Budgeting Process Steps

[Former Member]

Hi,

As far as i can comprehend from budgeting textbooks, the few major steps involved in coming up with a detailed budget involved (regardless of software at this moment):

1) study external and internal factors,

2) C level people (CFO, CEO..) come up with corporate objectives in the form of high level numbers, i.e. gross margin, ROI, net income,

3) C level people discuss with senior management on the numbers for corporate objectives,

4) senior management will be guided by the corporate objectives created in step 2, to come up with master financial statements, such as master PnL, master balance sheet, master cash flow

5) equipped with the master forecast financial statements, allocation is done so that certain entities have certain numbers, i.e. sales, cost... to hit

how different is this textbook suggestion from how the real world budgeting is done? and i am a bit confused about step 5 as well, i mean if allocation is done to such a detailed level, where is the contribution from the operation level people? certainly between senior management and operation level, there must be some space that allows creativity of detailed budget creation so that challenges on the numbers can happen. or maybe my concept of allocation is wrong, since i thought everything from sales, raw material cost, sales and admin cost is allocated. it doesn't really make sense, but what number is actually allocated then?

any input on this?

cheers

[Former Member · Accepted solution]

hi

there will be two approaches. Top Down And Bottom Up.

In top down the Senior Mgmt will consider the past historical data and do increase/decrease on

some %basis. Also considers general factors.(Economy Recently , Politial etc.)

In Bottom Up the planning will be done at the lower level and consolidation will be done at the higher level and the comparision will be done for both top down and bottom up and Finally Top Mgmt

gives a final budget version.

srinath

[Former Member]

Hi,

In real term scenario, the budget implementation is quite similar to the one specified. However, the one specified only talks about top down approach. In reality, most of the organizations follow a Counter - Current approach (combination of both top - down and bottom - up).

The following points should be kept in mind while implementing Budgeting Process with BPC:

1. Budget approach

a. Frequency and timing

Will budget information be entered monthly, quarterly, yearly?

Will budgeting and forecasting happen at the same time or at different times?

b. Iterations

If only the current budget or forecast is required, then a single budget or forecast category is all that is needed

If multiple versions of the budget need to be saved, members such as BUDGETV1, BUDGETV2u2026will be needed

If a monthly version of the forecast needs to be saved, members such as JANFCST, FEBFCSTu2026will be needed

c. Best Practice: Multiple iterations

Create a single category that users write to (e.g. Current Budget)

When a version of the Current Budget needs to be saved, the Administrator can use a Data Manager copy package to copy that data to another category member (e.g. Budget Version 1)

2. Modeling requirements and budget method

a. Level of detail

What level is being budgeted for in every dimension?

The best practice is to budget at the base level in every dimension

b. Approaches for budgeting at a level that is higher than base:

Separate application: If you need to model at aggregated levels across most of the dimensions and detail is not important

Separate hierarchy: For example, if certain budget inputs need to occur at the total year level, set up a separate hierarchy that includes only total year members for each year

Smart input: Use the existing full hierarchy, but only input to certain members. For example, to budget for all products in a dimension with individual products at the base level, input the value that represents the total for all products into one of the base level products and let it aggregate to the total

c. Top Down

Allows for inputs that represent aggregated values to be allocated down to the base members of one or many dimensions

d. Bottom Up

Allows for inputs at the base level of every dimension. This does not require that every member in every dimension is budgeted for.

Allows budget creators to input their numbers and provide justification, rather than having a number assigned to them

e. Budgeting Tips

Auditing: in most cases, auditing is not turned on for Budget or Forecast categories

Insight: the Budget detail should be the same as the Actual detail if using Insight

Utilize optimization to improve performance

3. Seeding requirements

a. Zero base: No data seeded as a starting point

b. Prior year actual: Data manager copy package is used to copy actual data from the prior year to the budget for the current year

c. Current forecast: Data manager copy package is used to copy current forecast data to the budget for the current year

d. Percent increase or decrease of actual or forecast: Data manager copy package is used in conjunction with conversion and transformation files which define the calculation

4. Input schedules

a. Input schedules can be broken out by:

Business process

Account classification

b. How input templates are broken out typically depends on:

Budget process

Budget creators

Structure of the company

c. Functions used to send data to the database:

EVSND gives cell by cell control and is best used when you need to convert an existing schedule

EVDRE performs better and is more scalable

d. Built u2013 in modeling functions

Spread: allows you to take a value and allocate it among selected destination cells. The value may be allocated equally, or it may be weighted by an optional range of weights

Trend: takes a value and grows it by a factor among selected cells. The source value will be placed in the first cell of the destination range. Each subsequent cell in the destination will increase by the amount specified

Weight: allows you to allocate an amount by weighted factors in your spreadsheet

5. Business process flows

a. EVASV Function: status reporting

Returns the approval status codes for a given category, entity, and time member

Can also be used to return the owner and the time the status was updated

EVASV (AppName, Category, Entity, Time, Parameter)

The Parameter returns one of the following:

u2022 0 u2013 (Default) Returns the status code

u2022 1 u2013 Returns the person who updated the status code

u2022 2 u2013 Returns the time the status was updated

Budget Method:

u2022 Direct Input: the account is being populated via direct input into an input schedule that interfaces with the database

u2022 Calculated (Driver Based): the account is calculated based on another account value. One account u201Cdrivesu201D the value in the second account. An account that is a driver may drive the value of more than one target account

Hope these points help. Please award points if helpful.

[Former Member]

Hi,

Thanks for your lengthy elaboration, however i have some more questions:

1. Budget approach

a. Frequency and timing

Will budgeting and forecasting happen at the same time or at different times? - How do you differentiate budgeting from forecasting?

2. Modeling requirements and budget method

b. Approaches for budgeting at a level that is higher than base:

Separate application: If you need to model at aggregated levels across most of the dimensions and detail is not important

Separate hierarchy: For example, if certain budget inputs need to occur at the total year level, set up a separate hierarchy that includes only total year members for each year - Total year level sounds like aggregated levels to me, why is "total year level" done using a separate hierarchy instead of using a separate application?

There are a couple of points mentioned that i have never tried, i.e. EVASV and Driver Based Calculation.

Where do you find EVASV by the way because they are not found in "Insert Function"?

Thanks again. Points have been awarded.

Cheers

Edited by: lip chean soh on Dec 9, 2008 7:34 AM

[Former Member]

Hi

Depending upon the organization, Budgeting will be done only before the start of the fiscal year.

and in some companies in the mid of the year also. Maximum will be done 2 times.

OB: Opening : For the complete Fiscal Yr

RB: Revised Budget : 2nd Half year.

But Forecasting will be done Monthly / Weekly etc. depending upon organization models requirement.

Srinath

[Former Member]

hi

Driver based calc: eg: Sales Amt = Sales Qty * Sales Price

Direct Method Eg: Sales Qty : Input (Direct Input)

Srinath

[Former Member]

Hi,

Suppose, you want to enter values directly for 2005.total and 2006.total. However, according to the time dimension, 2005 and 2006 are not base members and you cannot enter data in any other nodes. So, the better way is to create one more hierarchy for years, so that you can directly enter values for the entire year instead of entering for every month.

The time dimension has been takes just as an example. This kind of scenario will hold good for other dimensions as well.

Driver based calculations are, in general, what we know as dimension logics. One dimension member drives another member.

Hope this helps. Kindly reciprocate by awarding points.

[Former Member]

We have often added a time dimension section that looks like the following, for the YEAR input process. Just be mindful that this hierarchy is seperated from the core reporting time hierarchy and is widely used for many purposes, and is one of the best methods for capturing and distributing Annual or non-month related data. So this hierarchy has its OWN top in the PARENTH1 hierarchy, but the great thing is that you can relate all these members using properties for logic and input/reporting schedules.

ID NEWID EVDESCRIPTION PARENTH1 SHORTDESC YEAR

xxxx.Anl Annual Member xxxx Yx xxxx xxxx

xxxx.Qx Annual Member xxxx xxxx.Anl Qx xxxx xxxx

2003.INP 2003 xxxx.Qx 2003 2003

2004.INP 2004 xxxx.Qx 2004 2004

2005.INP 2005 xxxx.Qx 2005 2005

2006.INP 2006 xxxx.Qx 2006 2006

2007.INP 2007 xxxx.Qx 2007 2007

2008.INP 2008 xxxx.Qx 2008 2008

2009.INP 2009 xxxx.Qx 2009 2009

2010.INP 2010 xxxx.Qx 2010 2010

2011.INP 2011 xxxx.Qx 2011 2011

2012.INP 2012 xxxx.Qx 2012 2012

THIS IS ONLY a small portion of the set of properties, but I hope this helps.

[Former Member]

Petar:

Thank you for sharing. I do have a few questions regarding your response, and hoping you can clarify for me.

1. Is this an "additional" TIME dimension or within the same TIME dimension?

2. what is xxxx.Qx? Are they different on yearly budgeting period? For instance, for 2008.INP, the H1 is 2008.Q5, maybe? and 2008.Q1-Q5 has H1=2008.Anl?

3. In my Time dimension, I also have a "YEAR" column. What specifically does this YEAR column do? Do we use it in script logic condition (i.e. TIME.YEAR=2008) or in reporting condition ([TIME].[YEAR]=2007)?

My questions may sound simple, so please bear with me. I am fairly new to BPC, and would like to understand a bit more thoroughly.

Thank you! Have a great day!

Brian

[Former Member]

Well, lets see if this helps.

- The extra members are a new Hierarchy, with a node of xxxx.ax at the top, inside th eexisting Time dimension.

- You have xxxx.ax at the top, the next child is xxxx.qx, then the years foloow. The YEARS.INP must be at the same level as the months.

- The YEAR column is just a property to indicate a grouping for many different purposes. And yes you can use it in a script if you like.

[Former Member]

To clarify one point Petar made -- you do NOT need to (and should not) add in a ParentH2 on your member sheet, to create a second hierarchy / roll-up in the cube.

If you look carefully at his screen dump of the member properties, you'll see it says ParentH1 as one of the column headings, although this is hard to visualize due to the variable-width font on SDN.

By "a new hierarchy" he means simply that xxxx.ax is a sibling to 2008.Total, 2009.Total, etc. and is at the top level of the H1.

[Former Member]

Hi Petar,

From what i can gather, it looks very much like the structure of the Apshell time dimension members, i.e. xxxx.Anl is year, xxxx.Qx is quarter, 2003.INP to 2012.INP are months.

you said "and is one of the best methods for capturing and distributing Annual or non-month related data." what do you mean and do you have a simple example for that?

where do i find EVASV within BPC?

cheers

Edited by: lip chean soh on Dec 10, 2008 4:01 AM