Business forecasting methods part 1: Inventory forecasting

Business forecasting is key to success and growth and is a key part of the business planning process.  Many businesses, whether retailers filling their shelves, garages making use of spares, or manufacturers consuming raw materials, will find much of their time is spent on trying to manage their inventory. Leaving aside the whole supply chain aspect, the first challenge is working out how much stock to make or buy.

Getting the correct amount of stock into your business is a significant challenge – too much, and you tie up cash that you might need elsewhere; too little, and you lose sales and provide poor customer service. So how do you deal with this problem? Every business is different, and the drivers behind their forecasting needs will also be different.

Ideally, you will have an amount of inventory that sells at a reasonably constant rate until you reach a pre-determined re-order point (the level at which you need to replenish stocks allowing for supplier lead-times) and then you place another order and carry on.In my experience, there are very few products like this.Almost always, you have bulk orders taking huge amounts of stock with little warning, or you have a seasonal impact on the level of sales, and any number of other factors that can frustrate your efforts to forecast accurately.

A lot of forecasting is, by nature, estimation. The secret is to make the estimation as educated as possible, and history is a great teacher. Examine the performance of other similar products, ideally over the entire product lifecycle.

Look at mean weekly or monthly sales, and you will begin to see patterns emerging – perhaps new items sell 20% more per week, on average, than on-going items. Maybe sales tail off by a measurable percentage once an item has been in the range for more than six months or a year.

It is likely to be possible to generate a series of product lifecycle graphs that can be used to inform the forecasts for new items.You may end up with a best-case lifecycle and a worst-case lifecycle, so then it’s a matter of getting some feedback on your new product to see if it’s likely to be an outstanding seller or just average. You can get feedback from your own sales team, if you have one, or perhaps from some ‘tame’ customers.

If you are a new business, or are launching a completely new product, of which you have no prior experience, then this ‘market research’ approach is about the only place you can go, unless you have some indication of how a similar product has worked for a competitor.

Apart from the forecast itself, it’s important to consider production options.Clearly, if there is a cost-effective way to break production down into stages, or even gather initial orders before going into production, then this is a good way of protecting yourself against unnecessary cost.

For example, perhaps items can be kept in a component state and not actually assembled or finished until projections become a bit clearer.You need to consider potential handling costs in doing this, but again it’s something that can work in some cases.

Interestingly, for some of the reasons mentioned above, it can be more cost-effective to manufacture locally as opposed to farming production out to the Far East, for example. I have known of cases where the extended lead times for offshore manufacture have meant going into production on highly speculative forecasts which have ended up costing more in stock surpluses and/ or shortages than was saved by going offshore. In one case, over 10% of production costs would have been saved by manufacturing in the UK, so things are not always as they seem!

Then there is cost to consider. This is not only economy of scale in terms of production, but also the cost of maintaining full stocks of your entire range, with the implied overstocking that may result from that approach. Oddly enough, supplying 100% of every order may not be the best thing for your business!

Finally, make sure you keep an eye on margins.You may get a large order from a big customer that will clean out your stock and leave you unable to fulfil any other orders for a few weeks.If that customer receives a hefty discount, then is the margin sufficient for you to disappoint a lot of other smaller customers who may be paying you full margin prices attracting no discount? Having a conversation with the big customer about delivery options can not only help your stock situation, but help your overall margins as well.