The UK e-invoicing debate – purity and pragmatism

The UK e-invoicing debate – purity and pragmatism

Posted by Ian Burdon in AP Automation, e-invoicing, Electronic Invoicing, EU Commission 18 Dec 2013

I attended the round table event on e-invoicing in London on the 9th December and I shared some thoughts on the event here. I want to share a few more thoughts, but this time I want to look at the way the debate is framed technically.

I have previously criticised PEPPOL for being far too rooted in technology rather than business. I am not alone in making that comment although I seem to have my head above the parapet more often. A similar sense washed over me at the roundtable, partly from the contributions from the European Commission and Open PEPPOL but also from other industry insiders there.

My feeling is that there is an industry agenda in play here, an agenda which ignores business realities. The impetus is to seek a ‘solution’ which ticks all the boxes of the industry ivory tower whereas if we really want eInvoicing to gain traction we need to look at what is already working in the market and at what participants actually want.

Purchasing Insight logoFor example at one point in the discussion someone said that what we really want is a means for systems to talk directly to systems, a view which conveniently overlooks that such an approach has been available for decades either through EDI or XML and has been comprehensively ignored by the vast majority of suppliers and buyers. I can see no reason to think that they are suddenly going to conclude that it is workable.

Similarly it was mentioned that SEN is likely to propose a European standard based on UBL and UN/CEFACT. A look at the figures from one provider in the market shows that what is actually working now in the market can be split roughly into 3 types – one third cXML, one third eBIS XML and one third proprietary standards much of which was SAP iDoc which is also simple XML schema. Only 2% of their activity is based around UBL and that is entirely within Denmark (and even then is, I am told, a version tweaked for the Danish public sector). There is some us of JSON in the USA.

Let me be clear, I have no problem with SEN proposing standards in UBL and UN/CEFACT so long as it also recognises what the market is actually standardising around which is simpler forms of XML. But this isn’t a message that the purists seem to want to hear, and neither do the reasons why UBL is not popular make much headway – hint: if it is easier to use paper than UBL then people will use paper. Paper invoices don’t fail at a schema level for minor errors such as typos.

We live in a commercial world and, as Tim McGrath commented during the morning session, standards arise from use not from mandate. To mandate ‘standards’ which are not in use to the exclusion of those which are is a special kind of madness reserved for the IT industry and policy makers.

Ian Burdon can be found on twitter @IanBurdon

  • Christian Lanng December 18, 2013 at 4:44 pm /

    It’s sad to see what used to be a good blog, succumb to a complete lobbyist FUD tool. It’s also sad that Ian still don’t disclose that his employer have direct commercial interest in this discussion and he is very far from neutral. At least Charles Bryant, normally have the tact of telling everyone he is a lobbyist.

    UBL is used in 190+ countries including countries such as Denmark, UK, Sweden, South Korea, China, etc. Furthermore Tradeshift is the fastest growing network in the world today and we have active users in 190+ countries who are all using UBL every day without any complaints. With a network of more than 400,000 companies in 3 years and another 5m Quickbooks users in the US soon starting to use UBL as their core transaction standard I think you can say there is plenty of traction behind UBL.

    At Tradeshift we are not purists, we will support any open, internet-based standard that is developed in a neutral way with access for all. Unfortunately there are no alternatives, since most of the existing EDI and e-invoicing players have been more interested in thwarting any attempt at standardization to protect their aging closed proprietary networks.

    The only point I agree in, is that traction will win, not committees. The existing players have managed a total EDI adoption of 10-15% in 20 years (or roughly the same as we have done in 3 years), I hardly see how anyone can call that “traction” in fact they seem to be sliding backwards these days.

    We picked the same model as the internet when we launched Tradeshift, open, free for all to connect and no barriers on who you can do business with, I will always bet on that side of the equation in the long run when it comes to traction.


  • Pete Loughlin December 19, 2013 at 7:10 am /

    I don’t agree or disagree with Ian Burdon’s viewpoint on standards. But I do defend Purchasing Insight’s policy of encouraging intelligent, insightful and informed debate and I would categorise most of your comments as such Christian. You have a viewpoint and you argue it eloquently. But please don’t criticise the medium that allows you to do that simply because others have a different view. It doesn’t do you any favours.

    Purchasing Insight does not moderate comments (apart from SEO spam) and people are free to express what ever opinion they have. Offering people a platform is like offering people a rope. You choose what you do with it.

  • Richard Fitzwilliam December 20, 2013 at 4:02 am /

    That’s the great thing about standards, everyone has one.
    If they were to adopt a standard then let’s be open and honest.

    If 1/3 use cXML today and SAP show no signs of slowing things up – that would make the case for adopting that standard. I saw a stat from them at the time they acquired ariba that 70% of the worlds GDP touches their combined systems. I think a reality check is in order – no amount of hatred for Ariba looks like it will slow them down. Last time I looked the potential for 70% of the worlds GDP trumped the potential of an extra 5m SME’s.

    PEPPOL reminds me of the famous dead parrot sketch from Monty Python. It’s not a dead standard, it’s just resting!

  • Christian Lanng December 21, 2013 at 8:36 pm /

    @Pete It’s exactly because I respect Purchasing Insight I criticize the medium, in the past I and others have written guest posts on this blog, and they have all been marked as such with a clear disclosure, in fact you preferred to interview me and other vendors, to keep the distance between the medium and the content clear. This is not the case with Ian Burdon, who is not a neutral observer or analyst, but very much representing a certain vendor and set of commercial viewpoints, which are very clear in most of his writing, unfortunately his affiliation is not. I think it’s a shame for readers to read this and not know the difference between his and your writing, and I think just like any other serious blog a disclosure policy would do you a lot of good.

    Now to some of the arguments above:

    – There are no real data in Ian’s post, just anecdote’s from a “certain operator” would this be the same operator who used to brag that 0.4% of their traffic was coming from other networks? I find the constant FUD, subterfuge and lack of willingness to stand by clear data highly disturbing and probably a clear indicator that there is no interest in a real data driven discussion.
    – @Richard 1/3 use cXML and SAP iDoc on Ariba’s network because it’s a closed proprietary network, do you seriously suggest that the government should be basing their procurement and e-invoicing initiative on standards that are owned by one vendor in the market? Who will that benefit?
    – As for slowing Ariba down, the last 10 deals we won all came in competition or from Ariba and it’s not only us, ask our friends at Taulia, Coupa, etc. why they are so busy these days. They in fact lose even when they drop their supplier fees as they have started doing as they get more desperate? Maybe the customers have realized ridiculous fees, complete lack of service, ancient technology and pissed off suppliers is a bad combo?

    Funnily enough the Ariba trolls remind me of the worst side of the internet, ad hominem attacks, no real content, hiding facts and thinking that just trolling enough will turn everyone away.



  • Ian Burdon December 23, 2013 at 10:33 am /

    I’d like to pick up on one point made by Christian. My views are my own and are essentially those I expressed while I worked in government. My approach is the same as it was when we created eProcurement Scotland as expressed in a series of meetings in, amongst other places, Oslo, Copenhagen and Paris arranged when what became PEPPOL was still being born. Those who were present then could no doubt attest to that.

  • Richard Manson January 9, 2014 at 9:31 pm /

    @Christian, obviously these blogs are designed to generate debate – some people will agree with the views posted, some won’t. I agree with Ian – and before you ask, yes I do have a business relationship with the firm Ian works for. But it’s not for this reason that I agree with Ian – or because I’m keen to sow seeds of fear, uncertainty and doubt – it’s because a) Ian speaks from experience of holding senior procurement positions in UK public sector and from being involved in PEPPOL and standards initiatives; b) I / my firm do this for a living, operate in many geographic regions and rarely come across organisations where UBL is their default document structure for sending invoices.

    Although we cannot boast about 400,000 sending organisations (which I’ll come onto later), I do feel we are qualified to make such a statement. CloudTrade goes to market through channel partners who typically white label or OEM our technology under their own brand. Although a relatively young company, over the last 2.5 years we’ve gone live with almost 80 end customers in Europe, Asia Pacific and North America (maybe we should be laying claim to the fastest growing e-invoice network :). We on-board new suppliers every day and rarely encounter organisations who are willing and able to send UBL.

    The main place where we do see UBL is in parts of Scandinavia (though I accept countries such as Netherlands and Turkey are promoting it). Though even in Denmark – where UBL is a requirement for organisations that want to send via NemHandel – our partner is making a nice living working with the many, many suppliers who are unable to adopt UBL and is acting as a middle-man mapping data enriched PDF’s using our technology (unlike UBL every billing application can generate a PDF) into UBL in order to post to NemHandel. The experience gained from working in Scandinavia and other places – as well as the feedback we get from our partners – is that the complexities and rigidity inherent with UBL, make it extremely difficult for organisations to adopt. My more technically gifted colleagues feel that UBL was designed by technical consultants without an appreciation of the real world.

    @Christian – you make it clear in your post that you’re keen for transparency. Great. With this in mind I’d like to challenge the implications you are making in a couple of your statements and ask for clarity about your numbers. Although I don’t believe Tradeshift has ever published transaction volumes or information about the number of portal suppliers v direct connection, I’m pretty confident that the majority of your 400,000 suppliers are using your portal to submit their invoices. Assuming this is the case (feel free to clarify it isn’t), I have two concerns about the implications you are making, firstly: an organisation using a portal to submit their invoice has no visibility of the type of document generated at the backend. So to imply that 400,000 organisations across the world are adopting UBL is exaggerating the truth a little don’t you think?

    Secondly: though the number of individual suppliers may be large, an organisation that is willing to use a portal to submit their invoice is likely to be a micro business that doesn’t use an accounting application. I know from bitter experience, any organisation that uses an accounting package such as Sage, Dynamics or SAP is not going to raise their invoice in one application, then duplicate the effort and do it again in a 3rd party portal. It just doesn’t happen. My point is, the number of vendors may be large, but how many actual suppliers are sending UBL direct and not as a result of using your portal? I suspect the number is very small and the actually volume of transactions that start off their life as UBL (and not mapped by you) likewise. Please advise if my assumptions are incorrect.

    I may sound as though I am anti UBL or standards as a whole, I am not. I am anti marketing hype getting in the way of hard reality. What I am definitely in favour of however is removing the barriers for suppliers in order to make it as easy as possible for suppliers to join the party. Asking a supplier to send UBL – or asking a supplier raise an invoice in a portal – simply creates barriers that impact adoption (of those suppliers that have a finance application and by implication send a lot of invoices that is :).

Post a comment