Insights for Procure-to-Pay and Finance Leaders


Failing to deliver? Part 2: The draft e-invoicing directive – a detailed look

The draft EU e-invoicing directive misses the mark on four major areas. The directive calls for semantic standardization of e-invoices rather than leveraging the existing investment made in pan European cross border e-invoicing solutions made by several member states. The directive does not take a stand on how invoices should be exchanged between suppliers and buyers. The directive does not provide a compelling event for suppliers to switch to e-invoicing by only making it an obligation to receive invoices electronically by public authorities. The directive grants member states 4 years to implement the rules – effectively placing Europe behind the curve on the digitalization of the procure-to-pay process. It is the wrong tools, it is too little and it is too late.

This is the second blog post about the draft EU directive on e-invoicing. It is in my opinion severely flawed because it does not take advantage of the opportunity to give European businesses a level playing field and competitive advantage. In my first blog post I looked at the development of the GSM standard in Europe and how it gave the European mobile telephone industry a competitive advantage that lasted for many years. My argument is that we could do the same in the field of e-invoicing and end-to-end e-procurement, but it requires a much more aggressive timeline and it requires that the EU commission withstands the pressure from the banking industry and old school e-invoicing service providers. GSM and the mobile telephone industry was based on a multilateral model and it is exactly what Europe and the world need in the field of e-invoicing. It is in my mind the most important thing to ensure in an e-invoicing directive and it is completely ignored it the draft directive.

Lets dive into the details of the directive and look at the flaws one by one. The directive calls for the development of a semantic standard for e-invoices by CEN. The directive furthermore suggests that the standard should be based on UN/CEFACT Cross Industry Invoice (CII) and ISO20022. Congratulations to the lobbyists on this achievement.

Delaying e-invoicing rollout 1-2 years: This call for semantic standardization is wrong in so many ways – that it would require a whole blog post just to explain why this it is such a disaster. But the short explanation is that it takes many years to mature a standard to the level where it can be used in cross border scenarios. The UN/CEFACT Cross Industry Invoice can hardly be considered a real standard and the ISO20022 Financial Invoice is as the name suggests the financial industry’s invoice format. Both formats suffer from being relatively immature and with limited to moderate real use.

Not connected with up-stream processes: The biggest problems with both formats is that they are not part of a suite of messages that covers other upstream activities. Invoicing is not the beginning of a payment process (what the banks seem to think) – Invoicing is the end of a long upstream process involving the exchange of quotes, purchase orders, dispatch advises, receipt advices and catalogues. Admitted – we have to communicate enough information to facilitate payment, but we don’t need to force the banking industry’s hammer onto the rest of the process.

Universal Business Language has been battle tested since 2005: In Denmark and later in Sweden and it is now the core format of the PEPPOL infrastructure that has demonstrated that it can scale and has been accepted by several member states (e.g. The Netherlands, Ireland, Austria, Denmark, Sweden).

Europe already has a standard: The draft directive stipulates that the standardization effort of the new semantic standard should be handled by CEN. This is a paradox in itself since the European localization of UBL is already handled in CEN BII since 2007.

The best thing about standards is that there are so many of them: There is basically no need for yet another semantic standardization effort in CEN. The world has enough e-invoicing standards and CEN is already doing a great job on localizing one of them for European use.

The directive does not talk about much about transmission and addressing of invoices between buyers and sellers. It merely states that invoices must be transmitted in a secure manner and that the semantic standard should cater for this! At the same time the directive stipulates that the use of digital signatures must not be a requirement.

Congrats to the lobbyists: This is another big win for the lobbyists that have no interest in Europe converging on multilateral service provider infrastructure. Recall my analogy with GSM and the roaming that takes place between telco operators. In mobile telephony the content is data (audio conversation, TXT messages and internet traffic) and the protocol and the backend infrastructure secures the conversation and handles roaming between the parties in a way that hides the complexity. The telephone numbers and the country codes handles all the addressing issues.

A half baked cake: An invoice standard without a transmission protocol and a roaming / peering infrastructure is a half baked cake that will not solve the problem. How will I, as a Danish supplier to a public institution in Spain, deliver my perfectly formatted and semantically standardized electronic invoice? Should I put it on a USB stick and put it in the mail? Should I send it via unsecure email? My vision is that any company in the world is capable of sending a business document (e.g. an electronic invoice) to any other company or public sector institution somewhere else in the world. In my mind this can only be realized with a secure multilateral service provider infrastructure. The infrastructure must support seamless roaming or peering of messages between the service providers in a way that hides the complexity from senders and receivers.

30 million Euro down the drain: The directive completely ignores the 30 million Euro investment already put into the PEPPOL infrastructure and standards. PEPPOL covers both the content (Universal Business Language), the addressing and the transmission, but the directive and the accompanying text does not mention PEPPOL or OpenPEPPOL a single time. However the European E-invoicing Service Provider Association (EESPA) which is basically a lobby organization is mentioned as one of the organizations that has been consulted. Many lobbyists will hear nothing of multilateral models like PEPPOL because they risk losing control. They are not interested in any framework where the public sector plays a part the governance as it is the case with PEPPOL in the OpenPEPPOL association.

A semantic standard that guarantees personal data protection: It is not even worth spending too many words on the lack of understanding of technology in the directive when it mixes the desire to develop a semantic e-invoicing standard that also guarantees personal data protection. One thing is standardization of content (semantic) and the other is the desire to transport the content in a secure fashion. Securing a content standard would require digital certificates, payload encryption and possibly signing – but the draft directive says that digital signing must not be a requirement. The directive simply neglects to mention the need for a coherent and secure way to transport e-invoices.

One of the biggest problems with the draft directive is that it only calls for an “obligation to accept the reception of electronic invoices”. There are several issues with this approach.

Without an obligation to send electronic invoices the uptake will only be moderate: Think about it – what is the incentive for small to medium size suppliers to implement e-invoicing towards their public sector customers today as compared to waiting until the majority of their customers accept e-invoicing? They would only be able to send e-invoices to a minority of their customers and the impact on payment terms is negligible. Savings in paper, stamps and handling time will save hours and perhaps days on an annual basis in a typical small and medium sized business. It will not save many full time employees. The public sector in Europe misses an opportunity to drive digitization, automation and innovation in the private sector. This is not only missed savings in the public sector it is also missed savings and missed competitive advantages in the private sector.

Some small and medium sized businesses resent the idea of making e-invoicing mandatory towards the public sector: This was also the initial response in Denmark where e-invoicing was made mandatory towards the public sector in 2005. But now that it has been implemented you will rarely find any criticism of the initiative. On the contrary businesses are pushing the public sector to digitize processes further upstream (e.g. purchasing). So this is an example of an initiative where politicians have to show courage and implement change that will require an up front investment from businesses but will lead to significant improved competitiveness for European business.

The Return On Investment (ROI) in the public sector is going to be low: All the affected public sector institutions need to upgrade their ERP systems to be able to receive and archive electronic invoices. They will need new invoice approval workflow solutions and they will need to make an agreement with one or more e-invoicing networks (more if the directive does not insist on a multilateral model like PEPPOL). The worst part of this is that there will only be moderate use for the new ability and the public sector institutions will need to support the old paper invoicing channel in parallel. This means that most of the business case is lost except if the public sector institutions make an even bigger investment in data capture products that can digitize paper invoices and route everything into the same channel as the electronic invoices. In worst case scenarios this would lead to increased costs for some institutions in the first few years until e-invoicing becomes mainstream.

The final and most significant flaw of the directive is the timeline. The directive calls for a 48 month implementation window of the new standard.

48 months is 4 years: It is actually unclear if the 48 months kicks in after the new semantic standard has been implemented by CEN or if it is from the date of the directive. If the 48 months are on top of the time it takes to implement the standard – then we are talking about at least 60 months in total (in a happy-go-lucky scenario). No matter how you look at it – this timeline will effectively position Europe behind the curve of e-invoicing adoption. E.g. compared to many countries in Latin America where governments are mandating e-invoicing with a firm hand. And this is very sad – because European businesses comparably will be less effective than their foreign counterparts.

Billions of Euros in unharvested process savings: That is the cost of 12-18 months of semantic standardization plus 48 months of implementation by member states combined with an obligation to receive rather than an obligation to send. It is unharvested process savings and missed competitive advantage that we cannot afford to waste.

The last blog post is called The ideal e-invoicing directive where I outline vision for where Europe could be after 3-4 years with a more aggressive path for public sector implementation of e-invoicing. We cannot afford to waste this unique opportunity to increase productivity and competitiveness in both the public and private sector.