A Beginner’s Guide to Variance Analysis – Part 1
Variance analysis isn’t just an idea we’ve come up with for a Brixx feature – it’s actually the most requested feature by Brixx users. The demand for a method to measure business performance against a forecast can’t be denied!
A couple of weeks ago we previewed the new Brixx feature – Actuals vs Forecast. In this series of three articles, I’m going to show you just how important actual vs forecast based variance analysis is – and how to plan effectively using it.
So, if you’re not thinking about variance analytics yet, should you be? And what are the best ways to go about it?
In this series, I’ll explain what variance analysis is, how to collect actual and forecast data, how to compare the two, and how to read the lessons variance analysis can give you on the future of your business.
What is variance and why is it important in business performance measurement?
A forecast is a forward-looking prediction of what a business’ finances are going to look like.
Variance analysis plays an important role in making financial forecasts more accurate, by comparing the forecast with actual data.
Looking at a detailed comparison of the forecast and actual data allows financial planners (including you, after reading this series)
Let’s unpack this a bit. We’ll start with my favourite topic – forecasting.
Forecasts are built by predicting how every moving-part of the business will perform financially in the future. For some things like regular bills, this is easy to predict. Other aspects of the business, like sales of a newly launched product or service, are harder to get right.
Accurate forecasts are important as businesses use them to plan the course of action today that will bring about the best results in the long term. In short, they help businesses make better decisions.
A great, simple example is, is it better to outsource a business task, like technical support, or hire someone to do it in-house?
Financially, outsourcing may be cheaper in the short term, but hiring someone may be cheaper in the long term, as well as bring other, less tangible benefits to the business.
The financial side of forecasting is rarely completely independent of other, non-monetary business concerns.
This is why it helps to think about a business’ finances in terms of what the business does, rather than accounting terminology – it makes business decisions like this easier to weigh up and take action on.
Because forecasts are so important to the decisions businesses are making today, they have to be well-researched.
When you make a forecast it is usually based on a combination of two things:
- Assumptions about what will happen in the future (will inflation continue to rise as it is? Will my sales increase over time?)
- Financial information on the past performance of your business, or businesses like yours.
Assumptions are always just that, assumptions. You could call them predictions. You could even call them guesses – but this is disingenuous to how good those “guesses” should be.
Why variance analysis can help improve financial forecasting
All assumptions should be based on as much real data as possible. But even real data, like the business’ past performance, isn’t a completely accurate guide to the future.
For one thing, relying on just the business’ past performance to make a forecast ignores the fact that this data is, by definition, out of date! It doesn’t include what your competitors are planning, new entries to your market, or any information about your own new or unreleased projects or business ideas. It’s the past. In forecasting, you must think about the future.
And the trouble with this is that you’ll never be quite right.
I’ve written extensively on forecasting, particularly sales forecasting, which is often clouded by optimistic thinking. But sales don’t happen in a vacuum. Sales happen because people are aware of what you are selling and motivated to buy your product or service. For new businesses, in particular, there can be an immense lead-time on getting the sales you need.
It’s not all doom and gloom of course! As soon as you start selling, you’re gaining information. Making contact with real customers, and starting to better understand how you should be selling and to whom.
As well as sales, other aspects of the business can be hard to predict. Logistics, delays, costs of delivery and unexpected expenses like legal fees, fines or repair & replacement costs can take a business by surprise.
Variance analysis is here to help with all of this. The purpose of variance analysis is to make your forecasting more accurate.
By committing to regular variance analysis you’ll see the gaps between your predictions and reality. Understanding the cause of these differences will help you to spot emerging trends and gaps in your predictions, and make allowances for these in your next forecast.
“So how do you do it?” Or “measuring business performance without an accountant”
I’ve described forecasting in detail. And I’ve extolled the virtues of variance analysis. But how do you actually do it?
What should you compare?
Financial reports are a great starting point because they provide a standardised way of looking at any business.
If you’re used to looking at reports of the business’ past performance (perhaps in end of year reports) then forecast reports should be quickly familiar to you – they are the same, it’s just the month’s they refer to haven’t happened yet!
Forecasts are usually explained using a core of three reports – the Cash Flow, Profit & Loss and Balance Sheet. Each of these three reports show the business’ activities in a different light – providing different information about the future state of the business.
Compared to this forward-looking forecast, actuals are things that have actually happened. If in July you make a forecast that starts in August to try and predict how the business will perform over the next 3 years, within a month you will catch up with the start of that forecast and you’ll start to have actual information on how the business performed in August to compare with your forecast.
Forecast reports and past financial reports are laid out in the same way and display the same information. This makes it genuinely easy to compare the two. The same terminology is used in each, and figures can be matched line-by-line.
Imagine you are doing the exercise described above.
It’s August – and you have the business’ Profit and Loss report in your hand for the month of August, showing exactly how the business performed.
Next to this, you have the forecast you made a month earlier when you predicted what would happen in August. By matching and comparing these two sets of data you can quickly see whether the business is on target.
Comparing forecast values and actual values for a single month is useful in itself and can make you quickly aware of any dangerous shortfall or useful surplus in that month.
Detailed variance breakdowns
Variance analysis isn’t just about quick takeaway information.
Knowing that you are better or worse off than you expected to be is just the start. Understand why there’s a difference is the next level.
For this you need to do detailed variance analysis – at a level where you can see not just that income is lower than predicted this month, but where you can compare the actual values and forecast values for individual products or services, sales channels, expenses and projects.
If you have this level of fine detail available, you can see which of your predictions are causing the variance between your actual values and forecast values. How to set this up is covered in next week’s article!
Repeating this process every month will improve your financial forecasts – making them a closer and closer match to reality.
How is variance expressed? As a percentage or a number?
It’s a trick question!
Variance can be expressed as either a percentage or a number. Percentage differences between actual and forecast values tend to use the forecast as the base number and show how the actual figure differs from the forecast as a percentage.
Variance can also be expressed as the difference between the forecast value and the actual value.
Both of these ways of representing variance are useful, percentage variance showing any change in values in the context of the original value, and making it easier to see trends over a period of time.
We’ll look at some more examples of variance in the next article!
Tune in next time for…
Part 2 in the guide to variance analysis. This will cover:
- How to do variance analysis well – detailing the kind of actual and forecast data you will need to get the most out of variance analysis.
- How to measure the performance of a business, with practical examples.