The Dutch Statistical Agency (CBS) recently published an interesting study into the Dutch internet economy. Interesting because of the used methodology and the fact that they have tried to put a number on online services. CBS partnered with Dataprovider and Google.

The first step in the research  was to select Dutch websites from Dataprovider’s website database which covers 203 million sites from 40 countries. The next step was to extract website characteristics such as CoC numbers, contact details, website functionality and certain keywords which were subsequently linked to the business register of the CBS. After that, researchers were able to add sector, employee and turnover characteristics to the study using VAT and tax databases.

By using this method it is also possible to distinguish between e-commerce sites selling products and sites that offer (online) services, sites that have or do not have payment functionality, and more. As a result a very comprehensive study has been put together that gives a better and more in depth study on the economic impact of e-commerce in the Dutch economy.

By framing the study as “Measuring  the Internet Economy in the Netherlands”, CBS leaves room for its own interpretation of the internet economy: there are webshops, online services and ‘Internet related ICT businesses’. From the above graph it can be concluded that internet presence is ranked from no website to passive, active and finally e-commerce website. However, from the methodology it seems that CBS has not looked at website capability to define the group Internet Related ICT, but rather what type of businesses (app developers, hosters, internet marketing, etc.) these are.

The study shows a good picture regarding Internet use by Dutch based companies for conducting business. It shows the value of goods sold rather than the value of goods and services purchased by consumers, businesses and government. E-commerce through foreign sites such as Amazon (Dutch turnover of € 320 million), Bookdepository.uk.co, Alibaba.com, etc. are not included, however. The same goes for one man bands (ZZP) selling through social media pages from Facebook, Instagram and others. Nevertheless CBS estimates the study covers 95% of Dutch websites which gives enough credibility to support its conclusions regarding the state of the Dutch internet economy.

Some key results of the study:

• A whopping 65% of Dutch businesses do not have a website, 70% of which are self employed professionals (ZZP). Especially in the construction sector there are many self employed professionals that have no website.
• Some 50.000 websites belong to what CBS deems to be the ‘core internet economy’, meaning those companies that either sell online products or services, or are dealing in internet related services. Their combined turnover equals € 104 billion in revenue or 7,6% of Dutch business turnover in 2015. There is a caveat here. From the combined crunched website data and linked business data it is not clear how much of a company’s revenue is e-commerce based. Using regression analysis and visiting a sample of websites from small companies leaves ample margin of error. Especially since an increasing number of traditional retailers transition to a combined on- and off-line commerce model (or pure e-commerce players opening up physical shops). The content and functionality and other website characteristics however do not provide conclusive insight into a company’s e-commerce based revenue share.
• The turnover of Internet Related ICT companies totals 71 billion and these companies employ 273.000 people in 16.000 businesses. This is a mixture of B2B and B2C e-commerce and compounds a variety of activities ranging from web designers, software developers, hosting companies, consultancy, communication and business services, but also activities in wholesale, retail and repair. By categorizing it as ‘Internet related ICT’ it is a big container of activities which main effect is that it makes up 68% of the Dutch Internet economy according to this study.

Another interesting conclusion from the study is that there appears to be ‘no clear pattern’ between businesses, websites, and productivity. In other words there is no indication that labour productivity is higher in e-commerce companies. In fact the related labour productivity graph below shows that most Internet categories experience a lower labour productivity. Only online services such as booking sites are more productive. This labour productivity conundrum does make you wonder about the alleged productivity impact of ICT and the promise of the digital paradise.

Labour productivity and website presence

Source: CBS 2016

The study takes the sizing of the Internet economy to the next level. At the same time the economic output per labour hour suggest that businesses without a website are more productive than companies with a web presence, web stores and Internet related ICT companies. Whatever the explanation (sector, value chain), companies without a web presence prove that they still have a solid reason to exist.

And here-in lies the rub of the study. The focus on output volume doesn’t give us a clue on the Digital Value Add (DVA). While the internet plays an increasing important role in doing business transactions it is still not clear how much it impacts our economic growth in terms of Gross Digital Domestic Product (GDDP). Since the authors of the CBS discussion paper ask for research suggestions, the METISfiles suggests to conduct a big data approach into the DVA of Dutch business: the value add generated through the usage of ICT. In doing so we can create a strong indicator to measure the progress we make  towards our digital future.