You’ve done your financial model going several years out.  You’ve got some nice assumptions laid out that are documented well.  You have the full set of financials, profit and loss, balance sheet and cash flow.  Plus, you’ve got various supporting schedules to support your revenues and various departmental expenses.

So, are you done at this point?  How do you know if your model will work correctly going forward?

I am a big believer in looking forward, but this can be one example where looking back can be very meaningful.

Here is a test that I would want you to take.  Suppose you have a model going out three years covering 2007, 2008 and 2009.  You’ve got the assumption laid out for those different years, which drive this model.  Now, we want you to do this test.  Create another column to the left of 2007 for the entire model.  This column will be for 2006.  What we want you to do is to put an assumption for 2006 and see how the model would come up for numbers for 2006.  This then gives you a great basis for comparing the numbers.  For example:

1.  Were there any items that you did not have forecasted at all that should’ve been?  This would be for things that would continue going forward, not necessarily things that might’ve finished off in 2006, such as a particular customer base or product or service line.

2.  How close did the numbers match up?  If there was a wide gap in certain expenses, that could be a sign that maybe your model isn’t working quite right.  Perhaps, there might be a different driver that influences that expense versus what you were using in your assumption.  Or, perhaps, the driver’s okay, but just the waiting is off.  Maybe, you’re actual costs might be $1500 per square foot versus the $1000 per square foot that you have in your model.

We did this recently with a client that was putting together a business plan to raise capital.  They didn’t have historical numbers in there from 2006 and I wanted them to put them in.  They did exactly this test and they found out through one of their royalty expense line items that it was running way short.  Upon looking at it more closely, they realized they were underestimating the number of customers in a couple product lines.  This was underestimating revenues and in turn was underestimating a significant royalty expense.

So to make sure your model can help you drive the business going forwards, give it a test driving backwards.  If it can work driving backwards, then you’ve got a greater confidence that it can work going forward as well.