Blog · Territory, capacity and quota planning · Tax and accounting
How an accounting firm turns its engagement list and filing calendar into a load curve per manager: filings due per week, the hours each typically takes from the firm's own time entries, the hours available from the roster, the weeks where load exceeds capacity months ahead, and the reassignments and early starts that flatten the peak before the season does it the hard way.
An accounting firm knows its filing deadlines and knows its staff, and meets the collision between them every season in the last fortnight before the date. The engagement list, the time entries and the roster can draw the load curve per manager in advance. This guide sets out the curve, the peaks, and the moves that flatten them.
Per engagement:
Deadline, from the filing calendar Hours = last year's hours on the equivalent engagement, or the type median Lead time = weeks before the deadline over which the hours are spread
Per manager, per week:
Load = Σ hours of engagements in their lead-time window that week Capacity = roster hours − leave − non-chargeable allowance Excess = load − capacity, floored at zero
Client and manager identifiers only.
Σ managers' load per week = Σ engagements' hours spread over their windows
Every engagement has a manager and a deadline. Those without are listed; an engagement with no deadline is on nobody's curve.
One manager, the eight weeks to a filing peak.
| Week | Filings due | Load hours | Capacity | Excess |
|---|---|---|---|---|
| −7 | 2 | 18 | 34 | |
| −6 | 4 | 26 | 34 | |
| −5 | 6 | 31 | 34 | |
| −4 | 9 | 42 | 34 | 8 |
| −3 | 14 | 61 | 30 (leave) | 31 |
| −2 | 22 | 88 | 34 | 54 |
| −1 | 31 | 104 | 34 | 70 |
| 0 | 18 | 52 | 34 | 18 |
A hundred and eighty hours of excess in four weeks, on one manager, visible eight weeks out. Weeks minus seven to minus five have fifty hours of slack.
| Move | Effect on peak |
|---|---|
| Start the four largest engagements three weeks early | −38 hours from weeks −2 and −1 |
| Reassign six engagements to a manager with slack | −52 hours |
| Cover the leave week | −4 hours of lost capacity |
| Remaining excess | 86 hours, planned overtime rather than discovered |
The curve after the moves is on the same page as the curve before. The season is still busy; it is no longer a surprise.
Deadlines counted, not hours. Thirty-one filings in a week is a number; a hundred and four hours is a plan.
Hours from a guess. Last year's time entries are the fact.
Leave not in capacity. The peak week is worse than the curve says.
Curve drawn in the peak. The moves need weeks to work. Eight weeks out is the minimum.
Mapped once, the engagement list, the time entries and the roster produce the load curve per manager, the excess and the effect of each move every week. Covirage builds this from the exports as they are. The tax and accounting page describes the setup, and the season watch guide covers the clients whose work has not started, which is the other half of the same season.
The time entries on the same client's equivalent engagement last year, or the median for engagements of that type and client size where there is no history. Stated per engagement, and the same method for every manager.
The number of weeks before the deadline in which the work is typically done, from the firm's own history: a return due on the 31st is usually worked in the three weeks before. The load is spread over those weeks. The lead time is stated and can differ by engagement type.
Roster hours per manager per week, less leave, less a stated allowance for non-chargeable work. It is the same figure the firm uses for utilisation, so the two reports agree.