28

Predictive sales analytics

What is it?

Predictive sales analytics is the process of figuring out how successful your sales forecast is and how to improve your sales predictions in the future.

Many companies now employ an analytics group that is responsible for data mining (Chapter 6) and analysis. This data mining activity looks for trends and relationships in the sales data which could help sales and marketing make more accurate sales predictions.

Alternatively, for smaller companies this type of analytics may be outsourced to a specialist data mining company that will determine future sales.

Why does it matter?

Sales revenue is the lifeblood of any business so knowing how much you can expect to receive has important tactical and strategic implications.

The ability to accurately predict sales matters because it provides valuable information for inventory management, staffing and cash flow management. Clearly, if you can accurately predict sales volume in the future, you can make or source only the products you need and manage inventory more efficiently so as to meet demand but not hold too much stock at any one time. Sales forecasts can also help you to ensure you have enough staff to handle the workload and ensure the best possible customer service. Sales revenue also drives cash flow, which is essential for any business. Knowing what to expect and when can help to make sure the business makes provisions to manage the slower periods as well as the busy periods.

Predictive sales analytics can also be used to secure funding and provide valuable customer data. If you are looking to secure a loan or financing from a third party for equipment or expansion you will need to demonstrate revenue and potential future revenue to prove your ability to repay the loan. Sales analytics can provide that data which can help facilitate the loan’s approval. Plus, this data can help you and the sales people to predict trends or changes to buying patterns that could improve sales still further. When individual sales people have access to this data they know what customers bought in the past and when they bought, which can be used to initiate contact and re-sell or up-sell.

When do I use it?

Predictive sales analytics is an extremely useful tool for planning and peace of mind. If you know what to expect then you are in a much better position to manage the peaks and troughs of your business, without the fire-fighting and sleepless nights. As a result, it is wise to use predictive sales analytics all the time so that you can make better decisions in the business and scale up and down when needed.

For example, many businesses experience more and fewer sales at certain times of the year. If you know that year on year you make fewer sales in June and July then you can encourage staff to take a holiday during that time and not get too stressed that sales are dropping. Or you could take special measures to source and secure sales from other areas, or develop different products or services to dovetail into the slower periods.

What business questions is it helping me to answer?

Predictive sales analytics can help you to answer:

  • How much product am I likely to sell in the next month/quarter/year?
  • How are my sales fluctuating throughout the year?
  • Are there any longer-term trends within the sales activity?
  • How do my various product lines compare in sales throughout the year?

How do I use it?

There are several ways of predicting sales. The first is to use analytics on past sales data to see if you can identify any patterns or trends that can then help you to forecast future sales. Of course, past results are no guarantee of future results.

In order to apply predictive sales analytics there is certain sales data you will need to collect. First you will need to have detailed sales for each product broken down by month, number of returned sales broken down by month and any external factors, or one-off events that influenced sales.

You can also use predictive techniques like regression (Chapter 7) and/or correlation (Chapter 3) to identify aspects of your offer that are affecting sales so that you can predict sales in the future. Plus scenario analysis (Chapter 4), Monte Carlo simulations (Chapter 13) and neural networks (Chapter 17) are also used widely in this area.

For example, Shell have used scenario analysis for many years to create a number of different feasible scenarios and how those scenarios would impact oil price, potential demand and revenue over the next 20–25 years. Every month they will review the various scenarios to decide which are more likely to happen and what they can then do to mitigate their risks accordingly.

Depending on your business, scenario analysis can be incredibly useful to help you outline a number of different possible futures based on not only your own data but additional data sets such as economic forecasts, climate change data or data on the growing influence of pressure groups that could impact your business. That way you can be ready for the future – whatever that might be.

Practical example

Say you are considering acquiring an additional business or are considering expanding your operations. Both will require capital and the easiest way to expand or buy another business is to use your existing reserves. Not only do you need to predict sales to work out whether you will be able to fund your new venture internally, but even if you can’t you will need to be able to predict future sales in order to secure funding.

Traditionally, we would look at past sales in order to do this so that we can then try to predict future sales. But this type of analytics starts to get really interesting when we also include some of the new data sets such as market trends, competitor data and even weather data.

For example, supermarkets will use weather data extensively to predict sales in the future and manage stock in the store. If the weather is due to be warm and sunny then they may promote their sausages and beef burgers and run in-store campaigns encouraging customers to enjoy the sun with a BBQ.

Initially this type of sales analytics started to help with stock control. Obviously stores only have so much space and there is no point using that space to offer a dizzying array of ice-cream in the winter! But now supermarkets are using this type of analytics further down the supply chain where they look at longer-term weather patterns to decide what suppliers to use in the first place.

Walmart use predictive sales analytics very successfully. They analyse buying patterns among similar types of customers and what competitors are charging in real-time as well as monitoring what’s trending on social media. For example, they learned via social media that ‘cake pops’ were popular with consumers and the company was able to respond quickly and get them into stores.

Tips and traps

Predicting future sales is always helped by detailed and thorough sales data from the past so keep accurate records. Also, make sure you account for returned goods to ensure that your sales figures are not inflated.

Don’t forget to take any unusual activity into account as this could easily skew the results and lead to inaccurate predictions.

Further reading and references

To find out more about predictive sales analytics see for example:

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
18.191.172.180