You might think that sales forecasting is a simple enough exercise – after all, sales managers only need to look at historic sales figures, analyze them and extrapolate the findings while applying any external factors such as inflation and interest rates. 

It would make sense that such a process should give a fairly accurate forecast of forthcoming sales over the next two or three financial quarters of the year.

Most people might say that one wouldn't have to be a financial genius to work out the relevant factors that should be considered when sales forecasting. 

But when you dig deeper into the details, the inter-relationship between many factors becomes quite complex.

In no particular order, these elements might be: 

•  Sales Team Performance: clearly, overall team performance can and should influence your sales forecasting. Obviously, historical performance needs to be examined, then you would need to consider any personnel changes or strategy variations.

• Customer Behavior: Understanding customers' preferences, purchasing habits and potential socio-economic demographic changes by sales region is paramount. Technological advancements and cultural changes influence buying decisions. One example might be sales of downloaded or streamed movies; if a particular actor has been publicly 'canceled' due to discovered previous offenses like sexual misconduct or overt racism – the chances are that movies containing that particular actor would suddenly become less popular. Similarly, DVD and Blu-Ray sales always diminish when faster internet connections are installed across geographical areas where postal deliveries of DVD sales is strong, but when fiber broadband arrives, everyone streams their movies instead of purchasing disc media.

• Expert Opinions: It makes sense to canvas industry experts to gain fresh insights and perspectives on sales from those with specialist knowledge. For example, a company selling instant noodles would find their sales gradually being eroded over the years if their products contained very high levels of salt. However, an industry expert in dietary health might advise that a major study on salt levels in food is soon to be released - highlighting health problems via high blood pressure from salt intake. This might well prompt the noodle manufacturer to change towards low sodium ingredients.

• International Events: If your organization exports and / or imports products or services, then trade policies, currency exchange rates and geopolitical events can affect sales in various regions. An obvious example of this is the war in The Ukraine after the Russian invasion in 2022. Since then, statistics show that the price of coal and wheat internationally have increased by around sixty percent – as Ukraine was known for its coal exports and has been called 'the breadbasket of Eastern Europe'. Events like pandemics and natural disasters may well have enormous consequences for sales forecasting. But if you could foresee all those sorts of things, you'd be very rich, and you wouldn't need to be reading this!

• Promotion and Marketing: Specific marketing campaigns, seasonal discounts, and promotional events will all have a significant sales impact. Examining the results of previous promotions will enable sales managers to predict upturns in sales. Cultural and religious high days and festivals often feature in promotional activities. As an example, supermarkets that sell delicacies such as Medjool dates always increase the price of those items and increase their stocks during the Muslim festival of Ramadan. This is because Muslims do not eat any food during daylight hours during Ramadan, and the quickest way to get their blood sugar levels up the moment after sunset is to voraciously gobble up handfuls of plump juicy dates.

• Market Research: Sales managers should keep informed regarding competitor performance, business data, industry trends and overall market conditions. Market research can take the form of surveys, both online and in print, competitions asking for customer feedback as part of a prize draw or whatever. Knowing what your customers think of your products and services is one of the most crucial elements of sales forecasting.

• Product Lifecycle: In order to predict seasonal and fluctuating sales figures, you should ask where your product or service sits within its natural lifecycle. The most obvious questions might be –'is this a mature product with stable sales, or is this a new service that may well grow rapidly?'

• Competitive Analysis: Keep your friends close, but your enemies closer. Keeping a close watch on your competitors is essential. You should look out for changes in their products, strategies or pricing. Clearly, if your competitor is undercutting you, that is bound to have an effect on your own sales.

• Customer Feedback: Online reviews through Google, ad hoc comments on Facebook and studied examinations on sites like Trustpilot can make a huge difference to sales, especially in eCommerce businesses, where products can't be seen and touched as they can in brick and mortar stores. If a particular product has been causing negative reviews on social media – not only does that product need to improve, but a sales forecast should be adjusted accordingly.

• New Product Launches: Obviously, there will be an expected impact on your sales forecast whenever innovation causes a new product to be launched. The honeymoon period of sales almost always increases whenever something new hits the market – but likewise there is often a dip after the azimuth.

• Sales Channels: Selling via omnichannel (i.e. e-commerce, retail or wholesale) brings its own discrete challenges when forecasting sales. Wholesale will always be a B2B exercise, whereas e-commerce is almost always a B2C process. But if a company does both, it's important to evaluate the performance of each channel and assess how the inter-relationship affects sales in each sector.

• Inventory Levels: The amount of stock on hand can and should affect your sales forecast. For example, try as they might to prepare for Christmas, most retailers and wholesalers run out of sticky tape by the second week of December. Thinking about shipping times from imported overseas-sourced goods is important. In short, if you have orders and can't fulfill them, that may be worse than having overstocks of goods languishing on warehouse shelves.

• Regulations and Compliance: Changes in regulations like data privacy laws, health and safety standards and regional variations in regulations can all be huge barriers to sales and consequent forecasting. For example, when the United Kingdom left the European Union, the amount of customs-related red tape and uncertainty over necessary export procedures drove many businesses to bankruptcy as the associated costs could not be covered. Likewise, the UK meat processing industry suffered due to changes in regulations of animal medicines and the simple inability to find UK workers to work in factories and abattoirs.

There is the concept of Sensitivity Analysis, (SA) is a quantitative technique used to assess how variables in one or more input parameters can impact the output of a forecasting model. 

SA is a priceless tool for understanding the reliability and accuracy of predictions, especially in complicated situations as above. 

SA can be applied in various industrial and service sectors, from insurance risk management to engineering and manufacturing.

Using AI and computing to pull the threads together.

Using online market analysis tools such as Forrester Research, Google Trends and 'Exploding Topics' is all helpful when it comes to sales forecasting. But when you have the statistics in front of you, their complex interrelationships need to be analyzed by AI or some sort of proprietary software to make sense of how they all affect each other.

Many AI-assisted sales forecasting tools are available tacked on to various business software platforms like Salesforce or Customer Relationship Management (CRM) platforms. More importantly, it's also possible to use Configure Price Quote (CPQ) software to make predictions about future sales.

CPQ software exists to help companies make accurate quotes by considering all the interrelationships between concomitant factors, thereby offering a price that will be acceptable to the customer, while providing sufficient profit margin to make the whole exercise worthwhile.

For example, a manufacturer of UPVC windows for new buildings might offer several different designs of sash windows or 'tilt and turn' units. 

If the cost of steel cable increases due to external factors beyond the manufacturer's control, a CPQ software package can and will adjust the final price of the sash windows (which use the cable to hold any counterbalance weights) by the increased percentage cost of the cable; it 'knows' exactly how many centimeters of cable are required for each window model.

Furthermore, the CPQ package will also advise if a change of specification is inappropriate or problematic. Imagine if the steel cable used on a product is 4mm thick, but the window manufacturer decides to use 3mm cable to mitigate the increase in cable cost.

A CPQ package might be pre- programmed with the tensile strength of 4mm cable. 

It notes that the proposed 3mm alternative is simply not strong enough to hold the 10 kg counterweights, and that the cable would probably snap after only a few hundred operations.

A CRM or spreadsheet package could easily help upscale sales forecasting capacity by predicting the changed profit margin of using cheaper cable, but only a CPQ platform could proactively advise against the specification change. 

CPQ software uses rule-based architecture so that any change in specification of a component is checked against all other related factors used in the same product.

The paralysis of analysis

Having read the above, don't sweat the small stuff too much! Remember that sales forecasts are never concrete and rarely 100% accurate. 

As fresh information comes to light, sales managers must be prepared to adjust their forecasts, but using a CPQ style AI-assisted software platform will allow quick and easy re-calculations, even if several interrelated changes are made at once.

Just like the window manufacturer mentioned above found out, these things are rarely straightforward – you might even say that they are definitely not simply 'open and shut'. 

Accordingly, it's important to remember that one's goal as a sales forecaster might be high accuracy as opposed to absolute perfection. It's not the 100% on-target figure that's important, more one's ability to be flexible. 

Such an outlook will prove sales forecasting to be one of the most practical tools in the day to day running of any business.