancy or state is conditional on their passing, which is the point at which the tests become part of the catch rate (the proportion of a rule's known errors that verification detects, per Human Gates and Number Grades) rather than a monthly report.

See Also

• Workforce Data Definitions Framework — the hub this test suite belongs to
• Agent Time Accounting Model — the ledger and the closure identity under test
• Agent State Taxonomy and Activity Codes — the precedence rule test T-04 applies
• Workload Units and the Roll-Up to FTE — the ladder whose reverse walk test T-05 formalizes
• Headcount Reconciliation in Service Operations — the population reconciliation this page assumes
• Time and Production Accounting Audit (generic process) — the audit that runs these tests across every capture point; see also Production Accounting for Blended Agents