Research shows that consumers make 80% of their retail purchase decisions in store, which is why it is vital for consumer goods manufacturers like Nestlé, Ehrmann, and Beiersdorf to gain access to retailers’ point-of-sale (POS) data.
This article looks at how POS data integrates with other data streams and at the new business scenarios that a “shelf-centered collaboration” approach enables.
Six years ago, a study conducted by consultants at strategy& confirmed not only that four out of every five purchase decisions were made at the store shelf, but also that more than 80% of managers from the retail and consumer goods industries wanted their respective sectors to collaborate more closely.
For companies in the consumer goods industry, the main aim of retailer-supplier collaboration is to gain access to retailers’ POS data and thus benefit from crucial daily information about sales figures. However, what the study also confirmed was that the task of turning POS data into actionable insight is far from straightforward.
“Each retailer has its own data format and data content,” explains Joachim Klippel, Solution Director of Consumer Products – Food at SAP.
Metro, Walmart, Tesco: sharing POS data
Although some major retailers, including Germany’s Metro, US-based Walmart, and the UK’s largest supermarket chain, Tesco, are already sharing their POS data, there’s one fundamental problem: They don’t all use the same data format.
“Regrettably, there is no industry standard here,” explains Klippel. So it’s up to the consumer goods suppliers to find a way around the data-format issue so that they can tap into retailers’ POS data, transform it, and combine it with in-house data streams ― such as delivery information ― to create actionable insight.
And that, says SAP’s Klippel, is exactly what the SAP Demand Signal Management application has been doing for the last two years: combining external and market data with internal business information and state-of-the-art analytics for a rapid response to demand signals from consumers.
“Using out-of-the-box adaptors provided by SAP-certified partners, manufacturers can stream POS data captured from hundreds of retailers around the world into SAP Demand Signal Management relatively quickly and easily,” says Klippel. The data acquisition process suddenly becomes much simpler. SAP partners do the acquiring and SAP Demand Signal Management cleanses and harmonizes the acquired information with the manufacturers’ internal data to create valuable business insights.
Despite the obvious benefits of enhanced analysis for their own sales operations, only Metro and a handful of other German retailers are currently willing to share their POS data. These benefits include:
A consumer goods manufacturer that supplies 10 retailers can filter several terabytes of data to ascertain whether sales figures are living up to expectations. If, for example, sales of a certain product are zero at a particular store on a particular day, the manufacturer can run the “enrichment algorithm” to find out whether an out-of-stock situation is to blame. If the product in question is out of stock, action can be taken immediately to replenish the shelves.
Armed with this kind of information, merchandisers can operate more efficiently, too. “For one thing, they can schedule their field work more accurately,” says Klippel, “because they can prioritize their store visits according to where the need is most urgent.”
Market research analyses
Syndicated data provided by the likes of German market research institute GfK (Gesellschaft für Konsumforschung) and US marketing research firm AC Nielsen are highly sought after. Large consumer goods manufacturers are willing to pay millions for information about how their products measure up to those of their competitors.
However, if it’s international comparisons they’re after, they won’t have much luck. That’s because market researchers source their data from many different databases with heterogeneous data content, explains Klippel. To find out how its products are selling in the UK or France compared with Germany, a Hamburg-based consumer goods group has therefore opted to deploy SAP Demand Signal Management. In doing so, it stands to benefit from the fact that SAP Demand Signal Management powered by SAP HANA not only analyzes massive volumes of data (big data), but unstructured data too.
“Analytical insight can now be paired with POS data and online sentiment analysis for enhanced decision-making,” says Klippel.
Sampling stations (like those at EDEKA), “three-for-two” product offers, and major advertising campaigns are all aimed at boosting product sales. But do they actually work? Or are some campaigns more effective in certain locations than in others?
“To improve your campaign planning, you need point-of-sale or syndicated data,” says Klippel. By analyzing demand history, consumer goods suppliers can see what effect a price discount is having on sales and decide on a day-to-day basis whether or not to continue a promotion. Thanks to the latest technology, they can pinpoint in near-real time the stores in which a promotion is not proving successful. “Getting hold of information like that used to take weeks. And by the time you got it, the promotion had already ended.” Syndicated data can be added here to gain further insight.
Ideally, all three scenarios dovetail. So, for example, if a manufacturer launches a new product, it can quickly find out how that product is selling in individual stores (daily POS data), what sort of reception it is getting from consumers (sentiment analyses on the Internet), and whether the product is causing sales of competitor products to fall (syndicated data).
SAP Demand Signal Management can incorporate context information into sales forecasts too. This can be particularly helpful for manufacturers of fast-selling products like ice cream and beverages. If weather reports show that the next few days are going to be sunny, with temperatures above 25 degrees centigrade, manufacturers can quickly alter their replenishment planning to meet the anticipated increase in demand.