AI, Data & Tech

Thou shalt collect data!

By Freek Aertsen | June 12, 2017 | 2 min read

Google, Facebook, Twitter, websites, the weather forecast, cash sales, customer cards: data, data and nothing but data! We are being swamped by a data tsunami. How can we use this in the supply chain to achieve a better planning?

Marketing colleagues have already discovered the gold mine, but up to now applications for the supply chain seem to have been somewhat scarce. For instance, last year we started collecting data from different sources for a manufacturer of headphones. At that time we didn't know whether - or how - it might be useful. A group of enthusiastic customers, a number of good data scientists and the belief that something good would come from it was enough to make a start.

Besides traditional ordering and checkout (point of sales) data from retailers, data is also added from Google Search behavior as well as emotions connected with the product that are obtained through sentiment analyses on Facebook or Twitter. By employing web scraping technology, prices, features, reviews and promotions from all sorts of Internet sites are collected with a high level of frequency, not only for own products but also for those of competitors. The amount of data really is unlimited, the database is consequently enormous! But now it needs to be converted into sensible information.


After half a year, the first promising results are coming in. The data can be used to improve the demand forecast and it also gives a better - and in particular - faster insight into the success of the products on the market so that a far quicker response can be given to unexpected consumer behavior.

The short term demand forecast has improved by basing it on Point of Sales data and modeling the relationship between headphone sales and features, reviews, prices and competitive promotions. But the biggest advantage of having consumer information is that the response time to market disruptions has significantly been shortened. In the traditional chain, it takes a number of weeks before disappointing consumer sales become known at the electronics manufacturer. The use of information such as POS data, search behavior, product reviews, features, promotions and price trends for competitive products has reduced the response time to a few days. When retailer stocks are seen to be static, a quick analysis can be made of the cause; is it because the prices of competitive products have fallen (available on a daily basis), are the product reviews bad or has one of the competitors launched a promotion? Web scraping technology makes all this information available almost in real time. Because the information is available fast, it is possible to set up actions to influence demand. The response speed increases significantly; from eight weeks to just a few days. Influencing demand ensures that, in close collaboration with the Marketing and Sales colleagues, sales are brought more in line with the demand forecast. The online optimization of product offers, targeted marketing campaigns, dynamic price setting and the improvement of retail operations makes a quick and appropriate response possible to established deviations. In the end, it is not the demand forecast that is in error, but the sales are incorrect.

Tangle of data

What needs to be done to convert this big data tsunami into an effective tool? Simply start collecting data, even if one doesn't yet know what it will be used for. Don't restrict yourself and more than anything, don't think beforehand whether you will need it, that will become apparent from the data at a later stage. In order to fathom this tangle of data, a good group of quality data scientists is an essential condition. They are able to convert the data into significant models and reports. But in order to use it to the greatest effect, it is eminently important to convert the traditional monthly demand forecasting process into a frequent demand-realization process. The planning position will play a leading part herein and will have to organize the cooperation between the Marketing and Sales colleagues.

Optimizing your S&OP?

TIAS offers the Executive Master of Operations and Supply Chain Excellence which is a 11-module program, completed over a 22-month period. It offers you the latest theories, techniques and skills needed to successfully define and execute your company’s operations strategy.

Read more about this master

Knowledge area's

Grow at TIAS

At TIAS we believe in Life Long Development, continuous personal, professional and network development during and after your studies. With more knowledge, better skills and a broader network, you will be able to create more impact and be successful.

More about learning at TIAS » 

Brochure TIAS School for Business & Society

Get to know TIAS: The business school for tomorrow's leaders.