Blog

Retail Demand Forecasting – Then and Now

Facebook Twitter LinkedIn Google Plus

Gone are the days when heuristics ruled the world!  There was one chief with all the knowledge about his shop, and would know how much he would sale for each product he carries. He knew when the demand will peak, and when he has to push the sales, through discounts. He was mostly correct. [Even if he is not correct, nobody would know as no one accounting for lost sales. And no one dares to ask too].

But now the world is much more complex – for better! You sale through many stores – and you don’t necessarily own them. They call the shots many a times too. So if you supply there more, they will put you on discount. If you supply less and go out of stock, they will replace you and fill the shelves with the competition. You lose in both sides!

So now it is ever important that you predict your demand better – for each of your product and in each of the store you operate. Estimate consumer dynamics well, understand retail behaviour – get more control over things. Walmart has been so successful because their inventory cost is half of the competition!

But things have become more complex too. You need to manage hundreds of stores, and you are now selling few thousand products. The challenge is how do you manage such a scale, day after day, month after month? Simple tools like spreadsheets fail here.

Data and analytics can now help you much better – as more data is available across places, analytics is more powerful with forecasting / predictive tools available.

Few technology breakthrough that a retailer must know now, that will help him forecast better:

1. The Power of Computing – many of the time we are actually limited by the sheer size of the data to manipulate, and our traditional computing resources (PCs, Laptops, even servers) fail badly. Time to look at cloud computing – that provides you cheap computing power on the tap. And you can achieve scale as and when needed.
    
2. The Power of Analytics – Analytics has matured from simple descriptive of the history to futuristic prediction and forecasting. Event specific prediction, net impact of promotions and discounts, true impact of season – all are possible now with the power of predictive analytics.
    
3. The Power of Collaboration – Demand estimation and sales planning is a collaborative effort. We have moved ahead from spreadsheet sharing and lost in translation (data, ideas) days from real time collaboration and planning. Processes like CPFR has been greatly benefitted with collaboration tools.
    

The Power of Data (availability) – we had been cribbing about non-existence of data on certain critical factors that may explain demand. But in this connected world – data on such factors are available – may be in different places, but well accessible. Advance retailers are using factors like weather data, search volumes, social media signals, region specific events and festival calendars – to augment the internal data and build better forecasts.

Facebook Twitter LinkedIn Google Plus