| |
The Brode Report | August, 2013
How to Nail This Year's Budgeting Process
 |
|
Hi,
I apologize for the long lag between newsletters—this is the first significant gap I’ve had since 2010. It was for good reasons, primarily being highly focused on a large telecom project. But with my kids starting school today (a senior and two sophomores in high school), I thought it high time to touch base again. I hope you had a great summer and are energized for the year to come.
Best regards,
David |
| |
|
|
|
| |
How to Nail This Year's Budgeting Process
Nine years ago I wrote about budgeting best practices. Below I review excerpts from that newsletter in light of the recent budgeting exercise I just completed (Careful there, lad. That’s just asking for the gods to ask for another change, kind of like calling a file “FINAL.”).
2004 View: The budget is primarily a measuring stick, not a
forecast.
 The real forecast is set forth in the strategic plan. Thus the annual budget is only useful for filling in details that we can track against. Furthermore, only budget a line item if the accounting system can report against it. The budgeting exercise is useless if you cannot track progress
against the budget numbers.
2013 Update:
There are really two ideas here: (a) that the budget is about filling in details versus real forecasting, and (b) that you should only budget items you can track against. I still believe both of these are true, but I’ll add some nuance.
 With respect to (a), this is true only if the budgeting process starts with senior management providing guidance as to the broad parameters of the budget by using the strategic planning model. But not every company does this, and in those cases the budgeting process really does become a forecasting process as well. I think this is especially true in startup environments.
With respect to (b), this is also tricky in a more dynamic startup environment. If the accounting system is still pliable, you may budget for what’s important to you and then hope that the accounting system will catch up.
2004 View:
Headcount costs deserve a detailed look.
 Salary should be planned at the employee level. When listing employees, be sure to code them by department so the expenses can appear in the
group’s final budget. Benefits and taxes should be calculated according to general rules and charged to the department.
2013 Update:
There’s nothing controversial here. I’d add that a budgeting system also needs to track full time equivalents (FTEs) so senior management can understand the level of effort going to each department. A recent budget analysis I did found that FTEs explained 95% of the opex variance among departments, so a strong personnel module is essential to the budgeting process.
2004 View: Other categories should follow the 80/20 rule.
 One recent client was budgeting over forty items for each department. Relatively simple analysis showed that 15 categories accounted for 90% of expenses. So these categories were budgeted directly and the remaining
categories were budgeted using one “Other” line. Should corporate require displaying detail for the extra 25 accounts, create an allocation process to fill out detailed general ledger accounts from the Other line using historical ratios. Granted, it is an extra step, but the time savings up front from not focusing on minor items frees up the time to do this simple process at the end.
2013 Update:
There is often a process of translating the final budgeting product to something that can be loaded into the general ledger system, and the idea of allocating from a bucket of the last 10-20% above is one example of that. You need to make sure that this category doesn’t fall prey to cuts when managers are trying to force the budget into a specific number. When the line items were individual and smaller this was a harder target, but by grouping them it’s now large enough to be vulnerable to attack.
2004 View: Plan often.
 Annual plans are a relic of a bygone era. Conditions change rapidly and require us to respond rapidly. Fortunately, we have the technology to do so. Budgets should be re-set quarterly and the annual budget process should not be significantly different from the quarterly budget forecast. Of course this raises issues about whether managers will sandbag the last three quarters of the forecast since it will be redone in three months, but there are numerous ways to discourage this behavior.
2013 Update:
I see only the fastest-changing, most dynamic organizations engaging in quarterly budgeting exercises, and typically it’s done in a smaller team than the annual effort. For most companies, the annual budget approved in the fall is still the touchstone against which the company will be measured throughout the year.
Back in 2004 I had specific recommendations:
- Plan and review at a higher level than is traditional.
- Create a process efficient enough to be used quarterly, if not monthly.
- Ensure spending can be tracked against the budget.
 Now I’d add a few more:
I see only the fastest-changing, most dynamic organizations engaging in quarterly budgeting exercises, and typically it’s done in a smaller team than the annual effort. For most companies, the annual budget approved in the fall is still the touchstone against which the company will be measured throughout the year.
-
Have a centralized database for shared access to information.
You know I’m a major spreadsheet fan, and it’s still easy to admit that an Excel-only budgeting process has significant weaknesses. Only one person really has access to the final data for reporting purposes, and this slows production during crunch time. There are many commercial products that can take data from Excel and share it with a group, and finance groups should adopt one of these.
- Implement strong version control.
Changes happen quickly at crunch time. If you’re using spreadsheets, save frequently with date and time stamps and note the changes from version to version. Don’t let the base file fork as when two people work on copies at the same time!
- Design output before the process begins.
This is really a generic step for any analysis, but it sure helps if you know where you’re trying to go. Knowing the output required at the end lets staff customize tools before you’re reviewing numbers after midnight.
|
|
|
|
